Abstract

Pressure ulcer (PU) risk assessment practices in adult intensive care unit (ICU) patients remain varied. Purpose: The authors assessed the performance of the Sequential Organ Failure Assessment (SOFA) scale and its subcategories in predicting the development of PUs.

Methods: A retrospective cohort study was conducted of all adult patients admitted to the mixed medical-surgical ICU of a Finnish tertiary referral hospital between January 2010 and December 2012. Data (diagnoses, demographics, clinical information, treatments, and instrument scores) were retrieved from the ICU database. Wilcoxon and chi-squared tests were used to examine patient subgroup (medical or surgical ICU and intensive care or high-dependency care patients), length of ICU stay (LOS), modified Jackson/Cubbin (mJ/C) scores and SOFA subcategory variables, and first-day SOFA scores. PU association was determined by logistical regression. Results: Among the 4899 patients in the study population, the overall PU incidence of acquired PUs was 8.1%. Medical patients had significantly more PUs (145/1281; 11.3%) than surgical patients (212/3468; 6.1%) (P<.0001). In all subgroups, significantly more patients with PUs had higher SOFA scores (mean 8.24) than patients without PUs (mean 6.74) (P =.001). The difference persisted when patients with LOS ≥3 days in the ICU were considered. Among the SOFA subcategories, the Glasgow Coma score, renal and respiratory disorders, and hypotension were significantly (P<.0001) linked to PU development. First-day total SOFA score and its cardiovascular and respiratory subcategory scores were the most important predictors of PUs. Conclusion: The total SOFA score provides an additional tool to assess PU risk in ICUs and should be used together with the Braden or the mJ/C Scale.

According to cohort studies and systematic reviews,^1-5 patients in intensive care units (ICUs) are at a high risk of developing pressure ulcers (PU) because of their condition and limited ability to move due to critical illness and therapeutic interventions. The prevalence of PUs among ICU patients varies from 5% to 30%; these numbers have decreased during the last 2 decades.^1-4 PUs predispose the patient to a considerable risk of complications (eg, prolonged hospitalization and infection), and the deleterious consequences of PUs cause suffering to the patient, carry high societal costs, and place extra demands on the nursing staff.⁵ 

PUs have a multifactorial etiology, and more than 100 different risk factors have been identified^6-9; recently, 46 PU risk indicators in 6 broad categories were acknowledged as relevant to patients in ICUs.^6-8,10 Multifactorial analysis has identified an association between the development of PUs and the patient characteristics age and length of stay (LOS), the comorbidities diabetes and cardiovascular disease, the intrinsic factor hypotension, and the iatrogenic/care factors prolonged mechanical ventilation and use of vasopressor agents.¹⁰ Furthermore, PU risk scales such as the Braden Scale or Jackson/Cubbin Scale and severity of illness/mortality risk scales such as the Simplified Acute Physiology Score (SAPS) II calculator¹¹ or the Sequential Organ Failure Assessment (SOFA) scale¹² predicted PU development.¹⁰ 

The Braden Scale is the most widely used risk assessment instrument in intensive care, although it is not designed for use in an ICU setting.^4,13 The Jackson/Cubbin risk scale was specifically designed for the assessment of PU risk in intensive care, but it has not gained wide acceptance even though it has been shown in cohort and validation studies to more accurately identify patients at PU risk than the Braden Scale.^14-16 According to 2 retrospective cohort studies,^3,17 the modified Jackson/Cubbin (mJ/C) risk scale is a useful but not an optimal instrument for PU risk assessment in the ICU setting. The first study examined the suitability of the mJ/C risk scale for PU risk assessment among adult ICU patients³ and the second¹⁷ examined the specific accuracy of the 12 main subcategories of the mJ/C for predicting PU development.

Two (2) categories of the mJ/C — oxygen requirement and hemodynamics — were found in a retrospective cohort study¹⁷ to be important prognostic indicators for PU development. The Braden Scale does not take these variables into account.¹³ The SOFA score assesses consciousness using the Glasgow Coma score (GCS) and takes into account respiratory and hemodynamic variables, albeit differently from the Jackson/Cubbin scale.^3,14,18 

A prospective cohort study¹² involving 9 medical-surgical ICUs (N = 299 patients) assessed SOFA as a risk indicator for PU in intensive care patients. Average patient age was 60 ± 17 years, 68% were men, and all study participants were on mechanical ventilation ≥24 hours. The possible risk factors for PU development in this study were assessed; first-day respiratory and fourth-day cardiovascular SOFA scores, winter season, and duration of mechanical ventilation were found to significantly affect PU risk, but the total SOFA score was not associated with the risk of PU. 

Currently, only LOS and hypotension/use of vasopressor agents can be considered uniform and clinically meaningful risk indicators for PU development in ICUs. The aim of this retrospective study was to examine the roles of the SOFA scale, mJ/C risk score, and LOS in predicting PU development among different patient groups in an ICU. 

Materials and Methods 

Hospital unit. The Turku University Hospital (Finland)  serves as a tertiary referral hospital for approximately 700 000 individuals from the cities of Turku and Vaasa and their rural areas. The adult ICU, staffed by 160 nurses, has 24 beds and serves as a national center for hyperbaric oxygen therapy. This ICU treats surgical and medical patients needing intensive care or high-dependency care (HDC). Patients are classified on admission by the treating physician according to treatment needs; patients with major burns and organ transplantation are treated elsewhere. The physician makes the admission diagnosis(es) to determine whether a patient requires medical or surgical care and enters these data into the electronic ICU database (Clinisoft, GE Healthcare, Buckinghamshire, UK). 

Nurses with special training in using the mJ/C scale for PU risk assessment³ and in wound identification and care enter relevant information into the database, including type and frequency of skin cleansing, skin integrity, PU presence, and National Pressure Ulcer Advisory Panel/ European Pressure Ulcer Advisory Panel (NPUAP/EPUAP) PU classification. The database has the capability to calculate the mJ/C and SOFA scores.^14,18

Study design. This retrospective cohort study included all adult patients (≥18 years of age) admitted to the ICU from January 2010 to December 2012 (see Table 1). The PUs included Stage 1 to Stage 4 and unstageable ulcers.¹⁹ Patients who had a PU on admission were excluded. The primary endpoint was the development of a PU during the ICU stay. The primary variable was the SOFA score and its subcategories. The secondary variables were the mJ/C score and LOS in the ICU and patient subgroup (ie, medical or surgical and ICU or HDC patients). 

Risk assessment instruments.

SOFA. The SOFA scale includes 6 assessment categories: serum bilirubin concentration, platelet count, renal dysfunction, nervous system (GCS) status, hypotension, and presence of respiratory disorder.¹⁸ Each category is scored from 0 (low risk) to 4 (high risk), giving the SOFA score a range from 0 to 24. The SOFA score system is validated; the higher the score, the more severe the patient’s condition and the higher the mortality risk.^20,21

mJ/C. The original Jackson/Cubbin Risk Scale was created to assess PU risk in ICUs.¹⁴ Minor modifications have been introduced (ie, the mJ/C) to improve reproducibility of the original J/C scale in clinical use.^3,14,17 The categories weight/tissue viability, respiration, and incontinence, and the 3 subcategories of deduction points (transport to examinations or treatments, blood product, and hypothermia) were subject to minor modifications.^3,17 The mJ/C scale consists of 12 main categories graded from 1 (high risk) to 4 (low risk) to describe 12 specific variables of the ICU patient’s clinical situation. The minimum score is 9 and the maximum score is 48, with a lower score indicating a higher risk for PUs.^3,14 An electronic version of the mJ/C scale was introduced in 2009 into the clinical database of the ICU of Turku University Hospital for use by the ICU staff.

Patient management. ICU patients are observed according to clinical needs. One (1) nurse is responsible for 1 to 2 patients. Several physiological and laboratory variables (eg, hemodynamic, oxygenation, GCS, hemoglobin, electrolytes, C-reactive protein) are followed continuously every day to ensure proper management of these critically ill patients. The SOFA¹⁸ score is determined at admission and daily thereafter; mJ/C risk scale assessment follows a similar pattern. If the mJ/C score sum is <29 points, the PU risk is high or extremely high.^3,14 As soon as possible (ie, on admission or within 24 hours after risk assessment), high-risk patients are provided a dynamic pressure redistribution mattress if they do not already have one. Otherwise, PU prevention follows general care guidelines¹⁹ in terms of positioning therapy and skin assessment. 

Data extraction and statistical analysis. The predefined data for patient demographics (age, gender, body mass index [BMI]), ICD10 admission diagnoses, ICU type and medical or surgical patients as described in Table 2, LOS, SOFA and mJ/C scores and subscores, hemoglobin concentration, sedation status, body temperature, mattresses used, PU data, and ICU outcome (moved from the ICU as recovering or deceased) were retrieved from the ICU database by the database administrator. The analysis dataset was transferred to SAS version 9.4 (SAS Institute Inc, Cary, NC) for further analyses. SOFA and mJ/C scores were treated as discrete values, whereas LOS data were analyzed into 2 categories (<3 or ≥3 days after admission).²² The first-day SOFA or subcategory scores were used to test whether they predict the development of PUs. If the first-day SOFA score was not available for all subcategories, the subscore was given a value of 0 (ie, normal). If any other information was missing, no data from that patient were included in the analysis (N = 35, see Tables 2–5).

Patients with 1 or more PU, regardless of the stage, were included.¹⁹ The PU incidence (percent) was calculated by dividing the number of patients who developed 1 or more new PUs (irrespective of stage) during their ICU stay by the total number of patients in that group (times 100). The total mJ/C and SOFA scores of ICU and HDC patients and PU and non-PU patients were compared using Wilcoxon’s signed-rank test. The differences in proportions were determined with the χ² test. The association between the SOFA scores and PUs was determined by logistic regression.  

Ethical consideration. The study plan was approved by the Ethics Committee of Hospital District of Southwest Finland (T25/2011, 14.06.2011, §172).

Results

Of the 4899 adult patients admitted in the study time frame, 115 had PUs on admission and were excluded from further analysis. The 4784 study participants included 3017 (63%) men and 1767 women, with a mean patient age of 61.8 ± 15.7 years. Mean BMI was 27.2 ± 5.5 kg/m². More men (264, 8.8%) than women (126, 7.1%) developed a PU (P = .0028). Average LOS in the ICU was 3.6 (range <1–≥ 60) days; the proportion of patients with LOS ≥3 days was 30.3%. PU patients tended to be older (62.2 ± 15.6 years with a PU versus 60.7 ± 15.7 years without; P = .084). Among all patients, 1264 were HDC (26.6%); 1281 (27.0%) of all patients were medical (see Table 2).  PU incidence was 11.2% (120/1071) among medical intensive care patients and 11.9% (25/210) among medical HDC patients. PU incidence was significantly lower among surgical intensive care (164/2414, 6.8%) and surgical HCD (48/1054, 4.6%) patients (P<.0001), respectively (see Table 2). Of the 49.6% of patients at high risk of PUs (mJ/C score ≤29), 59.9% developed PUs (P<.0001, χ² test, data not shown). 

SOFA scores. On admission, the intensive care patients had significantly higher SOFA scores than HDC patients (P<.0001) (see Table 2). Patients who developed PUs had a mean SOFA score of 8.24 ± 3.44; the mean SOFA score of patients who did not develop PUs was 6.74 ± 3.16 (P =.001, 2-sided, Wilcoxon’s signed-rank test). The total SOFA scores on admission predicted the development of PUs (Wald’s χ² test 34.3517, DF 19; P = .0167).

Among all patients (4749), 362 (7.6%) were in the high SOFA score group (≥12 points). The higher the SOFA score, the higher the PU incidence in all patient subgroups (see Table 2). PU incidence in the <6 SOFA score group was 4.5% (74/1635) and 18.2% (66/362) in the SOFA score group ≥12 (P<.0001). Significantly higher SOFA scores (≥12) were noted more among intensive care patients than HDC patients (9% [315/3485] versus 3.7% [47/1264]; P<.0001). Thus, the distribution of patients (PU versus non-PU) among different SOFA score groups and between medical and surgical patients within ICU or HDC groups was statistically different (see Table 2). 

Among all patients, 1441 (30.3%) were treated for ≥3 days. Of the 357 patients with PUs, 304 (85.2%) had a LOS of ≥3 days (P<.0001, see Table 3). The overall PU incidence in the group with a LOS ≥3 days was 21.1% (304/1441) versus 1.6% (53/3343) in LOS <3 days. In the population where LOS was ≥3 days, proportionately more patients with PUs had a higher SOFA score (≥12) (20.7% [63/304]) than those without PUs (13.5% [153/1137]; P =.0037; see Table 3). 

In terms of the SOFA subcategories, PU development was not statistically linked to serum bilirubin concentration (P =.1188; see Table 4), but it was related to the platelet count (P =.0042), the GCS, renal dysfunction, hypotension, and respiratory disorder (P<.0001 for each subcategory). The subcategories bilirubin concentration, platelet count, GSC, renal dysfunction, respiratory disorder, and hypotension were found to have 0 points in 97%, 70%, 66%, 40%, 21%, and 18% of patients, respectively (see Table 4). 

The distribution of patients within the respiratory category (see Table 5) was more even than in the hypotension category, where only 12 out of 4794 patients collected 2 points (data not shown). The worse the oxygenation in terms of the ratio of partial pressure of oxygen in the blood to the fractional inspired oxygen, the higher the incidence of PUs (P<.0001; see Table 5).

mJ/C scores. The mJ/C scores of different patient populations were divided into equal-size groups (median, Q1, and Q3). The intensive care and HDC populations differed regarding mJ/C score distributions (see Table 2), showing that the SOFA scores are valid over a wide range of mJ/C score points. The mJ/C scores of patients with LOS ≥3 days were not different between patients with and without PUs, showing that the various risk factors within the mJ/C score do not influence the interpretation of the importance of different SOFA score groups (see Table 3).

Discussion

The population of this study represented the typical mixed ICU population with regard to age, male gender dominance, LOS, and SOFA scores.^{3,8,9,11,12,18,22,23} PU incidence was lower than in many previous studies,^{1,2,5,8,9,11,20,21} which is probably due to improved preventive measures introduced in recent years.⁵

PUs have a multifactorial etiology and although many potential risk factors have been identified,^5,7 only a few seem to be independent risk factors in intensive care patients.^6,10 Male gender dominates among ICU-admitted patients and among PU patients with some exceptions.^9,10,12 Older age is a generally accepted risk factor for PU development.^{6,10,11,12,21} Patients in the present study exhibited a trend toward higher age among PU patients, but the mean age differences were small, as has been reported previously in cohort studies.^11,12,22 Thus, it is questionable whether gender and age ultimately affect clinical decision-making related to PU risk. 

Confirming previous reports,^9,12,22,23 in this study medical patients also had a higher rate of ICU-acquired PUs than surgical patients. This was probably due to the fact that medical patients have more comorbidity and are more severely ill than surgical patents, as indicated by their SOFA and mJ/C scores (see Table 2). Irrespective of the patient’s subcategory, the higher the SOFA score, the higher the incidence of PUs. This is not surprising, given that a higher SOFA score indicates multiple organ dysfunction and is associated with increased mortality.^18,20,21 Manzano et al¹² did not identify an association between total SOFA score and PU development, probably due to the rather small number of ventilated patients in that study. In a review of intensive care risk factors by Cox,¹⁰ the association between the severity of illness/mortality risk scales such as SOFA based only on the Manzano et al¹² study or PU risk scales for PU development was not considered sufficient. The present study shows total SOFA scores are associated with PU development, regardless of patient subgroup. Although the specific role of the mJ/C score in the development of PUs was not examined in the current study, patients with PUs had significantly lower mJ/C scores than non-PU patients. This result supports the important role of PU risk assessment in the ICU setting.^4,14,15 

Severely ill patients require longer treatment in the ICU.^1,18,22,24 In the current study, proportionately more patients with an extended LOS (≥3 days) — a uniformly accepted risk factor for PUs^{1,6,8,10,12,22,24}— developed PUs when their SOFA scores were high. This suggests that in addition to prolonged ICU LOS, a high initial total SOFA score is related to the development of PUs. A retrospective cohort study²⁴ found that if the mJ/C scores did not differ markedly between the PU and non-PU patients in the group of patients with LOS ≥3 days, the mJ/C score and the LOS complemented each other when the risk of PU was assessed for ICU patients. 

To the authors’ knowledge, the association between platelet count and plasma bilirubin concentration and PU development has not been studied previously.^5,12 The ability to demonstrate only a trend favoring an association between these variables may have been due to normal platelet counts and bilirubin concentration values or a lack of these data on the first day of patient management at the ICU, because most patients had 0 points in these subcategories on admission.

Frankel et al²⁵ and Nijs et al² reported renal impairment is a risk factor for PU development, and the current study results concur. This contrasts with the study by Manzano et al¹² that did not detect an association between renal failure and PU development. However, in the current study, an association was found between GCS and the development of PUs. Neither Manzano et al¹² nor Cremasco et al¹¹ reported GSC data in relation to PUs; still, GCS is part of both the SOFA and SAPS II scales. Neurological failure among non-PU patients in the study by Manzano et al¹² was more common than renal failure among PU patients, possibly related to the fact that the patients were mechanically ventilated and thus immobile.^5,17

Data from the cardiovascular and respiratory SOFA categories were evenly distributed among patients. These categories were the most reliable indicators of PU risk; they also contained the fewest 0-value data points at admission. This is in line with the report by Manzano et al¹² that identified the first-day respiratory SOFA and the fourth-day cardiovascular SOFA subcategories as risk factors for PU development in ventilated ICU patients.  

Tissue oxygenation is important for tissue viability. As such, it was not surprising that the SOFA cardiovascular and respiratory subcategories provided important data with regard to PU risk. Hypotension often is related to cardiovascular disease, and both are regarded as risk factors for PUs.¹⁰ Cardiovascular disease represents a broad range of conditions; a 5-fold difference in the incidence of PUs recently was shown among the various groups of cardiovascular diseases.²⁶ Thus, the SOFA cardiovascular category score gives a better indication of the risk of PU than the broad categories of cardiovascular diseases. One of the subcategories of the mJ/C score – hemodynamics – has been significantly associated with PU development.¹⁷ The definitions of the cardiovascular/hemodynamics subcategory are different between the SOFA and mJ/C scales, but both take into account the use of vasopressor agents and the risk of PU.¹⁰ 

Mechanical ventilation seems to be only a surrogate risk indicator of PU development in the Manzano et al study¹² because it is used to improve oxygenation.¹⁷ The respiratory subcategory of SOFA considers direct tissue oxygenation and is defined differently from that of the mJ/C scale, but both scales link the respiratory status to PU development.^12,17

The mJ/C risk scale includes 12 main categories, 5 of which correspond to the Braden categories.^3,13 Of the 12 subcategories, 8 (incontinence, mental condition, mobility, medical history, hygiene, oxygen requirements, hemodynamics, and general skin condition) appear to contribute to PU development.¹⁷ The Braden Scale does not take into account respiratory/hemodynamic variables.¹³ The present study showed the SOFA score predicts PU development in an unselected cohort of intensive care patients. The total SOFA score is currently recommended to be used as an adjunctive risk scale in addition to either the Braden or the (m)J/C scales.

Limitations

This study was retrospective by design. Some SOFA score values were not available for the day of admission (this was especially the case for platelets, bilirubin concentration, and the GCS), casting some uncertainty on the interpretation of the ultimate impact of these variables. The authors also cannot ascertain that PU preventive measures were provided as prescribed or that they remained the same throughout the 3-year period of the study. The large sample size of this study probably minimizes the potential impact of preventive measures on the outcomes of this study.

Conclusion 

A retrospective examination of the role of the SOFA scale, mJ/C risk score, and LOS in predicting PU development among different patient groups in an ICU found the higher the first-day SOFA scores, the higher the PU incidence. This association was independent of the subgroup of patients treated in the ICU (medical or surgical) and included patients whose LOS was 3 or more days in the ICU. Among the SOFA subcategories, cardiovascular and respiratory scores were the most important for predicting the development of PUs (P<.0001). Therefore, the authors recommend the use of the total SOFA score as an adjunctive risk scale in addition to the either Braden or the mJ/C. If no PU risk assessment scale is in use, the total SOFA score should be used; clinical PU risk assessment by health care professional should always be performed and documented. Additional multicenter cohort research is needed to support the evidence base for these recommendations. 

Acknowledgment

Statisticians Hanne Koskela and Riku Kivimäki provided valuable contributions for the analysis of the results. The language of the article was reviewed by Robert Paul, MD, PhD, certified translator.

References

1. Takala J, Varmavuo S, Soppi E. Prevention of pressure sores in acute respiratory failure: a randomized, controlled trial. Clin Intensive Care. 1996;7(5):228–235.

2. Nijs N, Toppets A, Defloor T, Bernaerts K, Milisen K, van den Berghe G. Incidence and risk factors for pressure ulcers in the intensive care unit. J Clin Nurs. 2009;18(9):1258–1266.

3. Ahtiala M, Soppi E, Wiksten A, Koskela H, Grönlund J. Occurrence of pressure ulcers and their risk factors in mixed medical-surgical ICU — a cohort study. J Intensive Care Soc. 2014;15(4):2–4.

4. VanGilder C, Lachenbruch C, Algrim-Boyle C, Meyer S. The International Pressure Ulcer Prevalence™ Survey: 2006–2015: a 10-year pressure injury prevalence and demographic trend analysis by care setting. J Wound Ostomy Continence Nurs. 2017;44(1):20–28.

5. National Pressure Ulcer Advisory Panel, European Pressure Ulcer Advisory Panel, Pan Pacific Pressure Injury Alliance. Prevention and Treatment of Pressure Ulcers: Clinical Practice Guideline. Cambridge Media: Osborne Park, Western Australia; 2014. Available at: www.epuap.org. Accessed September 9, 2018.

6. Tayyib N, Coyer F, Lewis P. Pressure ulcers in the adult intensive care unit: a literature review of patient risk factors and risk assessment scales. J Nurs Educ Pract. 2013;3(11):28–39.

7. García-Fernández FP, Agreda JJ, Verdú J, Pancorbo-Hidalgo PL. A new theoretical model for the development of pressure ulcers and other dependence-related lesions. J Nurs Scholarsh. 2014;46(1):28–38.

8. Bly D, Schallom M, Sona C, Klinkenberg D. A model of pressure, oxygenation, and perfusion risk factors for pressure ulcers in the intensive care unit. Am J Crit Care. 2016;25(2):156–164.

9. Becker D, Tozo TC, Batista SS, et al. Pressure ulcers in ICU patients: incidence and clinical and epidemiological features: a multicenter study in southern Brazil. Intensive Crit Care Nurs. 2017;42:55–61.

10. Cox J. Pressure injury risk factors in adult critical care patients: a review of the literature. Ostomy Wound Manage. 2017;63(11):30–43.

11. Cremasco MF, Wenzel F, Zanei SS, Whitaker IY. Pressure ulcers in the intensive care unit: the relationship between nursing workload, illness severity and pressure ulcer risk. J Clin Nurs. 2013;22(15-16):2183–2191. 

12. Manzano F, Navarro MJ, Roldán D, et al; Granada UPP Group. Pressure ulcer incidence and risk factors in ventilated intensive care patients. J Crit Care. 2010;25(3):469–476.

13. Bergstrom N, Braden BJ, Laguzza A, Holman V. The Braden Scale for predicting pressure sore risk. Nurs Res. 1987;36(4):205–210.

14. Jackson C. The revised Jackson/Cubbin pressure area risk calculator. Intensive Crit Care Nurs. 1999;15(3):169–175.

15. Seongsook J, Ihnsook J, Younghee L. Validity of pressure ulcer risk assessment scales; Cubbin and Jackson, Braden, and Douglas scale. Int J Nurs Stud. 2004;41(2):199–204.

16. García-Fernández FP, Pancorbo-Hidalgo PL, Soldevilla Argeda JJ, Rodrigez Torres MC. Risk assessment scales for pressure ulcers in intensive care units: a systematic review with meta-analysis. EWMA J. 2013;13(2):7–13.

17. Ahtiala M, Soppi E, Kivimäki R. Critical evaluation of the Jackson/Cubbin pressure ulcer risk scale — a secondary analysis of a retrospective cohort study population of intensive care patients. Ostomy Wound Manage. 2016;62(2):24–33.

18. Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med. 1996;22(7):707–710.

19. National Pressure Ulcer Advisory Panel and European Pressure Ulcer Advisory Panel. Pressure Ulcer Prevention and Treatment: Clinical Practice Guideline. Washington, DC: National Pressure Ulcer Advisory Panel;2009. Available at: www.epuap.org. Accessed September 9, 2018.

20. Minne L, Abu-Hanna A, de Jonge E. Evaluation of SOFA-based models for predicting mortality in the ICU: a systematic review. Crit Care. 2008;12(6):R161.

21. Moreno R, Vincent JL, Matos R, et al. The use of maximum SOFA score to quantify organ dysfunction/failure in intensive care. Results of a prospective, multicentre study. Working Group on Sepsis-Related Problems of the ESICM. Intensive Care Med. 1999;25(7):686–696.

22. Theaker C, Mannan M, Ives N, Soni N. Risk factors for pressure sores in the critically ill. Anaesthesia. 2000;55(3):221–224.

23. Bergquist-Beringer S, Gajewski BJ, Davidson J. Pressure ulcer prevalence and incidence: Report from the National Database of Nursing Indicators® (NDNDQI®) with the National Pressure Ulcer Advisory Panel (NPUAP). In: Pieper B, ed. Pressure Ulcers: Prevalence, Incidence, and Implications for the Future. Washington DC: NPUAP;2012:175–188.

24. Ahtiala M, Soppi E, Tallgren M. Specific risk factors for pressure ulcer development in adult intensive care patients – a retrospective cohort study. EWMA J. 2018;19(1):35–42.

25. Frankel H, Sperry J, Kaplan L. Risk factors for pressure ulcer development in a best practice surgical intensive care unit. Am Surg. 2007;73(12):1215–1217.

26. Ahtiala M, Kivimäki R, Soppi E. Characteristics of ICU patients with pressure ulcers present on admission, acquired in ICU or no ulceration: a retrospective cohort study. Wounds Int. 2018;9(1):10–16.

Potential Conflicts of Interest: Dr. Soppi was chairman of the board, Carital Group, Finland, a group of companies manufacturing and providing global marketing of mattresses for the prevention and treatment of pressure ulcers, through 2017. Financial support for the study included state research funding (grant 13693), the Turku University Hospital Foundation (year 2016), and the Foundation for Nurse Education.

Ms. Ahtiala is an authorized wound care nurse, Service Division, Perioperative Services, Intensive Care Medicine and Pain Management, Turku University Hospital, Turku, Finland. Dr. Soppi is a senior consultant in internal medicine, Eira Hospital, Helsinki, Finland. Dr. Saari is an associate professor, Department of Anaesthesiology and Intensive Care, University of Turku and Turku University Hospital. Please address correspondence to: Maarit Ahtiala, RN, Perioperative Services, Intensive Care Medicine and Pain Management, Turku University Hospital, Hämeenkatu 11, FI-20520 Turku, Finland; email: maarit.ahtiala@tyks.fi.

Abstract

The effectiveness of music therapy for relieving pain and anxiety during burn dressing changes has not been reported from India. Purpose: This study was conducted to assess the effect of music therapy on pain, anxiety, opioid use, and hemodynamic variables during burn dressing change.

Methods: Patients in a tertiary care burn unit who were >10 years old, conscious, able to respond, and oriented to time, place, and person participated in a 2-month, quasi-experimental, cross-over pilot study. Each served as his/her own control. Dressings were changed every other day alternating between the control (standard pain management) and experimental (control plus patient-selected music) intervention. Pain was assessed using a numerical rating scale, anxiety was scored using the State Trait Anxiety Test (higher scores indicated more pain and anxiety), and hemodynamic parameters and analgesics were recorded. Wilcoxon Test and chi-squared tests were utilized for statistical analysis. Results: Median pain scores (5, interquartile range [IQR] IQR: 3-7; and 6, IQR: 5-8) and median anxiety scores (12, IQR: 8-17; and 14, IQR: 10-19) were significantly lower during the experimental than during the standard dressing change, respectively (P<.001), and opioids were used significantly less frequently during the experimental change (P = .002). Conclusion: Music therapy helps reduce anxiety, pain, and opioid use during burn dressing change.  

Burn injury results from excessive exposure of tissue to thermal, chemical, electrical, or radioactive agents.^1,2 According to a relevant textbook,³ pain is integral to survival. Although no research describes the proportion of burn patients reporting inadequate pain relief, pain relief measures have been anecdotally reported to be insufficient in both adult^4-6 and pediatric⁷ burn populations, mainly due to fear of side effects and addiction to opioids, lack of routine pain evaluations, and standardized analgesic protocols.⁴ In their review article, Patterson et al⁵ recommended a structured approach to burn analgesia that includes performing routine pain assessment and incorporating both drugs and individualized alternative therapies based on psychological, cognitive behavioral, and operant learning techniques.⁸ A variety of psychological techniques such as hypnosis, relaxation, and music therapy may be helpful in managing burn pain by providing multimodal distraction.⁴ 

An exploratory study⁹ showed the effect of music therapy depends upon not only the type of music, but also upon the associations and memories of the music. Various studies performed in countries other than India have documented a positive effect of listening to music on burn patients’ pain and anxiety during dressing change. Hsu et al¹⁰ conducted a prospective, randomized controlled trial (RCT) (N = 70) to assess the impact of music on burn patients’ pain and anxiety at the time of dressing change. Patients were randomly assigned to 2 groups; the control group was provided standard interventions (ie, routine dressing change) and the experimental group listened to music during the routine dressing change. Pain and anxiety were assessed before, during, and after dressing change. Morphine use in both groups was recorded. A significant reduction in pain and anxiety was reported by the fourth day in the experimental group (P<.05), although morphine dosage remained the same in both groups. Similarly, in a quasi-experimental pretest-posttest design study by Son and Kim,¹¹ 32 patients were assigned into control (routine burn dressing changes) and experimental (listening to self-selected music during dressing changes) groups for 3 days. Pain and anxiety scores were self-reported using the State-Trait Anxiety Inventory (STAI)¹²; significant reductions were noted in both anxiety (P<.017) and pain scores (P<.012) before and after the dressing changes when patients listened to music in the experimental group as compared to the control group. 

Whitehead-Pleaux et al¹³ conducted a similar RCT study on pediatric patients. Fourteen (14) patients were randomly assigned to control (verbal interaction) and experimental groups (listening to live music). The participants of the experimental group anecdotally reported positive effects of music on pain and anxiety, although the difference in the pain and anxiety scores was not statistically significant between the experimental and control groups. Fratianne et al¹⁴ conducted a repeated measures study among 25 patients (age 7 years and older) serving as their own control to test the efficacy of music-based imagery and musical engagement (ie, activities such as listening to music, identifying or singing familiar songs, practicing deep breathing according to music rhythm, and playing simple musical instruments) in relieving pain and anxiety among burn patients during the debridement process. A significant reduction in self-reported pain was noted among persons who received music therapy versus those who did not (P<.03). Anxiety scores also improved, but the difference of the anxiety scores between the 2 groups was not statistically significant. 

A systematic review and meta-analysis¹⁵ of 17 RCTs (804 patients) on the effects of music intervention on burn patients found a significant positive effect of music intervention on pain alleviation (standard mean differences [SMD] = -1.26; 95% confidence interval [CI] -1.83 to -0.68), anxiety relief (SMD = -1.22; 95% CI -1.75 to -0.69), and heart rate reduction (SMD = -0.60; 95% CI [-0.84 to -0.36]). In a systematic review of 26 studies,⁶ 17 showed positive outcomes when nonpharmacological interventions were used to manage procedural pain among adult burn patients. 

Music therapy also has been documented to have a positive effect on pain and anxiety in patient populations other than burns. A repeated measures, pretest–posttest design study (N = 36 men) performed by Hwang and Oh¹⁶ compared 3 different types of music therapy (singing, listening to music, and playing instruments for 30 minutes, twice a week for 6 weeks) on levels of depression, anxiety, anger, and stress among alcohol-dependent clients. The authors reported participant anxiety scores were significantly reduced with music therapy (P<.05). Gutgsell et al¹⁷ conducted a RCT (N = 200) to assess the efficacy of a single music therapy session to reduce pain in palliative care patients. A significant decrease (P <.0001) was noted in the numeric rating scale (NRS) pain scores in the music therapy group.

Nilsson¹⁸ conducted a systematic review of 42 RCTs published from 1995 to 2007 on the effects of music therapy on pain and anxiety in the perioperative period among patients undergoing elective surgery. The type of music employed was soothing (ie, not more than 60 to 80 beats per minute); in 29 studies, a patient self-selected music. In 19 studies, the STAI¹² was used to assess anxiety. In the 22 studies that evaluated the effect of music on pain, a visual analogue scale was most commonly used (n = 12); other instruments used to measure pain included a NRS, the McGill Pain Questionnaire, and a verbal rating scale. The results showed (as demonstrated in approximately ~50% of the outcomes) music intervention had a positive effect on reducing patient anxiety and pain in the perioperative setting. 

A RCT by Wang et al¹⁹ evaluated the effect of music therapy on preoperative anxiety. The experiment group (n = 48) listened to 30 minutes of patient-selected music 30 minutes before the surgery and self-reported anxiety using the STAI; physiological measures of anxiety included heart rate and blood pressure. Results showed music therapy reduced anxiety levels of the patients in the experimental group by 16% compared with their pre-interventional level. The reduction in the anxiety scores in the experimental group as compared to the control group was also statistically significant (P =.001). In a review of the literature by Henry,²⁰ music therapy was found effective in decreasing pain and anxiety related to acute illness, injury, and painful procedures among critical care patients.

The current authors noted a lack of clinical evidence regarding use of music therapy among burn patients in India and conducted a study to assess the effect of music therapy on pain, anxiety, and opioid use among patients admitted in a tertiary care hospital in northern India.

Methods

Study location. The study was conducted in an 8-bed burn unit of a tertiary level hospital (Post Graduate Institute of Medical Education and Research, Chandigarh, North India) during July and August 2015. 

Study design. A quasi-experimental, cross-over (repeated measures) design was employed. Every burn dressing change was considered a study data point. In order to ensure complete homogeneity of the control and experiment group, each patient undergoing dressing changes served as his/her own control. All the patients irrespective of their post burn day were identified and screened for eligibility using the inclusion and exclusion criteria. The enrolled patients had a control dressing change on the first day of assessment. Following the cross-over design, the same patients went through an experimental dressing change on the third day (ie, during their next dressing change), and this cycle was repeated until discharge. Each patient was studied for approximately 10 days. The washout period was an important consideration in the cross-over design; dressing changes were performed at least 48 hours apart to reduce recall bias among the patients. 

Participants. Using the total enumeration technique, all the dressing changes performed in the burn unit during the study period were assessed. Inclusion criteria stipulated study participants must be burn patients >10 years old, conscious, able to respond, and oriented to time, place, and person. Burn patients on ventilator support, hearing impaired, and/or declared hemodynamically unstable by the attending plastic surgeon were excluded. In addition, patients had the choice of opting out of the study at any time if they were not comfortable with the intervention. Had this occurred, they would have been reported as drop outs.

Conceptual framework. The conceptual framework for this research plan was based on Melzack and Wall’s Gate Control Theory of Pain.²¹ This theory proposes that pain can be reduced by stimulating sensory neurons, which are larger and faster than pain neurons and effectively close the gate-carrying pain stimuli. Cognitive processes such as distraction reduce pain perception by consuming attention and stimulating the sensory neurons.¹²

Study tools. The primary outcome variables were pain, anxiety, and opioid use during the burn dressing change. Secondary outcome variables were hemodynamic parameters. An interview/assessment schedule was prepared to obtain patient sociodemographic (age, gender, education, occupation, and marital status) and burn profile (etiology and degree of burn, total burn surface area) data; these data were validated by experts from nursing education and clinicians from the department of plastic surgery of the authors’ institute. Using a NRS,²² a standardized ordinal level scale was used to measure pain perception. Patients rated their pain from 0 to 10, where 0 means no pain and 10 means extremely intolerable pain. The Indian version of the State-Trait Anxiety Test (STAT), developed by Vohra,²³ of the STAI for adults developed by Speilberger et al,¹² was used to assess patient anxiety levels 30 minutes before dressing changes. The STAT is a standardized tool with established reliability and validity.²³ Forty (40) items (20 measuring state anxiety and 20 measuring trait anxiety) are used to calculate the score. State anxiety is defined as an emotional state that exists at a given moment in time (ie, how the respondent feels “right now, at this moment,” in this case immediately before the dressing change). Trait anxiety is described as the likelihood of the person to experience anxiety when perceiving a stressful situation as dangeous or threatening.¹² Total scores for state and trait sections separately range from 20 to 80, with higher scores denoting higher levels of anxiety.²³

Care protocol. According to routine protocol of the unit, dressings are changed on patients in the burn unit every other day. Patients are bathed with water and a betadine scrub, and the dressing change is performed by the attending plastic surgeon. The wounds are dried using sterile dressing pads (gauze dressing wrapped over cotton). Antibiotic ointment (silver sulfadiazine or colloidal silver) then is applied over the wounds. The wound areas then are covered with a double layer of sterile dressing pads (12 inch x 12 inch) and the dressing is secured using sterile (6-inch) bandages. The routine pain management in the unit included pharmacological interventions prescribed by the physician and administered to the patients as needed. 

The burn unit’s analgesic routine was not manipulated during the study. The drug name, dosage, and route of the analgesics administered to the patients by the burn unit team during the dressing change were recorded per se. A clinical assessment proforma was developed to collect data about hemodynamic parameters (blood pressure, pulse rate, respiration, and temperature) using the cardiac monitors of the unit. All of the above study tools were paper/pencil instruments, and the questions were asked by the investigator who completed the forms. 

Music therapy protocol. A protocol for music therapy during burn dressing change was prepared on the basis of a review of literature and suggestions from medical experts. As advocated in literature, nonlyrical instrumental music pieces^8,17 were included in the therapy under 4 classes: Spiritual/religious (Bhajans and Sikhism shabad), Western instrumental (guitar, piano), Classical instrumental (tabla, flute, sitar), and Bollywood instrumental (ajeeb dastan hai ye, khamoshiyan). The 15 music pieces were assessed by 20 nursing research experts. The experts were asked individually to rate each music piece by giving it a score out of 10 for suitability to be used in music therapy in the patient population (1 = unsuitable and 10 = most suitable). The scores for each music piece were totalled and the music pieces with the lowest scores were eliminated, leaving 10 instrumental music pieces. The patient was given an opportunity to listen to all the music pieces on the first day of therapy; the music selection made by the patient was used every time s/he received music therapy during the experimental dressing change. 

The music was played using MP3 players and earphones at the bedside of the patient 30 minutes before and for 30 minutes after each dressing change. Ear phones were used to help the patient focus on the music and to remove distracting sounds of the burn unit (eg, cardiac monitors or people talking). The MP3 players and earphones were kept separate for each patient and were cleaned with alcohol swabs before and after use to prevent infections. Ear phones were used at bedside, and loudspeakers were used where dressings were changed. During control dressing changes, patients received standard care as per the unit’s protocol.

Ethical considerations. The research protocol was approved by the Ethics Review Committee of the institute. Informed written consent was obtained from all participants after explaining the research, objectives, and duration of the study. All the participants were informed about their right to refuse to participate or withdraw from the study at any time. Patient anonymity and confidentiality were maintained while collecting data and reporting the study by restricting data access to the core investigators. Nothing interfered with the routine treatment of the participants. Care was taken not to cause any harm or discomfort to them. The STAT was purchased with legal rights of use. The study was registered with the Clinical Trials Registry of India (Registration number: CTRI/2016/09/007281).

Data collection and analysis. The sociodemographic (gender, age, employment status, marital status, level of education), burn profile (type of burn, percent of body affected; ie, total burn surface area [TBSA]), and hemodynamic variable (blood pressure, heart rate, respiration, and temperature) data were collected by interview and clinical assessment. Using the NRS, patient pain scores were recorded 3 times: 30 minutes before, during, and 30 minutes after the dressing change. To obtain the pain score during dressing change, patients were asked to rate their overall pain during the procedure after the dressing change was over; this score was analyzed using median and interquartile range (IQR) data. Degree of anxiety was determined 30 minutes before burn dressing change as per the total score obtained on STAT. Analgesic use was recorded and classified into 2 categories for purposes of analysis: opioids (eg, tramadol, morphine) and nonopioids (eg, paracetamol and diclofenac sodium). The hemodynamic parameters were assessed at 3 instances: 30 minutes before, during, and 30 minutes after dressing changes. 

The data collected were directly entered into the Statistical Package for Social Sciences (SPSS for Windows, version 16.0, Chicago, IL) for statistical analysis. Descriptive analysis was used for the sociodemographic and burn profile data to assess frequency and percentage. Because the data did not follow normal distribution, nonparametric tests were used, with a level of significance of 0.05. The median values of pain, anxiety, and hemodynamic parameters of the control and experimental groups were compared using Wilcoxon test. The frequency of opioid use between the control and experimental dressing change was compared using chi-squared test. Post hoc analysis of the power of the study was performed using the “Gpower” software, version 3.0.10 (Universität, Dusseldorf, Germany). With an effect size of 0.5, with the given sample size of 52, a power of 0.95 was achieved.

Results

Demographic and burn characteristics. During the data collection period, 25 patients were treated in the authors’ unit. Of these, 10 were eligible for the study according to the inclusion and exclusion criteria. Participants included 8 men, 2 women, mean age 28.5 ± 13.7 (range 14–51) years; 6 were educated up to the secondary level, 3 up to primary level, and 1 was illiterate. Five (5) were unemployed, 6 were married, 7 had electric burns, and 3 had second-degree flame burn injury. The mean TBSA was 27% ± 19.0% (range 5%–60%). 

A total of 104 dressing changes (52 control, 52 experimental) were completed as study data points. The mean number of dressing changes of each participant was 10 ± 6.09 (range 2–18), 5 as part of the experiment (with music therapy) and 5 as control (without music therapy). The majority of the participants (7) chose spiritual music (gayatri mantra and shabad) for the therapy.

Procedural pain and anxiety outcomes. The overall pain score (median, IQR) for the experimental dressing changes (3, range 1–5) was significantly lower than for the control dressing changes (4.5, range 2–6) (P<.001) (see Table 1). Median state anxiety scores before the experimental dressing changes (12, IQR: 8-17) were significantly lower as compared to the control dressing changes (14, IQR: 10-19) (P<.001). The trait anxiety scores were not significantly different during the control and experimental dressing changes (see Table 2). 

Analgesic use outcomes. Most patients required analgesics during the dressing change. Of the total 104 dressing changes, an analgesic was provided 38 times in control group and 24 times in the experimental group. The frequency of opioid use was significantly lower during the experimental dressing change (9.6%) as compared to the control dressing change (34.6%), (P = .002) (see Table 3). Interestingly, the overall frequency of analgesic use was not reduced, but the type of medication changed. Patients were given weaker analgesics such as paracetamol more often during the experimental dressing change pertinent to the lower levels of pain perception. In all the patients, morphine was charted as 3 mg intravenously as needed. Thus, the objective to reduce the opioid use (as-needed use of morphine) was successfully achieved. 

Hemodynamic parameters outcomes. None of the hemodynamic parameters (blood pressure, heart rate, respiratory rate, temperature) was significantly different between the control and experiment groups (see Table 4). 

Discussion

Since time immemorial, music has been widely used to enhance well-being, decrease emotional difficulties, and distract patients from symptoms such as pain.²⁴ Despite this fact, no studies have been conducted in India that explore the use of music therapy to relieve patient pain and anxiety during burn wound dressing changes. In the current study, every patient served as his or her own control. This helped ensure the control and experimental dressing changes had total homogeneity in terms of participant characteristics. As previously described, Fratianne et al¹⁴ employed a similar repeated measures design with patients serving as their own control to test the efficacy of music-based imagery and musical engagement to manage pain and anxiety during burn debridement. A quasi-experimental pilot study was conducted by Kahar et al²⁵ in Singapore to assess whether music helps reduce the pain experienced by burn patients during dressing changes. Using convenience sampling, 30 dressing changes were alternatively assigned to control (without listening to music) and experimental (listening to music) groups. The authors used the NRS, a pain behavioral tool, and physiological monitoring. The music was soothing and distracting, but its effects were not found to be statistically significant. The authors concluded that further high-quality studies are needed to evaluate the impact of music. Also, they did not allow patients to select the type of music; they acknowledged that allowing the patient to select the music may have enhanced the study. Thus, the current study was built on the work of Kahar et al²⁵ by using patient-selected music. Self-selected music also was used in the Son and Kim¹¹ and Wang et al¹⁹ studies. In the systematic review by Nilsson,¹⁸ 29 out of 42 studies used self-selected prerecorded music. However, researchers also have used live music^13,17 and participatory music interventions such as playing instruments or singing,^14,16 showing music therapy has a wide scope in terms of use and practice. 

Pain intensity was measured using a NRS because it frequently has been reported to be a useful tool.^17,18,22,26 The current study is in consonance with many previous studies in documenting that music has a significant role in reducing the intensity of pain among burn patients during dressing changes.^10,11,14,15 However, Whitehead-Pleaux et al¹³ reported that music therapy for pediatric patients during donor site dressing changes generated mixed results; therefore, the study was statistically inconclusive, suggesting more studies in pediatric burn patient populations are needed. The current study confirms that music therapy is effective in reducing pain, with results similar to studies exploring music in other patient populations besides burns.^17,18 

The STAT was used because it is a widely accepted, reliable, and valid tool.²³ A significant reduction was noted in the median state anxiety scores before the experimental dressing changes as compared to the control dressing changes (P<.001). These findings are in line with the Hsu et al¹⁰ and Son and Kim¹¹ studies that found significant reduction in anxiety scores of burn patients when music-related interventions were employed. Similar to the current findings, studies involving other patient populations also have found music to be useful in relieving anxiety.^16,19 However, the Whitehead-Pleaux et al¹³ and Fratianne et al¹⁴ studies differ; they both reported an improvement in the self-report of anxiety scores of the burn patients, but the difference was not significant. The Wang et al¹⁹ study utilized physiological measures of anxiety (eg, heart rate, blood pressure, electrodermal activity, serum cortisol levels) along with STAI. Further research using these physiological measures in burn patients is warranted.

In the present study, the frequency of opioid use was found to be significantly lower in the experiment group (P = .002). Because literature is scant regarding the effect of music therapy on opioid use among burn patients, further exploration to generate satisfactory evidence is warranted. None of the physiological parameters such as blood pressure, pulse, respiratory rate, and temperature — assessed in this study as secondary outcomes because they have been reported to be indirect measures of anxiety and pain¹⁹— was significantly different between the control and experiment groups. Wang et al¹⁹ had similar findings, with no significant difference found in heart rate and blood pressure values (P = .5) between persons who did/did not have music therapy. However, Almerud and Petersson²⁷ reported a significant decrease in both systolic and diastolic blood pressure during music therapy sessions among intensive care unit patients temporarily on a respirator.

Limitations

Lack of blinding (observer and patient) is a major concern. A larger number of patients is needed to increase the power of the study and determine if music truly has an effect on pain and anxiety. The small tertiary care unit provided services to patients of all ages (including neonates) from all over northern India; the majority of these patients have severe burn injuries and are either on inotropic or ventilator support. These patients were not included in this study because their anxiety and pain perception is not easily measurable, limiting the sample size of this pilot study. Further research in this dimension is under consideration, including a multicenter study. 

Conclusion

Pain management in burn patients, which has both physiological and psychological outcomes,²⁸ is an ever-growing challenge faced by the burn care team. Music therapy can help reduce the level of pain, anxiety, and opioid use during dressing change among burn patients. Similar studies can be conducted in different research settings and different patient populations to include a larger sample size, and the effect of instrumental and lyrical music types can be compared. Future research would benefit from blinded observer assessment of pain and anxiety. Also, self-report of pain and anxiety can be compared with objective assessment by an observer. Generating further evidence with a larger sample size to support current and historic findings and facilitate the development of an evidence-based standard music therapy protocol is recommended. n

References:

1. Venes D. Taber’s Cyclopedic Medical Dictionary, 20th ed. Philadelphia, PA: F.A. Davis Company;2005:306.

2. Smeltzer SC, Bare B. Brunner and Suddarth’s Textbook of Medical-Surgical Nursing. 10th ed. Philadelphia, PA: Lippincott Williams and Wilkins Publishers;2008:1704–1721.

3. Tortora GJ, Derrickson B. Principles of Anatomy and Physiology, 12th ed. Hoboken, NJ: John Wiley and Sons;2009:574–575.

4. Latarjet J, Choière M. Pain in burn patients. Burns. 1995;21(5):344–348.

5. Patterson DR, Hoflund H, Espey K, Sharar S; Nursing Committee of the International Society for Burn Injuries. Pain management. Burns. 2004;30(8):A10–A15.

6. de Jong AE, Middlekoop E, Faber AW, Loey NE. Non-pharmacological nursing interventions for procedural pain relief in adults with burns: a systematic review. Burns. 2007;33(7):811–827.

7. Miller K, Rodger S, Bucolo S, Greer R, Kimble RM. Multi-modal distraction. Using technology to combat pain in young children with burn injuries. Burns. 2010;36(5):647–658.

8. Kozier BJ, Erb G, Berman AT, Snyder S. Fundamentals of Nursing, 7th ed. London, UK: Pearson;2004:1178–1189.

9. Rudd E. Can music serve as a “cultural immunogen”? An exploratory study. Int J Qual Stud HealthWell-being. 2013;8:20597. doi: 10.3402/qhw.v8i0.20597.

10. Hsu KC, Chen LF, Hsiep PH. Effect of music intervention on burn patients’ pain and anxiety during dressing changes. Burns. 2016; 42(8):1789–1796. 

11. Son JT, Kim SH. The effects of self-selected music on anxiety and pain during burn dressing changes [in Korean]. Taehan Kanho Hakhoe Chi. 2006;36(1):159–168.

12. Spielberger CD, Gorsuch RL, Lushene R, Vagg PR, Jacobs GA. Manual for the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press;1983.

13. Whitehead-Pleaux AM, Baryza MJ, Sheridan RL. The effects of music therapy on pediatric patients’ pain and anxiety during donor site dressing change. J Music Ther. 2006;43(2):136–153.

14. Fratianne RB, Prensner JD, Hutson MJ, Super DM, Yowler CJ, Sandley JM. The effect of music-based imagery and musical alternate engagement on the burn debridement process. J Burn Care Rehabil. 2001;22(1):47–53.

15. Li J, Zhou L, Wang Y. The effects of music intervention on burn patients during treatment procedures: a systematic review and meta-analysis of randomized controlled trials. BMC Complement Altern Med. 2017;17:158. 

16. Hwang EY, Oh SH. A comparison of the effects of music therapy interventions on depression, anxiety, anger and stress on alcohol-dependent patients: a pilot study. Music Med. 2013;5(3):136–144.

17. Gutgsell KJ, Schluchter M, Margevicius S, et al. Music therapy reduces pan in palliative care patients: a randomized control trial. J Pain Sympt Manage. 2013;45(5):822–831.

18. Nilsson U. The anxiety- and pain-reducing effects of music interventions: a systematic review. AORN J. 2008;87(4):780–807.

19. Wang SM, Kulkarni L, Dolev J, Kain ZN. Music and preoperative anxiety: a randomized controlled study. Anesth Analg. 2002;94(6):1489–1494.

20. Henry LL. Music therapy: a nursing intervention for the control of pain and anxiety in the ICU: a review of the research literature. Dimens Crit Care Nurs. 1995;14(6):295–304.

21. Melzack R, Wall PD. Pain mechanisms: a new theory: a gate control system modulates sensory input from the skin before it evokes pain perception and response. Pain Forum. 1996;5(1):3-11.

22. Hawker GA, Mian S, Kendzerska T, French M. Measures of adult pain: Visual Analog Scale for Pain (VAS Pain), Numeric Rating Scale for Pain (NRS Pain), McGill Pain Questionnaire (MPQ), Short Form McGill Pain Questionnaire (SF-MPQ), Chronic Pain Grade Scale (CPGS), Short Form 36 Bodily Pain Scale (SF-36 BPS), and Measure of Intermittent and Constant Osteoarthritis Pain (ICOAP). Arthritis Care Res. 2011;63(Suppl 11):S240–S252. 

23. Vohra S. Manual for State-Trait Anxiety Test (STAT). New Delhi, India: Psy-com services; 2001:18–19.

24. Chi GC, Young A. Selection of music for inducing relaxation and alleviating pain. Holistic Nurs Pract. 2011;25(3):127–135.

25. Kahar NABA, Kanageswari S,Tay YB, Koh KX. A pilot study of the effects of music listening for pain relief among burn patients. Proc Singapore Healthcare. 2011;20(3):162–173.

26. Jafari H, Emami Zeydi A, Khani S, Esmaeili R, Soleimani A. The effects of listening to preferred music on pain intensity after open heart surgery. Iran J Nurs Midwifery Res. 2012;17(1):1–6.

27. Almerud S, Petersson K. Music therapy and patients on respirators. Intensive Crit Care Nurs. 2003;19(1):21–30.

28. Castro RJA, Leal PC, Sakata RK. Pain management in burn patients. Rev Bras Anestesiol. 2013;63(1):149–158.

Potential Conflicts of Interest: none disclosed

Ms. Rohilla is a Public Health Nursing Officer; Ms. Agnihotri is a tutor; Dr. Trehan is a lecturer; Dr. Sharma is Professor and Head, Department of Plastic Surgery; and Dr. Ghai is Principal, National Institute of Nursing Education, Post Graduate Institute of Medical Education and Research, Chandigarh, India. Please address correspondence to: Latika Rohilla, MSc, Public Health Nursing Officer, Post Graduate Institute of Medical Education and Research, Sector-12, Chandigarh, India. PIN code:160012; email: swatirohilla.19@gmail.com.

Abstract
Hospital-acquired pressure injuries (PIs) present a significant challenge to pediatric providers. Purpose: The purpose of this quality improvement program was to develop and implement a debrief protocol and to evaluate compliance with and the implementation of a comprehensive prevention bundle to decrease the overall incidence and severity of pediatric pressure ulcers (PUs)/PIs in a free-standing children’s hospital. Methods: As a member of the Children’s Hospitals Solution for Patients Safety national network, a PU Hospital Acquired Conditions (HAC) team was created in 2013, followed by the development and implementation of a PU occurrence debrief tool and discussion guide and implementation of multiple staff educational strategies and a comprehensive prevention bundle. The PU occurrence debriefing occurred within 24 to 48 hours of a PU. Incidence data were collected annually from 2014 until 2017. Results: Compliance on implementation and documentation of bundle elements ranged from 88% to 94%, and PU/PI incidence decreased by 30% from 2014 to 2016 and by 40% in 2017. The overall PU rate was 0.0057 in 2014, 0.0050 in 2015, 0.0036 in 2016, and 0.0023 in 2017; 65% of all PUs were device-related. Of those, >50% were related to respiratory devices, 25% to peripheral intravenous catheters/central lines, 10% to tracheostomies, and 15% to other devices. Respiratory device-related PUs decreased by 50% in the pediatric intensive care unit, by 80% in the neonatal unit, and eliminated completely in extracorporeal membrane oxygenation patients. Conclusion: The debriefing process, debriefing tool, educational programs, and prevention bundle reduced the rate of hospital-acquired PIs in pediatric patients and propagated a culture of safety.

Hospital-acquired Stage 3 and Stage 4 pressure ulcers (PUs) or pressure injuries (PIs), including those in pediatric patients, are reportable and considered “never events” by several national benchmarking organizations.¹ Reductions in reimbursements for health care-acquired PUs have been implemented by the Centers for Medicare and Medicaid Services (CMS) since 2012.¹ The reported prevalence of pediatric PUs ranges from 0% to 43%^2-4; Baharestani and Ratliff¹ reported an incidence of 23% in neonatal units. In a multisite study of >5000 children in 9 pediatric critical care units, the overall PU incidence was 10.2% (range 0.8%–17.5%) across sites.⁵ Curley et al⁶ found a 27% incidence of PUs in a study of 3 critical pediatric care units; most patients (97%) had Stage 1 or Stage 2 injuries.⁶ In a  survey involving >1000 children in a pediatric hospital, McLane et al⁷ reported a 4% rate of hospital-acquired PUs. Razmus et al³ reported a PU prevalence of 1.4%, 1.1% of which were hospital-acquired.

New medical technologies have increased the survival rates of extremely preterm newborns and infants with critical congenital conditions such as central nervous system (CNS) injury, cardiac diseases, or mechanical ventilation; respiratory devices, central lines, internal catheters, and tracheostomies contribute to cutaneous injuries such as PI.² Intrinsic factors such as immobility, altered sensation, sedation, inactivity, nutritional deficiencies, and immaturity of the skin also are known to heighten the PI risk.² Few risk assessment tools have been modified for pediatric patients and none have been validated that include medical devices as a risk factor for PI. Although adult practitioners are fairly proficient at managing PI, medical and nursing education on pediatric PIs is generally considered inadequate.¹

So far, no specific approach/tool has been shown to have an impact on staff education/management of pediatric PI. The purpose of this quality improvement project was to develop and implement a debrief protocol and evaluate compliance with and the implementation of a comprehensive prevention bundle to decrease the overall incidence and severity of pediatric PUs/PIs in a free-standing children’s hospital.

Methods
Setting. Cohen Children’s Medical Center (New Hyde Park, NY) is a 200-bed, free-standing, quaternary care academic facility that serves patients from around the region, nation, and globe. The 57-bed, neonatal intensive care unit (NICU) and 34-bed pediatric intensive care unit (PICU) are state-of-the-art facilities serving children with the most complex conditions, offering extracorporeal membrane oxygenation (ECMO), therapeutic hypothermia, and innovative cardiac, neurologic, and orthopedic surgery. Extensive bone marrow transplant, cancer, organ transplant, and level 1 trauma patients are routinely treated on the medical floors and intensive care units (ICUs).
 

History of preventive concern (2012-2013). Children’s Hospitals Solution for Patients Safety (CHSPS) national network was launched in 2011. The goal of this initiative was to prevent hospital-acquired harmful conditions through the use of standard definitions; training in the Model for Improvement and Plan/Do/Study/Act cycles8,9; the creation, implementation, and measurement of event prevention bundles; data analysis; and transparency across the collaborative. In 2012, Cohen Children’s Hospital became the first pediatric hospital in New York state to join CHSPS. The program was awarded a multiyear contract by the CMS as part of its Partnership for Patients initiative, a priority project designed to reduce hospital inpatient harm by 40% over a 3-year period. A new safety program, “Commit to Zero,” that addresses reducing errors to none, was introduced; it includes implementing daily safety briefings, safety behavior education, leadership rounding, a “Great Catch” employee recognition program, development of hospital-acquired condition (HAC) teams, and a safety coach program. In this framework, a PU HAC team was created and a PU Prevention Bundle was developed and propagated throughout the hospital.

Before the team’s formation, surveillance of PUs and data collection were inconsistent. Five (5) random, 1-day studies revealed a PU prevalence of 3% on medical floors, 15% in critically ill pediatric patients, and 8% in the neonatal unit. Prevalence was calculated by dividing the total number of hospital-acquired PUs by the total number of patients surveyed times 100. Total rate of PUs (total PU per year/total number of admissions) was 0.0069 for 2013 (likely underreported because data collection was inconsistent).
 

PU Prevention Bundle. The PU Prevention Bundle focused on the National Pressure Ulcer Advisory Panel (NPUAP) recommended elements10: skin assessment, repositioning, device rotation, bed support surfaces, moisture management, and nutrition. All nursing staff were required to complete mandatory education modules presented as didactic lectures and computerized webinars with postlecture tests that addressed skin assessment scales, PU staging, risk, prevention, management, and bundle element documentation. Random patients’ computerized records were reviewed for each bundle element. The bundle review sheet, completed and assessed by PU HAC co-leads, revealed inconsistencies in implementation and documentation of bundle elements stemming from suboptimal knowledge of PU prevention strategies, staging, and treatment. According to the subjective opinion of the co-leads after interviewing staff, staff were hesitant to report PUs; thus, reports were not timely or accurate. As a result, opportunities for prevention often were missed.
 

Evaluation of gaps (2013). In order to accomplish the primary goal of PU/PI reduction, the members of the PU HAC group developed a key driver diagram (see Figure 1), identifying 4 primary drivers: prevention process, accurate recognition, timely and accurate reporting, and change of culture. Didactic lectures and mandatory computer presentations were recognized by the PU HAC co-leads and nursing education leadership as ineffective in bridging theoretical knowledge with practical care. Most nurses did not feel comfortable implementing preventive measures, staging, diagnosing, and treating PUs, despite the initial wave of mandatory education. Many staff members, including nurses and clinicians, were complacent about existing injuries. A culture of prevention, urgency of reporting, and preoccupation with harm was not prevalent. Some staff did not feel comfortable reporting PUs, fearing punitive reaction and blame.

The team’s evaluation of existing practices as part of subjective discussion between nursing education leadership, quality departments, and the PU HAC co-leads suggested a need for a different approach that tied all driver elements together in a safe but realistic environment.

As evidenced by the lack of information in the literature prior to 2014, quality improvement is a relatively new concept for medicine in general and very new to PI specifically. In the past, quality was generally defined as achieving best clinical outcomes and safety was defined as not harming patients; however, these concepts have converged over the last decade and the idea of “aiming for zero harm” has emerged.^9,11 According to a review of descriptive studies discussing core concepts of quality improvement and patient safety,¹¹ 3 key principles are required for achieving zero harm: development of safety culture, staff accountability, and transparency. Simulation-based or event-based medical education has been shown in reviews of descriptive core concepts and studies involving on medical simulation^12-14 to be useful for creating a safety culture in events of critical resuscitations, trauma, and negative surgical scenarios. Appropriate feedback (including debriefing) has been shown to be central to this process.^13-15 Because no literature was available that described the role of debriefing in PU reduction, our group borrowed the concept of simulation and debriefing from industries such as aviation, automotive, and trauma where it had been used successfully.^12,13,16
 

Conceptual framework — debriefing. Feedback is a post-experience analytic process. It involves discussion and analysis of an experience and reviewing and integrating lessons learned into one’s knowledge bank. A review of the literature¹² shows that in adult teaching, “active” participation is an important factor in effective learning. Critical learning is achieved by reflecting on experiences, recognizing strengths and weaknesses, and reviewing alternative choices. Feedback is an integral part of medical debriefing.¹²
 

Debriefing is defined by Merriam-Webster as questioning someone about a completed mission with an aim of obtaining useful information. Generally, it is a process of inquiry and evaluation. First developed as a formal process in the military during World War II when troops were gathered after missions to reconstruct and describe what happened in an effort to reduce psychological stress,¹² post combat discussions or “performance critiques” eventually became a fundamental component of battle simulation exercises for soldiers in training.^12,13 The focus of after-action reviews shifted over time from subjective emphasis on error identification to nonpunitive guided group discussions and self-reflection. Debriefing also has deep roots in aviation, mass transportation, and nuclear power, areas in which overt or latent human and system weaknesses can lead to loss of life.¹³ As a result, it appears these industries have developed cultures of safety that are far less tolerant of conditions that place human lives at risk than are prevalent in the health care industry.

A relevant review found effective communication is crucial to patient safety.¹⁶ The Joint Commission found communication failures are the root cause of 60% to 70% of sentinel events. In health care, it is important to find ways to open lines of communication. Post-event debriefings in medicine are defined as a “discussion of actions and thought processes after an event to promote reflective learning and improve clinical performance”; they are facilitated discussions of a clinical event focusing on learning and performance improvement.¹⁶ Essential elements of post-event debriefings include active self-learning, a primary intent for improvement, reflection on specific events, and the inclusion of input from multiple team members.¹⁴ Post-event debriefings are a foundational behavior of high-performing teams.¹⁶

The American Heart Association endorses debriefing as a strategy to improve cardiopulmonary resuscitation quality.^15,16 A recent meta-analysis^11,17 found organizations can improve individual and team performance by up to 25% by conducting effective debriefings; in these simulation-based studies, debriefing has been associated with enhancements in team performance and improvements in both technical and behavioral skills. In clinical medicine, post-event debriefings have been shown to increase overall performance, reduce the frequency of equipment-related problems, and improve communication and teamwork.
 

Debrief Tool (2013–2014). The PU HAC lead (an intensive care attending physician)  developed the “PU Occurrence Debrief” tool, a 1-page summary to guide a discussion after PU events in real time led by PU HAC leads (see Figure 2). A frequent attendee of post event debriefs, she searched the literature for information on safety approaches, adult learning, and debriefing specific to medicine and applied the concept to PU staff education and management. She presented her instrument to the group, her co-lead, and a quality partner, and the tool was trialed in a few pilot instances where the concept was well-received. Staff were supportive and appreciative of the framework for discussion. Concomitantly, staff were re-educated on the reporting process via an electronic medical database, with emphasis on time (goal to report within 12 to 24 hours), accuracy, and completeness.

Debrief process. The PU implemented debrief process is triggered by assistant nurse managers on each unit, who report each event via a centralized computerized system. These reports are transmitted to a quality department member and to the PU HAC lead. The PU HAC lead then conducts the debrief with the team taking care of the affected patient within 24 to 48 hours of the occurrence. This short gap serves to minimize recall bias, allowing timely patient care and potentially addressing global issues such as equipment failure, gaps in communication, lack of support, or staffing problems. To improve team performance, all members of the team who are physically present participate in the post-event debriefing. The goal is to include all staff, patient, and/or family if appropriate, because each offers a unique perspective and each perspective is important to understanding the individual and team strengths and weaknesses.

Three (3) stages of debriefing process are described in literature¹²:
Stage 1: The reaction phase: Allow for initial responses. Summarize what happened. Ensure common understanding.
Stage 2: The understanding phase: the heart of the debriefing inquiry and analysis. Inquire about assumptions/actions.
Stage 3: The summary phase: Lessons learned. Provide take-away points for the future as soon as possible after the event to review the facts, identify system errors, and improve the process with goals to reduce future recurrences.

We found debriefing to be the most effective when structured and facilitated. This structure is based on the “Gather, Analyze, Summarize” approach to debriefing endorsed by the American Heart Association and incorporated in its life support courses.¹⁷ We begin by reviewing the details of the event, examining the patient and discussing staging accuracy. This is followed by discussion of the risk factors, mitigating circumstances, presence or absence of preventive measures, and the nature of treatment to minimize sequelae. Finally, a summary or a simulation for the next time is generated, with emphasis on “What would I do next time?” We try to empower staff to speak frankly and offer their opinions and suggestions. The atmosphere of psychological safety, where team members feel secure in critically analyzing their own performance, is best achieved when the debriefing proceeds in a nonpunitive fashion. We focus on high-value issues such as adherence to prevention guidelines, equipment or assessment issues, and appropriateness of treatment. Debriefing is kept brief, taking no more than 10 minutes.
 

Bundle implementation and tracking. The consistency of bundle element implementation and documentation has been tracked since 2014. Random audits of electronic PU/PI prevention documentation are performed every quarter (5 patients are selected from every unit) by PU HAC champions and documented on a PU HAC Bundle Audit sheet; the audits focus on completeness, accuracy, meaningful interventions based on the assessed risk, and (if the patient develops a PU/PI) accuracy and timing of the reporting. Audit questions require Yes/No responses (eg, Performed vs. Not performed, Interventions documented vs. Not documented, based on the patient’s clinical risk). When the audit is complete, the patient PU/PI assessment score as well as clinical risk is determined by the auditors first and then compared to the documented data. The PU/PI champions are well-trained and are able to assess the accuracy of the bedside nurses’ clinical judgment.
 

Data collection and analyses. The primary outcome was total  number of hospital-acquired PUs/PIs per year. New PUs/PIs occurring after admission were counted and prevalence was calculated as the number of PUs/PIs divided by the total admissions per year. Key process measures (bundle compliance, preventive measures implementation, and timeliness and accuracy of the PU/PI incident report) were assessed quarterly via random patient audits and documented in a hospital-based Health Insurance Portability and Accountability Act-compliant database  via a computerized data collection tool.

Results
Since implementation of the program and debrief protocol, compliance on implementation and documentation of bundle elements has ranged between 88% and 94%, with an ongoing aim to maintain >90%.

Following the initiation of the debrief process and coupled with continuous education efforts, we have seen a significant decrease in hospital-acquired PIs. The total number of PUs decreased 30% from 2014, and as much as 40% in 2017 (from 2014; see Figure 3). The overall rate of PUs/PIs (total PUs/PIs per year/total number of admissions) was 0.0057 in 2014, 0.0050 in 2015, 0.0036 in 2016, and 0.0023 in 2017. While 60% of hospitalizations were short stay (2 to 3 days), 70% of PUs occurred in longer term admissions, with an average time to PU occurrence of 8 ± 2 days. In 2017, PICU patients accounted for 40% of PIs; 37% occurred on medicine/postsurgical floors, 13% in the NICU, 5% in operating rooms, and 5% in emergency room/other (see Figure 4). Device-related PUs/PIs accounted for 65% of all PUs/PIs, consistent with previous reports2-5; of those, >50% were related to respiratory devices, 25% to peripheral intravenous/central lines, 10% to tracheostomies, and 15% to other devices (see Figure 5).

As of the writing of this article, most PUs/PIs reported in the hospital have been in the early stages (Stage 1 and Stage 2), suggesting our approach has resulted in improved surveillance, prevention, and timely recognition in the 3 years of study (2014–2017) (see Figure 3). PU/PI occurrence has been reduced in ECMO patients from 40% in 2013 to 0% in 2016–2017), and PICU respiratory device-related PIs have decreased by 50% over the last 2.5 years. No neonatal respiratory device-related injuries have occurred in the last 2 years. The trajectory for 2018 follows the same pattern.

Discussion
In 1999, the Institute of Medicine released its report, To Err is Human.¹⁸ It stressed that creating a learning environment is crucial for the development of a safe and reliable organization. The report recommended 5 elements. First, simulation should be used in training of staff performing safety-critical functions. Sophisticated gadgets are unnecessary; rather, practice to perfection and effective coaching should be part of training. Second, a culture of reporting errors is required. Third, the reporting of errors must be accountable for both the occurrence and the actual notification of errors but free of shame and blame. Fourth, good communication among disciplines is important. The fifth element is to analyze errors and identify their roots.

As indicated in the Driver diagram (see Figure 1), the debrief process interacts with several of the key drivers and processes in PU/PI prevention. The debrief process brings discussion (of a present case) and simulation (for future cases) to the forefront of learning. This has proven to be effective in fostering small multidisciplinary group learning, facilitated by the PU HAC lead.
The “timely reporting” component was much more difficult to implement. The support of hospital leadership and close collaboration with the hospital quality department were crucial for implementing timely reporting and effecting culture change. The culture had to be changed from perceptions of fear of punitive actions and thoughts that maybe the ulcer will improve on its own and that the ulcer is not life-threatening to positive feelings with regard to recognizing the problem, that this concerns a process not an individual, and that we should feel team pride for our efforts. At the same time, we did not want staff to take PUs/PIs lightly. Multiple guidelines, policies, and safe-guards have been put in place stemming from root cause analyses of PU/PI events. Still, we recognized that certain PUs/PIs are unavoidable and “occurrence debrief” would remain an essential opportunity to reinforce education, review, and plan for “the next time.”

We were able to apply all 5 Institute of Medicine¹⁸ elements in our journey to reduce the rate of HACs in pediatric patients. We nurtured the internal motivation and competency of our staff through education, discussion, availability, and development of necessary tools. Today we are focused on continuing and strengthening our clinical quality and safety program. The future goals are to sustain improvement and to strive for better outcomes.

Limitations
The biggest limitation of this descriptive study is that data are limited to 1 hospital’s effort to decrease the incidence of PUs/PIs. The quality improvement models we adopted, the campaigns we launched, and the debrief tool we developed worked in our environment. The tool has not been validated outside of our hospital, and further studies with multiple sites would be needed to see if the concept and the tool itself can be generalizable. We cannot sufficiently underscore the importance of the continuous education that took place. It is difficult to analyze if education in itself is a variable that affected the success of the debriefing concept. Having said that, we believe the tool and the approach improved staff ability to retain and to apply the information given.

Conclusion
A quality improvement project was undertaken to develop and implement a debrief protocol and to evaluate compliance with and the implementation of a prevention bundle to decrease the overall incidence and severity of pediatric PUs/PIs in a free-standing children’s hospital. Through the plan’s initiation in 2013 and implementation in 2014, we learned that implementing and sustaining quality improvement initiatives to decrease PUs/PIs is a Herculean task that requires many tools. The strength of this program is its relatively low cost (time commitment from qualified personnel to conduct the debrief), simplicity, and ease of implementation. During the 3 years of implementation, the rate of hospital-acquired PUs/PIs decreased by 40% and compliance with PU/PI prevention bundle is consistently ~90%, the targeted aim. In our experience, the concept of debriefing (utilizing a debrief tool) is an effective way to promote PU/PI education, enhance preventive efforts, reduce hospital-acquired injuries, and propagate a culture of safety. Additional research to examine the effect of this approach on preventing PIs in adults and ascertain staff education needs and the effect of education delivery models is warranted.

Abstract
Preventing, identifying, and treating deep tissue injury (DTI) remains a challenge. Purpose: The purpose of the current research was to describe the characteristics of DTIs and patient/care variables that may affect their development and outcomes at the time of hospital discharge. Methods: A retrospective, descriptive, single-site cohort study of electronic medical records was conducted between October 1, 2010, and September 30, 2012, to identify common demographic, intrinsic (eg, mobility status, medical comorbidities, and incontinence), extrinsic (ie, surgical and procedural events, medical devices, head-of-bed elevation), and care and treatment factors related to outcomes of hospital-acquired DTIs; additional data points related to DTI development or descriptive of the sample (Braden Scale scores and subscale scores, hospital length of stay [LOS], intensive care unit [ICU] LOS, days from admission to DTI, time in the operating room, serum albumin levels, support surfaces/specialty beds, and DTI locations) also were retrieved. DTI healing outcomes, grouped by resolved, partial-thickness/stable, and full-thickness/unstageable, and 30 main patient/treatment variables were analyzed using Kruskal-Wallis, chi-squared, and Fischer exact tests. Results: One hundred, seventy-nine (179) DTIs occurred in 141 adult patients (132 in men, 47 in women; mean patient age 64 [range 19–94]). Of those patients, 110 had a history of peripheral vascular disease and 122 had hypertension. Sixty-nine (69) DTIs were documented in patients who died within 1 year of occurrence. Most common DTI sites were the coccyx (47 [26%]) and heel (42 [23%]); 41 (22%) were device-related. Median hospital LOS was 23 (range 4–258) days and median ICU LOS was 12 (range 1–173) days; 40 DTIs were identified before surgery and 120 after a diagnostic or therapeutic procedure. Data for DTI outcome groups at hospital discharge included 28 resolved, 131 partial-thickness/stable, and 20 full-thickness/unstageable; factors significantly different between outcome groups included mechanical ventilation (15/42/12; P = .01), use of a feeding tube (15/46/12; P = .02), anemia (14/30/9; P = .005), history of cerebrovascular accident (12/27/7; P = .03), hospital LOS (67/18/37.5; P <.001), ICU LOS (23/10/12; P = .03), time-to-event (13.5/8/9; P = .001), vasopressor use after DTI (13/31/11; P = .003), low-air-loss surface (10/9/3; P = .005), and device-related (14/24/4; P = .002). Conclusion: DTI risk factors mirrored those of other PUs, but progression to full-thickness injury was not inevitable. Early and frequent assessment and timely intervention may help prevent DTI progression.

Despite progress in pressure ulcer (PU) knowledge during the past decade, the identification and treatment of deep tissue injury (DTI) continues to present a challenge for bedside clinicians. At the time this study was conducted, the definition of a DTI from the National Pressure Ulcer Advisory Panel (NPUAP) was “Purple or maroon localized area of discolored intact skin or blood filled blister due to damage of underlying soft tissue from pressure and/or shear. The area may be preceded by tissue that is painful, firm, mushy, boggy, warmer, or cooler as compared to adjacent tissue. DTI may be difficult to detect in individuals with dark skin tones. Evolution may include a thin blister over a dark wound bed. The wound may further evolve and become covered by thin eschar. Evolution may be rapid exposing additional layers of tissue even with optimal treatment.”¹

In their descriptive article, Black et al² described conditions not considered to be DTIs. Although the conditions the authors reported included intact skin that appeared purple or maroon, they also developed a process for differential diagnosis of the condition. Of note: During the period covered by the current study, the term pressure ulcer was the official term utilized in clinical practice and cited research likewise utilizes that term. As such, the terms pressure ulcer and deep tissue injury are used in this study. In 2016, the NPUAP revised the terminology from pressure ulcer to pressure injury (PI) to encompass the spectrum of tissue damage due to pressure. Since 2017, the Mayo Clinic (the employer of all authors) has adopted the updated NPUAP nomenclature.1^,2

The accurate and timely identification of DTI is important for several reasons. Based on anecdotal experience, the current authors have found that early discovery of DTI allows prompt identification of possible causes, initiation of treatment, and potential development of preventive strategies. In addition, 24 to 72 hours can lapse between the precipitating pressure event and the onset of purple or maroon skin.³ This delayed manifestation becomes particularly important when the precipitating event occurred before the patient’s admission, yet the DTI appears beyond the 24-hour window for present-on-admission status, at which point the admitting facility becomes financially responsible for care. In Minnesota, for example, the law requires that this complication be publicly reported.⁴

Background
In 2005 and 2013, the NPUAP consensus conferences focused on DTI in order to gain a deeper understanding into the causes, manifestations, and evolution of these injuries. A substantial advance was the understanding of DTI as more of a bottom-up than top-down phenomenon that was related to the deformation of deep tissue with muscle ischemia and reperfusion injury under intact skin, despite adequate surface pressure relief.^5-9 The effectiveness (or lack thereof) of nursing interventions on the development of DTIs and their outcomes is highly relevant to wound, ostomy, and continence (WOC) nursing because the WOC nurse is usually the person consulted to assess and collaboratively manage these wounds over time.

Several recent studies describe characteristics of patients with DTI.^9-11 In a large, annual, inpatient prevalence study, VanGilder et al¹⁰ surveyed between 79 000 and 92 000 patients from 2006 to 2009. The survey showed the overall and nosocomial PU prevalence decreased by approximately 1% in 2009 after remaining fairly constant in the years 2006 to 2008. However, the proportion of suspected DTIs increased 3-fold to 9% of all observed PUs in 2009; these wounds were more prevalent than either Stage 3 or Stage 4 PUs, with the heels the most prevalent DTI site. The authors of the study¹⁰ acknowledged that differences in demographic characteristics (that were not available to them) might help clarify the cause of more suspected DTIs in the study and provide insights to guide prevention, treatment, and design of future clinical studies. A prospective, multisite, exploratory study by Richbourg et al9 described the progression from suspected DTI to full-thickness skin loss and explored associated conditions of a sample of 40 inpatients. In this smaller study,⁹ the sacrum was shown to be the most frequent site of DTIs.

Since 2001, the NPUAP and others have made a substantial effort toward clarifying and refining the definition, clinical characteristics, and causes of DTI.¹² As part of a retrospective review of a 2-year cohort of patients who developed DTIs, Sullivan13 reviewed and described patient factors contributing to the evolution and outcomes of these injuries.

In current study authors’ institution, DTIs are the second most common documented type of PU behind Stage 2 PUs. Although DTIs can be devastating injuries, progression to a full-thickness PU is not inevitable; DTIs have been observed to heal or improve before hospital discharge. The purpose of the current research was to describe characteristics associated with the development of DTIs and to better understand how these characteristics might be associated with PU outcomes at the time of hospital discharge. The authors elected to evaluate each DTI rather than each patient as the variable of interest, because patients with multiple DTIs did not acquire them all at the same time or in the same manner, nor did the injuries all resolve to the same degree.

Research questions were: 1) What are common demographic and intrinsic factors (eg, mobility status, medical comorbidities) of patients who develop a DTI during hospitalization? 2) What are common extrinsic factors (ie, surgical and procedural events, medical devices, head-of-bed [HOB] elevation, incontinence) of patients who develop a DTI during hospitalization? 3) What is the care and treatment applied to DTIs that resolve, improve by hospital discharge, or progress to a full-thickness PU?

Methods
This single-site, descriptive, retrospective analysis of electronic health records (EHRs) was approved by the Mayo Clinic Institutional Review Board (Rochester, NY). The study setting was a 2207-bed Midwestern academic medical center composed of a 2-campus acute care hospital designated by the American College of Surgeons as a level 1 trauma center. Inclusion criteria were documentation of DTI in hospitalized, adult patients who were 18 years or older at the time of admission, hospitalized for >1 day, and who had provided research authorization. Exclusion criteria were existing DTIs at admission, DTI in patients with a length of stay (LOS) <1 hospital day, and patients who had not agreed to review of their EHRs for research purposes. Suspected DTIs were excluded from the study analysis after the chart was reviewed by the study team, including a board-certified rehabilitation physician/physiatrist.

Additional data points related to DTI development or descriptive of the sample (age, gender, body mass index [BMI], race, comorbidities, Braden Scale scores and subscale scores, hospital LOS, intensive care unit [ICU] LOS, days from admission to DTI, time in the operating room [OR], serum albumin levels, support surfaces/specialty beds, and DTI locations) also were retrieved from the EHR on each patient by Mayo’s Division of Biomedical Statistics and Informatics. Information on other factors thought to be related to PI or DTI development (ie, devices, incontinence, vasopressor use, procedures exclusive of the OR, OR duration, comorbidities) and factors used in treatment following DTI occurrence (ie, dressings, noncontact low-frequency ultrasonic [NLFU] therapy) were manually retrieved by the study team for analysis. Specialty information related to physical therapy or the OR was examined by the appropriate specialty investigator. Several factors (ie, mechanical ventilation, need for a feeding tube) were used as surrogates for prolonged HOB elevation (≥30˚).¹⁴ The authors did not attempt to (nor could they) accurately capture degrees of HOB elevation from a retrospective chart review; however, from clinical experience, these factors could be used as rough surrogates for elevated HOB.
 

Data analysis. All data were extracted from the EHR and entered into an Excel (Microsoft Corp, Redmond, WA) database for statistical analysis. Statistical analyses were based on individual DTIs as the variable of interest because each was considered as an independent injury. For patients with multiple DTIs, not all DTIs occurred under the same circumstances or on the same date or resolved to the same degree.

Continuous features (age, hospital and ICU LOS, time to event, and Braden Scale for Pressure Risk Score) were summarized with medians, interquartile ranges (IQRs), and ranges. Categorical features (gender, race/ethnicity, and comorbidities) were summarized with frequency counts and percentages.

The outcome categories at time of hospital discharge were resolved, stable, partial-thickness, full-thickness, and unstageable. Because the small numbers of patients in 5 subgroups made meaningful statistical comparisons impossible, the subgroups were combined into 3 larger categories: resolved (defined as no evidence of DTI by EHR documentation), stable/partial-thickness (DTI fading toward normal pigmentation, no margin expansion at time of discharge by photo or EHR documentation, or epidermal loss not extending through  the dermis), and full-thickness/unstageable (progression to Stage 3, Stage 4, or unstageable requiring reporting to Minnesota Adverse Health Care Events by photo or EHR documentation).

Comparisons of factors between DTIs that resolved, remained stable/became partial-thickness, or progressed to full-thickness/unstageable by hospital discharge were evaluated using Kruskal-Wallis, chi-squared, and Fisher exact tests. All analyses were performed using the SAS software package, version 9.3 (SAS Institute Inc, Cary, NC). P values <.05 were considered statistically significant.

Results
Between October 1, 2010, and September 30, 2012, 928 patients were documented with DTIs; of these, 151 were excluded because patients had not provided research authorization. An additional 592 DTIs were excluded as being present on admission or not verified by photo or note by an Advanced Practice Registered Nurse Clinical Nurse Specialist (APRN CNS), along with 6 additional ulcers lacking documentation. After the exclusions, 179 documented DTIs remained and were used as the cohort for analysis in this study.

Patient data. Of the 141-patient cohort (median age 64 [range 19–94] years), 25 (18%) had multiple DTIs: 16 had 2 injuries, 7 had 3 injuries, and 2 had 5 injuries. Characteristics of the cohort of patients who developed DTIs are described in Table 1, and the summary of patient characteristics (demographics and intrinsic and extrinsic factors) relative to DTI outcomes is shown in Table 2. The median BMI was 27.0 (range 12.7–61.6), and the median total Braden scores were 14 (range 8–23) at 3 days, 13 (range 8–22) at 2 days, and 13 (range 7–23) at 1 day before DTI appearance. The median hospital LOS was 23 (range 4–258) days and median ICU LOS was 12 (range 1–173) days, with a median time from admission to DTI event of 9 (range 1–129) days. The primary comorbid condition was hypertension (122, 68%), followed by peripheral vascular disease (110, 61%), congestive heart failure (71, 40%), diabetes (56, 31%), and acute renal failure requiring dialysis (56, 31%). Other factors that were identified and are known to contribute to DTI development and progression included altered mobility measured by a Braden subscale score of <2 (127, 76%), presence of incontinence (109, 66%), surgical intervention before DTI (40, 22%), diagnostic and/or therapeutic procedures requiring limited mobility (120, 66%), and use of medical devices (41, 23%).

Braden total and subscale scores were assessed 72, 48, and 24 hours before DTIs were identified. At all 3 times, the median total scores showed a high risk for PU development, with only a minimal downward trend. At 72 hours before DTI, the median total score was 14 (range 8–23), and at 48 and 24 hours the median total score was 13 (range 48 hours, 8–22; 24 hours, 7–23). No statistically significant difference was noted in the Braden total or subscale scores 72, 48, or 24 hours before DTI was identified, with the exception of the nutrition subscale score.
 

DTI data. Among the 179 DTIs studied, 47 (26%) occurred on women and 132 (74%) on men; 162 (91%) occurred on non-Hispanic whites, 7 (4%) on Hispanic whites, 2 (1%) on American Indian/Alaskan Natives, 2 (1%) on blacks, 1 (<1%) on an Asian, and 2 (1%) “other.” Two (2) patients who had a total of 3 (1.5%) DTIs did not disclose their race.

DTIs occurred in a variety of locations, including the sacrum (11, 6%), intergluteal (4, 2%), buttock (25, 14%), trochanter (1, 1%), ischium (1, 1%), calcaneal (42, 23%), coccyx (47, 26%), and other (48, 27%). There were no mucous membrane injuries in the “other” category.

Sixty-eight (68) DTIs occurred in patients who had received some care in the ICU, with a median ICU stay of 12 (range 1–173) days. Sixty-nine (69) of the DTIs occurred in patients who died within 1 year of the injury, with 38 of these DTIs occurring in patients who died during hospitalization.

Among the 179 DTIs, 41 (22%) were device-related: 16 (39%) were related to compression wraps, 1 (2%) to a sequential compression device, 2 (5%) to a walking splint-boot, 5 (12%) to a cast or splint, 3 (7%) to a cervical collar, 5 (12%) each to oxygen devices and orthotic braces, 3 (7%), to a bed/chair, and 1 (2%) to a sling.

NLFU MIST therapy (Cellularity, Inc, Warren NJ) was used for 56 (31%) DTIs, 2 before and 54 after DTIs developed. The therapy was used before a DTI developed for patients with a partial-thickness wound that subsequently developed additional tissue damage with DTI occurring postoperatively or post-procedurally.

Thirteen (13) DTIs were on patients who had some form of documented skin alteration at the same anatomic location before the DTI was identified (mean 77.5 hours; median 49 [range 20–192] hours). Of the 13 skin alterations documented, 8 were erythema, 3 were Stage 1, and 2 were Stage 2 PUs.

For 118 (66%) of the DTIs, patients had undergone some type of diagnostic or therapeutic procedure within 72 hours of DTI development, including computed tomography scan (38, 32%), radiography (35, 30%), ultrasonography (27, 23%), echocardiography (22, 19%), and interventional radiologic procedures (15, 13%).

DTI outcomes. At the time of hospital discharge, 28 DTIs had resolved, 131 were partial-thickness/stable, and 20 were full-thickness/unstageable. There were no statistically significant differences between the demographic characteristics related to the DTI outcome groups. Factors that did significantly differ between outcome groups included the intrinsic factors of a history of cerebrovascular accident (12/27/7; P = .03) and anemia after DTI (14/30/9; P = .005). Extrinsic factors included mechanical ventilation before DTI (15/42/12; P = .01) and after DTI 15/34/12 (P < .01), use of a feeding tube after DTI (15/46/12; P = .02), hospital LOS (67/18/37.5; P <.001), ICU LOS (23/10/13; P = .03), time to event (13.5/8/9; P = .001), vasopressor use after DTI (13/31/11; P = .003), low-air-loss surface after DTI (10/9/3; P = .005), and device-related injury (14/24/4; P = .002). (It is important to recall mechanical ventilation and feeding tube use were surrogates for HOB elevation to 30° or more.) All other intrinsic and extrinsic factors were not significantly different between the outcome groups, including surgeries and procedures.

In the patients who died, the DTIs appeared between 0 to 63 days before death. Although some of these may have been Kennedy Terminal Ulcers, insufficient documentation of the characteristics (eg, ulcer shape, borders, color) precluded the ability to confirm this diagnosis.15 A comparison of variables relative to DTI outcomes is summarized in Table 2.
 
Discussion
This study was begun just before publication of the retrospective review by Sullivan13 of suspected DTI evolution in adult acute care patients, and the results share several similarities. Similar to Sullivan, the initial intent of this study was to describe and understand the characteristics of patients who developed a DTI and the contributing intrinsic and extrinsic factors involved. The long-term intent was to use this knowledge to inform clinical practice and future research. The current retrospective study examined factors associated with the development and outcomes of DTI during a 2-year period. Although no definite intrinsic or extrinsic factors could be directly attributed to the development of DTI, incidence of DTIs was higher in patients with cardiovascular diseases (including hypertension, peripheral and cerebrovascular disease, congestive heart failure) and diabetes mellitus.

In the current cohort of DTIs, (11%) progressed to full-thickness/unstageable at the time of discharge. The majority (131, 73%) were stable or were categorized as sloughing of the epidermis at the time of patient discharge. Twenty-eight (28, 16%) of the DTIs were completely resolved; 69 DTIs (38%) occurred in patients who died at a median of 51 days after the DTI was first identified, 38 (55%) of whom were still hospitalized. Cohorts in this and previously published work13 both had similar patterns of comorbidities, with a high incidence of patients who had cardiac, vascular, and/or renal diseases or diabetes mellitus.

Pre-DTI identification assessment. In 13 (7%) of the 179 DTIs, a skin alteration was documented before the DTI developed, and most of these (8 DTIs) initially showed blanchable erythema. The range of time that erythema was present varied greatly (ie, from 20 hours to 192 hours). For 3 DTIs, the PU was initially identified as Stage 1 and was present for 22 to 29 hours before being documented as a DTI. Two (2) DTIs were initially identified as Stage 2, with 1 present 25 hours before the DTI was documented. In this case, the patient underwent a procedure in which he was positioned over the Stage 2 PU; 48 hours after the procedure, the wound bed became dark purple and nonblanchable based on physical exam by the APRN CNS. Farid et al¹¹ suggested blanchable erythema develops 168 to 336 hours before DTI formation, a range much higher than that of the current study. In the authors’ EHR system, the CNS on the unit was automatically alerted of skin alterations documented in the EHR to trigger the need for further assessment and intervention, which may explain this difference.

Chart information of the patients with the 13 DTIs and pre-existing skin alterations was reviewed to determine the types of diagnostic and surgical procedures associated with injury development. In a pilot study by Honaker et al,¹⁶ precipitating events occurred 1 to 5 (mean 2.41) days before DTI skin changes were noted. Therefore, the current study reviewed records for the 5 days before DTI documentation and identified 7 DTIs in patients who had surgical procedures lasting from 30 minutes to 5 hours. For 1 DTI, erythema was present for 168 hours before the DTI was identified, and the patient had undergone 7 diagnostic procedures in the 5 days before the DTI was documented. Another had erythema documented for 49 hours before DTI appearance; this patient had undergone 5 diagnostic procedures within that period of time.

Surgical and procedural considerations. In a surgical environment, DTI prevention involves identifying patients at risk through a skin assessment and reliably implementing prevention strategies. In the inpatient setting in the United States, the Braden Scale is a widely used tool for identifying at-risk surgical patients.¹⁷ However, in an OR environment, the preoperative Braden Scale score is inherently skewed because most patients have lost the ability for self-protection and communication. The Scott Triggers tool was developed as a predictive scale developed on evidence-based factors specifically tailored for high-risk perioperative PI or DTI development.¹⁷ The Munro Tool for Pressure Ulcer Risk-Assessment Scale for Perioperative Patients was developed as both a communication tool and documentation of risk assessment in each phase of perioperative care.¹⁸ The Munro Tool incorporates 15 evidence-based risk factors for the perioperative period. However, reliability and validity testing have not been completed on either of these tools. Good hand-off communication and visual indicators can alert OR staff as to which patients are most at risk because of preoperative instability and diagnostic procedures, as well as specialty knowledge of required positioning and equipment used during the case. If a patient develops a DTI postoperatively, root cause analysis needs to include OR or procedural environment staff, so all involved personnel can examine and possibly improve their practice.19 Although the differences between the outcome groups in the current study were not statistically significant, more postoperative DTIs evolved to full-thickness/unstageable PIs postoperatively (7 DTIs, 35%), and there was a higher incidence of procedures before DTI appearance.

Treatments and NFLU therapy. No discernible pattern of dressings used on DTIs, either before or after the DTI, was identified, although silicone-border dressings were used most frequently (37, 95%); these dressings were used before DTI occurrence in 20 DTIs. Use of specialty beds increased after DTIs appeared, with the greatest increased use in the group with resolved outcomes (P = .002). The only specialty bed used that showed significance between DTI outcome groups was a powered, multizoned, low-air-loss mattress system (10/9/3; P  = .005) compared with (partial) high-air-loss mattress (1/8/3; P  = 0.45).

DTI has been shown in a review of the literature by Berlowitz and Brienza²⁰ to result from tissue distortion related to vertical and shear forces; in addition, the review by Stekelenburg et al²¹ and the rat model study by Cui et al²² demonstrated that compression of the blood supply, impaired lymphatic function of the tissues, and reperfusion injury also can be factors. Several investigators including Peirce et al (rat model)23 and Oomens et al^24,25 and Agam and Gefen²⁶ (biomedical engineering models) showed damage from oxygen-derived free radicals released during reperfusion can cause inflammatory and cytotoxic effects on the tissues, rendering muscle tissue more susceptible to this damage than the skin. NLFU provided at 40 kHz has been shown by multiple investigators^27-31 to improve healing by decreasing proinflammatory cytokines, increasing vascular endothelial growth factors and improving microcirculation. In retrospective studies by Honaker et al³¹ and Thomas,³² an average of 10 to 12 sessions were required to produce a 75% to 100% resolution rate of DTIs. During the time frame covered by this study, NLFU was used for patients who did not have measurable signs of healing with standard treatments including dressings, specialty beds, and turn/reposition programs. The mean number of sessions for the NLFU group in the current series was 9.9, but 33% of the patients had 5 or fewer sessions because the DTI resolved or the patient was discharged or died. To achieve the best outcomes with NLFU, the intervention should be started as soon as the DTI is identified. In the current study, the average delay between DTI identification and NLFU initiation was 2 days, but in 9 DTIs (16%), there were delays ≥5 days.

Unlike the results of the prospective randomized controlled trial by Prather et al²⁷ and a quasi-experimental study by McCormack and Hobbs³³ that showed improved healing of partial-thickness injuries, the outcomes for the current cohort of DTIs that received NLFU were not better than those who did not. Several confounding factors were more prevalent in the NLFU-treated group, such as longer LOS (47 days vs 38 days), vasopressor use following DTI identification (38% vs 23%), and mechanical ventilation (50% vs 25%).

Although no specific patient acuity data were collected, the NLFU-treated group of DTIs were in patients who had a higher acuity of illness, because death occurred in 28% of the NLFU group within 1 month of discharge and in 70% of those patients, death occurred on the day of discharge or 2 to 3 days later.

HOB elevation. Since the introduction of guidelines for the prevention of ventilator-associated pneumonia, clinicians have had to weigh competing patient care priorities with regard to HOB elevation. Guidelines for ventilator-associated pneumonia34,35 recommend a HOB elevation between 30˚ and 45˚, and NPUAP guidelines recommend a HOB elevation of <30˚.^36,37
A feasibility study by Schallom et al³⁵ found reduced oral secretion volumes and reflux at higher HOB elevations without pressure-related tissue injuries. However, the 11 patients who completed the Schallom study³⁵ all were on low-air-loss mattresses, and the study period was only 48 hours. In the current study, ventilator support was found to be a significant factor for PU development. This conflict in perceived priorities related to HOB elevation could be addressed with a more frequent turn-reposition program and use of specialty mattresses, such as those with low-air-loss, pressure-redistributing, or shear-reduction features.

LOS. In the current study, hospital LOS ranged from 4 to 258 days, with a median of 23 days, and ICU LOS ranged from 1–173 days with a median of 12 days. DTIs that were resolved at discharge had the longest ICU and hospital LOS, as well as the latest onset of DTI following admission. Most patients who had an extended LOS were awaiting transplant or dependent on supportive technology, such as extracorporeal membrane oxygenation or left ventricular assist devices. The authors’ analysis of DTI outcomes was limited to the status at hospital discharge, so although 131 DTIs (73%) were considered stable (not worsening) or improving (skin intact, discoloration fading), the ultimate DTI outcome was unknown. A total of 12 DTIs that resolved before discharge were device-related; the extended LOS due to more severe illness most likely allowed time for complete resolution.

Devices. In a presentation to the 2012 NPUAP Bi-Annual Conference, Baharestani³⁸ noted that PUs caused by medical devices have increased in recent years. Medical device-related PUs are defined by the NPUAP¹ as resulting “from the use of devices designed and applied for diagnostic and therapeutic purposes. The resultant pressure ulcer generally conforms to the pattern or shape of the device.” A secondary analysis of medical center data³⁹ has shown medical devices create pressure against the skin, and the microclimate is altered as a result of humidity and heat between the device and the skin. This combination makes PUs more likely to occur after device use. To prevent these injuries, health care facilities have put policies in place to ensure medical devices are removed or repositioned in a timely manner; when a device is used, the NPUAP recommends that it be cushioned and the skin under a medical device inspected at least daily unless medically contraindicated and more often in patients with localized or generalized edema, and that staff be educated on use of medical devices.³⁹

Of the 179 DTIs in the current study, 41 (23%) were related to medical devices. The device most likely to cause a DTI was a long-stretch compression wrap (Deluxe LF elastic bandage; Hartmann Inc, Rock Hill, SC) applied to the lower extremities. Wraps accounted for 15 (38%) of the DTIs caused by medical devices. As a result of this finding, the authors changed practice to use of short-stretch wraps (Rosidal K; Lohmann & Rauscher, Milwaukee, WI) for ambulatory patients and long-stretch wraps for nonambulatory patients. Cotton or foam padding was applied before wrap use to protect bony prominences and tendons from focused pressure. Casts and splints accounted for 8 (18%) of the device-related DTIs.

Other medical devices that caused DTIs in this study included cervical spine collars (3, 7%), cast/splint (5, 12%), respiratory devices (5, 12%), braces (5, 13%), boots (2, 5%), sequential compression devices (1, 2%), slings (1, 2%), and oscillating beds (3, 7%). In their prospective study, Coyer et al⁴⁰ found ICU patients who developed a device-related DTI had an average of 8.6 medical devices placed. Endotracheal tubes and nasogastric tubes caused the highest number of DTIs in their study.

In the current study, 27 (68%) of the DTIs caused by medical devices occurred in the ICU setting and of those 11(49%) were caused by long-stretch wraps. A unique situation arises when patients are unconscious and require a cast or splint. If a patient cannot communicate that the cast or splint is causing additional pain, a PU may be more likely to occur. In rare cases the authors have experienced, removing the splint or cast routinely can place the patient at risk for further complications, such as worsening an unstable fracture.
Incontinence. According to a systematic review⁴¹ and a prospective analysis,⁴² fecal and urinary incontinence usually do not cause DTIs, but they have been associated with PUs because of the weakening of the skin’s ability to tolerate shear/friction, moisture, caustic drainage, and bacteria.^37,38 Gefen⁴³ used mathematical and computational modelling to demonstrate the effects of wetness-related friction on shear loads, with resulting reduced strength of wet skin. In a systematic review and meta-analysis of incontinence-associated dermatitis and moisture as risk factors for PU development, Beeckman et al⁴¹ found significant associations between both urinary incontinence alone and double incontinence and PU development. In the current sample cohort, a low incidence of urinary incontinence was noted because patients were either continent or had an indwelling urinary catheter. However, patients with DTIs had an higher incidence of fecal incontinence in all outcome categories, ranging from 60% to 84%, with the highest percentage in the full-thickness/unstageable group.

Braden subscale scores. No statistical difference was found between the 3 condition-at-discharge groups in total Braden score 1, 2, and 3 days before DTI appearance. Each group’s total scores placed them in the high-risk category (12–14), with slightly lower scores in the group that developed full-thickness/unstageable DTI. The Braden nutrition subscale score was the only subscale that showed a significant difference between groups before DTI development, with the lowest scores found in the group that developed full-thickness PUs.

Although the prospective, quasi-experimental repeated measures study by Serpa and Santos⁴⁴ did not find the Braden nutrition subscale score to be predictive for PU development in hospitalized patients, the current authors were not surprised to observe that DTIs progressed to full-thickness injuries and were associated with the lowest scores. A minimal downward trend was noted in other subscale scores from 72 to 24 hours before DTI, but the trend seemed to be consistent with an increasing acuity of illness and intensity of therapy and care, which may have been associated with decreased mobility and increased tissue distortion.

Limitations
This study had several limitations. All data were abstracted retrospectively from the EHR, so granular details of DTI assessment and nursing documentation were limited. APRN CNS notes and photos were reviewed before DTI inclusion, but no interrater reliability was established. A natural history study without interventions was not conducted in that the 2-year period of data was derived from patients in a clinical practice where clinicians and providers were using various means to reduce the risk of PIs, which may have influenced the results.

Conclusions
This retrospective analysis of hospital data demonstrated that several factors are associated with development of DTI — namely, HOB elevation, use of therapeutic medical devices, and cerebrovascular diseases. What these factors may have in common is the potential for tissue distortion, which could contribute to additional unrecognized risk of tissue injury. When patients also experience hemodynamic instability requiring vasopressors or anemia, potential injury to tissue may be exacerbated.  

This study also demonstrated that DTIs do not inevitably progress to full-thickness reportable events, even in high-risk patients. A high index of suspicion when caring for patients with these factors should warrant more frequent monitoring of skin (especially under medical devices) and promote the earlier use of therapeutic support surfaces that redistribute pressure and reduce shear. Future DTI studies could involve concurrent or retrospective analysis of risk factors and associations of those factors in a descriptive or case control design. Inclusion of the role of microvascular disease and tissue distortion also should be considered for future research.

Acknowledgment
The authors thank and acknowledge Christine M. Lohse, MS, Department of Biomedical Statistics and Informatics, Mayo Clinic, Rochester, Minnesota, for performing the statistical analyses of the study data.

Abstract
The optimal timing of loop ileostomy reversal remains largely unknown, but evidence that delayed ileostomy closure may increase postoperative complication rates is increasing. Purpose: Retrospective research was conducted to compare outcomes between patients who had early (<6 months) or late (>6 months) loop ileostomy closure. Methods: Records of patients >18 years of age who underwent circumstomal reversal of a loop ileostomy over a period of 5 years in 1 hospital’s colorectal unit were abstracted and analyzed. Data from patients who had a planned or conversion to laparotomy, a concurrent bowel resection, reversal of double-barrel small bowel and colonic stomas, or closure of an end ileostomy or patients whose records were incomplete were excluded. Demographic information, American Society of Anesthesiologists (ASA) grade, primary operation indication, surgery and inpatient dates, readmission within 30 days of discharge, reasons for readmission, complication type, and Clavien-Dindo classification were extracted and compared between early and late closure groups using independent-sample t test and Fisher’s exact test. Results: Among the 75 study participants, 25 had an early closure (mean age 68.6 [range 26 – 93] years, mean time since primary surgery 3.8 months) and 50 had a late closure procedure (mean age 71.6 [range 46 – 93] years, mean time since primary surgery 12.8 months). Gender distribution, ASA grades, primary surgery indication, and total number of readmissions were similar between the 2 groups. Hospital length of stay was significantly shorter (5.5 days vs 9.4 days; P = .01) and average number of complications was significantly lower (0.33 vs 0.61; P = .04) in the early closure group. Rates of postoperative ileus, anastomotic bleed, and wound-related complications were not significantly different. Conclusion: Hospital length of stay and average number of postoperative complications following circumstomal loop ileostomy closure were significantly lower in the early than in the late closure group. Additional studies are warranted to help guide practice.

The loop ileostomy is a type of stoma created to divert the flow of intestinal content away from a distal colorectal anastomosis. It defunctions an anastomosis and limits the clinical impact of an anastomotic leak, which is one of the most feared complications in colorectal surgery.¹ According to retrospective and prospective studies,^1,2 a loop ileostomy is favored over loop colostomy for ease of construction and lower complication profile. Additionally, the efficacy of a loop ileostomy in reducing the consequence of anastomotic dehiscence and improving outcomes in distal colorectal anastomoses is well documented in prospective cohort studies.^3,4 However, a systematic review⁵ has shown nearly 20% of patients develop a variety of ileostomy-related complications that can negatively affect quality of life and lead to hospital readmissions. For these reasons, loop ileostomies tend to be temporary and many surgeons aim to reverse them within 3 months, although the optimal time for reversal of ileostomy remains unknown.

The reversal operation generally is regarded as safe, with reported mortality rates as low as 0.4%.⁵ This surgery is commonly performed through a circumstomal incision followed by full mobilization of the ileostomy, formation of ileo-ileal anastomosis, and closure of the fascia and wound. In some instances, a full laparotomy is required. However, the literature suggests that up to one third of loop ileostomies may never be reversed.⁵ A number of factors have been shown in a retrospective study⁶ of 964 patients with ileostomies to be implicated in the delay or failure of stoma closure; these factors included older age, comorbidities, delayed recovery after the initial operation, complications such as anastomotic dehiscence, and the need for adjuvant chemotherapy. It is hypothesized that the time delay may have an impact on postoperative outcomes when the loop ileostomy is eventually closed.

 The aim of this study was to compare outcomes between patients treated in a single colorectal unit who underwent early (<6 months) or late (>6 months) circumstomal closure of a loop ileostomy in order to determine whether timing of reversal influences postoperative outcomes.

Methods
A retrospective analysis of data from a single colorectal institution was undertaken. Electronic general surgical records were used to obtain a list of patients that had undergone closure or ileostomy reversal between June 2012 and June 2017. Approval for the collection of patient data was obtained from the Board of King’s College Hospital NHS Foundation Trust.
Patient criteria. Patients were considered for inclusion if they were >18 years of age and underwent circumstomal approach of reversal of loop ileostomy. Patients requiring a planned or conversion to laparotomy for reversal of ileostomy, reversal of ileostomy with concurrent bowel resection, reversal of double-barrel small bowel and colonic stomas, and closure of end ileostomies, as well as patients for whom information needed for the analysis was unavailable or incomplete, were excluded.
Patient data. Patient data abstracted included demographics, American Society of Anesthesiologists (ASA) grade, primary operation where the ileostomy was formed, indication of primary operation, date of reversal of loop ileostomy, interval between primary operation and date of reversal of loop ileostomy, length of inpatient stay, readmission within 30 days of discharge, reasons for readmission within 30 days of discharge, and complication types and numbers (see Table 1).

Data collection. Data were collected from patient electronic and paper files by the lead author. Data were extracted using hospital numbers only and stored in a secured password-protected file.
Data analysis. The following variables were calculated in Excel: time interval in months from primary operation to reversal of ileostomy, length of stay, and number of complications. The categorization of complications are shown in Table 1. The Clavien-Dindo classification was used for comparison of the postoperative complications. This validated grading system for postoperative complications was first described by Dindo et al⁷ in 2004 and is now widely used for grading postoperative complications in an objective and reproducible manner. All postoperative complications were manually graded using the Clavien-Dindo grading tool as shown in Table 2.⁷ Analyses were carried out on the total number of patients as well as 2 subgroups of patients, categorized in terms of early (<6 months) and late (>6 months) reversal of loop ileostomy, using IBM SPSS Statistics, version 23 (IBM Corp, Armonk, NY). Variables were analyzed using independent sample t tests for the difference in means between the 2 groups. Categorical data were reported as the number of patients/percentage of patients and compared using Fisher’s exact and chi-squared tests. A P value of <.05 was considered statistically significant.

To ascertain the results of this study in comparison to published data, a manual electronic search of PubMed (Medline) and Embase for literature published between January 1, 1987 and December 31, 2017, was performed using the search terms ileostomy, reversal, outcomes, and complications to find research that compared outcomes in patients undergoing ileostomy closure at different time intervals. All citations identified were reviewed independently by the lead author to assess suitability for comparison to study findings; 6 original studies were identified (see Table 3).

Results
Of the 101 patients that underwent reversal of loop ileostomy during the study years, 75 (25 with early and 50 with late reversal) met the inclusion criteria and their data were included in the study. Mean age was 68.6 (range 26–93) years for the early group and 71.6 (range 46–93) years for the late group. Gender distribution, ASA grades, and indication for primary operation were similar between the 2 groups (see Table 1).

The majority (76%) of loop ileostomies were constructed due to cancer; the remainder were created as a result of benign diseases such as diverticular disease (19%), inflammatory bowel disease (4%), and rectal trauma (1%). Within 30 days of discharge after reversal of ileostomy, 14% of the total number of patients had 1 or more hospital readmissions, 4 in the early group and 7 in the late group (see Table 1). Mean length of hospital stay following ileostomy reversal was 4.5 (range 0–16) days in the early group versus 9.4 (range 0–39) days in the late group (P = .01) (see Table 1 and Figure). The total number of postoperative complications was higher in the late group (35) than the early group (9), although consideration must be given to the fact that the number of patients in the late group is double that in the early group. Some patients developed no postoperative complications (42, 56%); 7 patients developed more than 1 postoperative complication (9.3%). The mean number of complications was 0.33 in the early and 0.66 in the late group (P = .04). Ileus was the most common complication in the early (4/9, 44%) and late (11/35, 31.4%) closure groups, but the difference was not statistically significant (see Table 1).

In terms of Clavien-Dindo classification, the late group experienced more grade I, II, III, and V complications, although these numbers were too small for statistical calculations (see Table 2).

Discussion
Systematic reviews and randomized controlled trials^8-10 have shown defunctioning loop ileostomies effectively reduce the rate of symptomatic anastomotic leaks and the need for reoperation in such cases. Restoration of gastrointestinal continuity by reversing loop ileostomies requires a second operation, the timing of which remains controversial. Furthermore, the morbidity rate following closure of ileostomy may be close to 20%,⁵ and systematic reviews and retrospective studies^5,11,12 have shown mechanical small bowel obstruction, ileus, and wound-related problems are common complications.

Because the optimal timing of ileostomy closure remains unknown, interest has been growing in the assessment of the association between timing of closure and postoperative outcomes. Prospective and retrospective reports^13-15 are available in the literature comparing ileostomy closures as early as 8 days following the primary operation to the traditionally timed closures at 12 weeks. Two (2) retrospective studies^15-16 and 1 prospective randomized trial¹⁴ compared closures at various time intervals: 1 month to after 6 months, 3 months to after 6 months, and within 3 months to after 3 months, with variable conclusions (see Table 3).

The current study compares the outcomes of ileostomy closures within 6 months to those after 6 months, the first published comparison of outcomes at these specific time intervals. A key finding from the current data is that patients who underwent early closure had a significantly shorter length of hospital stay than patients who had late closure (4.5 days vs 9.4 days, respectively; P = .01). Similar findings were reported in a prospective randomized trial by Alves et al¹⁷ that compared outcomes in 186 patients randomized into 2 groups that underwent either early (within 8 days) or late closure (at 60 days) of a loop ileostomy. The reported length of stay in the early group was significantly lower than the late group (16 vs 18 days, respectively; P = .013). Similarly, a retrospective study of 93 patients by Omundsen et al¹⁸ that compared ileostomy closures within 10 days and at 90 days found length of hospital stay in the early group was significantly lower than in the late group (14 vs 17 days, respectively; P = .05).

The current study also found the mean number of complications was significantly lower in the early group as compared to the late group (0.33 vs 0.61, respectively; P = .04), findings supported by a randomized controlled trial of 112 patients by Danielsen et al¹⁹ in which significantly fewer complications occurred in the early group as compared to the late group (1.2 vs 2.9, respectively; P<.0001). A retrospective study by Rubio-Perez et al²⁰ concluded the delay in ileostomy closure was associated with a significant increase in postoperative complications, specifically wound infections (13%; P = .007) and pseudomembranous colitis (4%; P = .003).

In the current study, the number of cases of ileus were higher in the late group (11) than the early group (4), but the difference was not statistically significant.

Williams et al¹³ analyzed the changes that occur in the distal, inactive limb of a loop ileostomy and found a significant reduction in the ileal smooth muscle contractility following a period of diversion of intestinal content. Based on this finding, the authors concluded such changes may result in reduced compliance and contractility of the defunctioned bowel segment, leading to the obstructive symptoms that may be observed following ileostomy closure. This may help explain the higher rate of ileus observed in the current late group, an observation also supported by the study by Alves et al.¹⁷

In order to objectively assess postoperative complications, study data were categorized according to the Clavien-Dindo classification (see Table 2). All postoperative complications were manually graded using the Clavien-Dindo grading tool as shown in Table 2. When comparing the 2 groups, the current authors found the late group had more grade I (14/35), II (16/35), III (3/35), and V (2/35) complications compared to the early group. However, these numbers were too small to determine statistical significance.

Some studies did not demonstrate significant differences in studied outcomes between early and late ileostomy closures. In their prospective controlled trial of ileostomy closures within 1 month and after 6 months, Zhen at al¹⁴ found rates of closure-related complications (20.9% vs 18.6%, respectively; P = .637) and length of hospital stay (7.94 vs 7.97 days, respectively; P = .588) were not statistically different between the early and late groups.

A prospective study by Zhou et al¹⁵ of early defunctioning ileostomy closure among 123 patients found no significant difference between the group that had ileostomy closure within 90 days and the group that underwent ileostomy closure after 180 days in relation to stoma-related complications (early group 10%, late group 16%; P = .387), anastomotic leakage (early group 1%, late group 2%; P = .691), ileus (early group 4%, late group 9%; P = .245), and peristomal hernia (early group 5%, late group 4%; P = .190).

A large retrospective study by Li and Ozuner¹⁶ on ileostomy reversal among 358 patients also reported no differences between the early (<3 months) and late (>3 months) ileostomy closure groups in terms of rates of ileus (12.3% vs 13.4%, respectively; P>.87), small bowel obstruction (15.6% vs 15.1%, respectively; P = .5), wound infection (2.2% vs 1.7%, respectively; P = .99), and surgery-related readmission rate (0.56% vs 0%, respectively; P = .85) (see Table 3).

The current study demonstrated that early ileostomy closure may lead to a significant reduction in length of hospital stay and average number of postoperative complications. These findings are supported by some of the aforementioned studies^17,19-20 and could encourage surgeons to expedite reversal of loop ileostomies whenever possible. This study is the first to compare outcomes of ileostomy closure within these particular time intervals; larger prospective studies are required to confirm these findings and help guide practice.

Limitations
This study is limited by its retrospective design and relatively small numbers. The difference in the group sizes also was a consideration during data analysis. A future follow-up study will have a larger number of participants more equally divided between the comparison groups and fewer confounders such as comorbidity status. The authors suggest a prospective, randomized study design that also will explore the effect of timing of ileostomy closure on parameters such as health-related costs and quality of life.

Conclusion
Although the optimal timing of closure of loop ileostomy is unknown, there is evidence to support early rather than late closure. The current study demonstrates the length of hospital stay and average number of postoperative complications were significantly lower in the early (<6 months) than in the late (>6 months) closure group. The current study supports results of previous studies and warrants prospective research to validate these findings.

Acknowledgments
The authors thank the general surgery theatre manager and colorectal clinical nurse specialist for their help with the provision of the data and Samantha Matin for her assistance with data analysis.
 

Abstract

Incontinence-associated dermatitis (IAD) is a common, painful, difficult-to-treat skin condition. Purpose: A 2-part, quasi-experimental, post-test study was conducted to evaluate the impact of prevention initiatives on IAD prevalence and incontinence practices. Methods: In part 1, from May 2017 to November 2017, a quasi-experimental post-test study design was conducted in a health district in Australia. Following an audit of IAD prevalence and identification of evidence practice gaps in 4 hospitals in a local health district (12 wards, 250 patients), an implementation science approach was used to implement evidence-based initiatives. An IAD committee was formed, staff were educated about correct incontinence pad sizing, washable and disposable underpads and plastic sheets were removed from the care setting, and barrier cream cloths for cleansing, moisturizing, and protecting skin were introduced. Patients admitted to 1 of the 12 wards who were ≥18 years of age were recruited for participation and evaluation in the post-intervention implementation IAD and incontinence care practices audit. Post-intervention data were entered into a software program and compared to pre-implementation data using descriptive and bivariate statistics. In part 2, nurses from the 12 wards were asked to participate in 1 of 6 focus groups to share their impressions about the barrier cream cloths. Discussions were transcribed verbatim and analyzed using descriptive content analysis. Results: The rate of incontinence among audited patients (N= 259, 132 men, 124 women; mean age 73.2 ± 16.8 years) was 47.2% (119/252) and 2/259 (0.8%) had a pressure injury (PI). IAD prevalence was significantly lower in the post- than in the pre-implementation audit (6/259 vs 23/250; P = .015), as was hospital-acquired pressure injury (9/250 [3.6%] vs 2/259 [0.08%]) and the use of bed protection layers (154/238 vs 6/259; P<.01). The focus groups included 31 nurses (25 women, 6 men). Four (4) themes emerged: 1) benefits to the patient (eg, improved skin condition), 2) usability (eg, fewer steps), 3) problems encountered (eg, not seeing the barrier in place), and 4) related factors. Patient comfort was cited frequently as an important benefit. Conclusion: Evidence-based initiatives led to a significant reduction in IAD prevalence and improved incontinence care practices.

Incontinence-associated dermatitis (IAD) is a painful skin condition that, according to a multisite epidemiologic analysis, an international pressure injury prevalence survey, the Global IAD Expert Panel, and a multicenter prevalence study,^1-4 is underreported, neglected, and difficult to treat in hospitalized patients. The condition is an irritant dermatitis characterized by erythema of the skin around the buttocks, perineum, gluteal clefts, and other areas where there is friction and moisture between the skin and clothing and bed linen.^5,6 IAD is caused by fecal or urinary incontinence or both and is included in a broader group of skin conditions referred to as moisture-associated skin damage (MASD).^5,7

IAD is a recognized risk factor for the development of pressure injuries (PIs).^1,2,8 A meta-analysis and systematic review of 58 studies by Beeckman et al⁸ established an association between IAD and the development of PIs. Gray and Giuliano¹ conducted a descriptive and correlational analysis of data on IAD and PIs in the sacral area involving 5342 patients in acute care facilities in 36 states in the United States. They found IAD significantly increased the likelihood of a patient developing a sacral PI. Furthermore, IAD is often incorrectly misdiagnosed as a PI.^9,10 Barakat-Johnson et al¹⁰ conducted a prospective, descriptive study in a large tertiary hospital in Australia of patients reported to have a PI. The authors found 176 out of 363 patients (48.5%) had various skin conditions that were not PIs. Moisture-associated skin conditions, namely IAD, accounted for 69 out of 176 (39.7%) of the skin conditions mistaken for PIs. Furthermore, IAD assessment, management, and prevention is not seen and discussed as a priority among clinicians, as found by the Global IAD Expert Panel.³ Prevention is the primary goal in the management of IAD and should comprise a number of evidence-based strategies aimed at maintaining skin integrity.³ In September 2014, the Global IAD Expert Panel met to review knowledge gaps in IAD and to advance best practice principles to address these gaps. They identified that preventing IAD can, in turn, prevent PIs and should be considered an essential component of PI prevention.³

The purpose of this study was to evaluate the impact of certain initiatives by examining the prevalence of IAD as well as incontinence practices. These initiatives comprised the formation of an IAD committee, review of existing IAD literature, education for nurses on the prevention and management of IAD and correct pad size, and the introduction of new skin cleansing and management products. These results were compared with the findings of the authors’ previous study.¹¹ Nurses’ overall perceptions about the use of a 3-in-1 barrier cream cloth also were assessed.

Methods

This study is 1 of 2 IAD initiatives¹¹ arising from a program of implementation science work to implement evidence-based IAD prevention and treatment in order to examine the effects of the program and address PIs in an urban Australian local health district (LHD). In 2015-2016, the authors conducted a cross-sectional, mixed-methods study across 12 wards in 4 hospitals in the LHD¹¹ to examine the prevalence of PIs and IAD as well as to determine evidence-into-practice gaps. Two (2) major practice gaps were identified: 1) IAD was misclassified as PI, and 2) existing incontinence nursing practice was inconsistent with contemporary evidence-based literature and guidelines and included use of antiseptic wash to cleanse skin after a soiled episode, slathering on zinc or ointment-based skin protectants that may impede pad absorbency, and incorrect sizing of containment devices. The consequences of these practice gaps are significant, with the potential to increase patients’ risk of developing a PI as well as IAD and increase patient discomfort and length of stay. Based on the results of previous research,^10,11 other cross-sectional and systematic review studies conducted in Australia and internationally,^8,12 and international guidelines,3 this LHD and the New South Wales State Pressure Injury Working Party¹³ identified IAD as a clinical priority in acute care hospitals. These findings led to collaboration between this LHD, the New South Wales State Pressure Injury Working Party, and international leads¹² for conducting IAD prevalence audits and statewide PI point-prevalence audits.

To address the acknowledged practice gaps, an implementation science approach, guided by the Promoting Action on Research Implementation in Health (PARiHS) Framework,^14,15 was employed as part of the broader project addressing PIs. Implementation science is the study of how to design and evaluate methods that enable interventions to be effectively put into practice, taking into account the context and internal mechanisms.^14-17 The current study’s intervention involved evidence-based initiatives to address IAD and existing incontinence practices. Specifically, the initiatives included the formation of an IAD committee, review of existing IAD literature, education for nurses on the prevention and management of IAD and correct pad size, and introduction of new skin cleansing and management products.

Study phases. The pre-implementation data collection phase was undertaken between November 2015 and January 2016, the implementation phase from June 2016 to May 2017, and the post-implementation phase between June 2017 and October 2017. The terms and operational definitions used in this study are presented in Table 1.^3,5,14,18-22

Design. This study used a quasi-experimental, post-test study design to evaluate IAD prevalence and the nature of incontinence practices following the implementation of evidence-based strategies across a district of 4 hospitals in Australia. A single point collection of IAD prevalence was conducted at different time points across the 4 facilities. Patients were audited once; to accommodate workload, each facility had a different audit day. Bivariate statistics were used to compare the resulting data with pre-implementation data, which also were collected once.

Ethical considerations. Ethical approval was granted by the Sydney LHD Ethics Review Committee of the Royal Prince Alfred Hospital zone (ref: HREC/15/RPAH/482). Participant confidentiality was maintained by the deidentification of all participant identifying information at the point of data collection. To obtain patient consent, a nursing staff member approached the patient and explained the purpose of the study and obtained verbal consent. Once consent was obtained, auditors, who were senior nursing members of staff, provided a detailed explanation of the study. If a patient was unable to provide consent, a relative consented on his/her behalf. For the nurse focus groups (FGs), written consent was obtained from each nurse.

Setting. The study was undertaken in 12 wards in a LHD comprising 3 acute tertiary hospitals, 1 subacute rehabilitation hospital, and 5 community health centers. The LHD is in a large urban setting that provides primary, secondary, and tertiary care to a local population of 600 000 people. In order to make direct comparisons with the previous study, the same 12 wards as the previous study¹¹ were audited. These included acute and subacute aged care, rehabilitation, neurology, cardiovascular, intensive care, and palliative care. Each ward consisted of 13 to 30 beds, with a mean occupancy of 72.3%. Each ward was staffed with between 7 and 80 nurses.

Sample/Participants.

Audit patient sample. Patients were recruited on the day of the audit if they were admitted to 1 of the 12 wards and ≥18 years of age. Patients were excluded if they were unable to consent or if they were not present on the ward (eg, undergoing a procedure). Agreement from the nursing unit manager (NUM) and nurses on the relevant wards was obtained in order to approach the patient before data collection.

Focus group sample. All nurses (bedside nurses, nurse leaders, nurse educators, and student nurses) working on any of the 12 wards at the time of the study were invited to participate in the focus groups. Student nurses could attend only if their registered nurse preceptor was attending the focus group. The focus groups were held in the post-implementation phase, 3 to 5 months after the completion of the implementation phase and 12 months after the initial implementation that began June 2016.

Intervention: evidence-based initiatives. The intervention initiatives included: 1) formation of a committee consisting of senior nurses, skin integrity and infection control nurses, continence nurse educators, and nurse consultants; 2) changes to clinical practice, such as correct sizing of containment products, the proper use of bed protection products, and knowing the difference between IAD and PI; and 3) an education campaign on IAD prevention and incontinence nursing practices (see Table 2).^14-17,23,24 The intervention strategy to address cleaning, moisturizing, and protecting patients’ skin was a major practice change. A barrier cream cloth (see Table 1)^3,5,14,18-22 was used to cleanse, moisturize, and protect patient skin. Practice changes were documented using the same audit tool as in the authors’ previous study (see Figure 1).¹¹ This approach was universally implemented.

Outcome measures. Primary outcome measures were: 1) the prevalence of IAD and 2) changes in clinical practice (eg, use of barrier cream cloth, removal of unnecessary bed protection pads).

Data collection.

Patient audit. The prevalence audit included a head-to-toe skin assessment using the framework outlined by the National Pressure Ulcer Advisory Panel/European Pressure Ulcer Advisory Panel/Pan Pacific Pressure Injury Alliance (NPUAP/EPUAP/PPPIA) guidelines,¹⁸ as well as observations of the incontinence practices. Researchers used the same audit tool as in their previous study (see Figure),¹¹ which was developed by the authors and conforms with IAD guidelines.³ The audit tool includes demographic data, continence and mobility status, IAD and PIs, incontinence products used, and nursing practice observations, which were documented by wound nurse leaders of each facility. IADs were classified as Category 1 or 2 (see Table 1),^3,5,14,18-22 and PIs were staged according to the NPUAP/EPUAP/PPPIA Pressure Ulcer Classification System.18 Data then were entered into Microsoft Excel. FGs. A discussion guide was developed by 2 senior wound clinicians and 2 researchers with expertise in qualitative research. The guide was pilot-tested on 5 nurses before use; no changes were deemed necessary. Questions included, What are your thoughts on the new 3-in-1 barrier cream cloth? and How has this assisted you in the care you provide your patients? Demographic information including gender, age, years of experience, and the role of each participant was collected. Interviews were digitally recorded and transcribed verbatim by an independent transcription company. A research officer conducted the focus groups which took 35 minutes on average. The NUM allocated time for staff to attend the focus groups.

Procedure. Steps involved in the development, roll out, and evaluation of the initiatives are outlined in Table 2.^14-17,23,24

Audit procedure. A senior nursing clinician at each site and the lead author were responsible for coordinating the post-implementation prevalence audit. NUMs and nursing staff were notified of the upcoming audit by email and at handover (change of shift) by senior clinicians. To ensure consistency, the same auditors from the authors’ previous study¹¹ collected the data. The auditors, who were senior nurses, were provided with a photographic chart of IAD grading. All auditors had the requisite knowledge to diagnose IAD. Interrater reliability was assessed to ensure consistency between auditors by undertaking a PI classification test for this study with the addition of IAD photos. This involved auditors correctly classifying PIs in line with the NPUAP/EPUAP/PPPIA Pressure Ulcer Classification System¹⁸ and IADs as per the Global IAD Expert Panel.³ The pass rate set was at 85% as guided by a method outlined by Prentice et al²⁵ for conducting PI prevalence audits in Australia. The passing grade was achieved.

The auditors performed a head-to-toe skin inspection as per Prentice et al²⁵; patients were asked whether they were experiencing problems with incontinence. Patients in this study who had a urinary indwelling catheter were deemed continent.^19-21 If they were wearing an incontinence pad, it was assessed for correct sizing and application based on heavy, medium, or light incontinence/soiling. As per the authors’ previous study,¹¹ the correct incontinence pad was determined based on the type of incontinence, flow, the patient’s size (weight and height matched to the manufacturer pad size guide), mobility status, and the amount of urine or feces reported by the patient or nurse. Information on the nature of incontinence management was collected.³ This included skin care, correctly sized incontinence products (as per Cottenden et al²²), and the number of bed protection layers on each bed, such as disposable underpads (Blueys; Cello Paper Pty Ltd, Sydney, Australia) and washable underpads (Kylies; Ontex Healthcare, Sydney, Australia), plastic sheets, and draw sheets). Participating wards subsequently received a report with the main findings for their ward.

FG procedure. Flyers were distributed to the 12 wards, inviting nurses to participate in the FGs. Before distribution, permission was obtained from NUMs. FGs were held following afternoon handover on the ward. Each FG ran for 20 to 40 minutes and offered nurses an opportunity to share their views on their use of the newly introduced 3-in-1 barriercream cloth to reduce IADs.

Data analysis.

Prevalence audit. Data were collected via the paper-and-pencil audit tool and entered into and analyzed using IBM SPSS Statistics, version 22 (IBM Corp, Armonk, NY).²⁶ Entered data were cross-checked for accuracy and incomplete data were coded as missing. Prevalence rates of incontinence and IAD were analyzed descriptively as counts and percentages. Bivariate analysis was performed and statistical significance was set at P<.05. Results were compared with the pre-implementation audit (N = 250). Comparisons of the age of patients from the pre- and post-implementation patient cohorts were made using a t test. Chi-squared tests were used to compare gender, mobility status, IAD and hospital-acquired PI occurrence, and incontinent product use in both audits. Fisher’s exact test was used when cell counts were <5. No adjustment was made for multiple testing because all outcomes were prespecified.

FGs. The focus group transcripts were uploaded into NVivo 10 software (QSR International, Melbourne, Australia) for content analysis.²⁷ Transcripts were reviewed by 2 authors (MBJ, ML) individually and then together. Authors read, reread, and then coded the transcripts in NVivo and then met to compare coding and reach consensus. Distinct categories were identified from the analysis and assigned to a relevant theme with the assistance of co-authors.

Results

Patients. In total, 259 patients (mean age 73.2 ± 17.0; 132 [51.6%] men, 124 women [48.4%]; gender missing for 3 patients) participated in the post-implementation audit. Additional characteristics of the patient sample can be found in Table 3. Characteristics of the post-implementation audit (N = 259) and the pre-implementation audit (N = 250)¹¹ are shown in Table 4. No significant difference was found between the mean ages of patients (P = .86) or the proportion of men and women (P = .81). Mobility status also did not differ between the pre- and post-implementation audits (P = .40). The patients within each audit were comparable in terms of basic demographics.

Comparison of incontinence. Of the 250 patients in the pre-implementation audit and the 252 patients in the post-implementation audit, 111 (44.4%) and 119 (47.2%), respectively, were incontinent; data on incontinence were missing for 7 patients in the post-implementation audit (P = .78). No statistically significant difference was noted in the proportion of patients with urinary incontinence between the pre- (58/111, 52.3%) and post-implementation audit (54/119, 45.4%) (P = .30) or in the proportion of patients with both fecal and urine incontinence in the pre-implementation (50/111, 45.0%) compared to post-implementation audit (46/119, 38.7%) (P = .33). However, the number of patients with fecal incontinence was lower in the pre-implementation audit (3/111, 2.7%) than in the post-implementation audit (19/119, 16%) (P = .001) and 139/250 (55.6%) patients were continent in the pre-implementation versus 133/252 patients (52.8%) in the post-implementation audit (P = .53). Thus, with the exception of fecal incontinence, the patients in each of the audits were comparable on continence measures. Overall, the patients in the pre- and post-implementation audit groups were similar enough to allow for an evaluation of the impact of the intervention.

Comparison of IAD and PI. The prevalence of IAD in the pre-implementation (23/250, 9.2%) and post-implementation audit (6/259, 2.3%) was significantly different (P = .015) (see Table 5). The prevalence of hospital-acquired PIs in the pre-implementation (9/250, 3.6%) and post-implementation audits (2/259, 0.8%) was significantly different (P = .034) (see Table 5).

Comparison of incontinent product use. Implementing evidence-based strategies resulted in a significant reduction in the use of bed protection between the pre-implementation (154/238, 64.7%, data missing for 12 patients) and 6/259 (2.3%) post-implementation audit (P<.01). Nurses did not use any plastics, washable underpads, or draw sheets as bed protection during the post implementation audit period (see Table 6), a significant improvement from the pre-implementation audit, which reported use of plastics in 67/242 patients (27.7%), washable underpads in 45/246 (18.3%), and draw sheets in 97/250 (38.8%). Only 6 disposable underpads (2.3% of patients) were found to be in use in this audit, which was an 87.5% decrease from the pre-implementation audit that noted use of 48 disposable underpads (19.2%) (P<.001) (see Table 5).

Another notable finding was the significant difference between the number of continent patients who were provided incontinent products: in the pre-implementation audit, 73/139 (52.5%) of continent patients had an incontinent product (either pad or bed protection), whereas 28/133 (21.1%) of continent patients had an incontinent product at the time of the post audit (P<.01) (see Table 5).

Nurse FGs. The demographic data for the 6 focus groups (31 nurse participants) is presented in Table 7. Nurses’ experiences with the barrier cream cloths and other issues with IAD were grouped into 4 categories: 1) benefits to the patient, 2) usability, 3) problems encountered, and 4) related factors.

Benefits to the patient. Participants provided positive feedback on the improved skin conditions of incontinent patients, particularly those with IAD, as a result of the barrier cream cloths.

There was one particular patient who was fecally incontinent and at high risk of developing IAD due to poor skin integrity from frequent loose bowels who did not develop any IAD due to the use of the cloths as a preventative measure. (FG1)

Participants stated that many patients felt a noticeable difference and improvement to their skin, including less tenderness and pain when they were cleaned.

I can especially think of one patient off the top of my head who was really badly excoriated and when you’d touch her she would just cry. And even after 2 days with the wipes she was like, “Oh, so much better.” (FG2)

In some cases, patients verbally requested the cloths, citing how comfortable they felt.

Some patients will ask us, “Can we use the wipes on the front, everything?” I say, “Yes.” And then say, “Okay, just put it on.” So we just put it on, it’s like “Oh, so comfortable.” (FG2)

Usability. The cloths also were used as a form of prevention and protection for patients who were incontinent to ensure that areas of the body were moisturized and that the condition did not escalate, potentially saving the use of products later on.

But if a patient’s starting to have loose bowels and you use them, well then you sort of prevent the problem to begin with and you don’t end up having to use them as much. (FG3)

Nurses found many uses for the cloths, which were used not only as a preventive intervention but also as a form of treatment for skin complaints such as “red and angry skin,” “excoriation,” “skin breakdown,” and “dry skin.”(FGs 1, 3, 5)

Participants also found the cloths to be effective at absorbing odor, making cleansing a more pleasant experience for both patient and nurse. Furthermore, nurses commended the hygienic nature of using disposable cloths, particularly compared with creams.

You’re not touching your gloves onto a tube that then is going to be touched again…and it absorbs the smell. (FG2)

Many participants reported the cloths were superior to conventionally used products such as barrier creams, particularly in terms of comfort, effectiveness, time efficiency, and ease of use. The moisturizing qualities of the cloth negated any need to use large amounts of barrier creams.

When you’re rubbing zinc on the patients they hate it, because it’s like this cold, horrible cream. It’s uncomfortable, it’s sticky. (FG3)

We cover our patients in zinc a lot and use a lot of zinc. And with those [cloths] you don’t need to. (FG4)

Conventional barrier creams, such as zinc creams, were deemed to be difficult to clean off without being rough and compromising the skin integrity, as well as being difficult to see through. Some participants stopped using barrier creams in favor of the cloths. Furthermore, several participants cited issues with how barrier creams were being used inappropriately. For example, too much cream was being applied, causing build up and time-consuming cleaning or improper cleaning. However, some nurses mentioned they were lathered thickly due to the ineffectiveness of the creams.

…you have to lather it so thick to actually get a decent effect. And you can just use one of those wipes and you’ve done it. (FG3)

Problems encountered. Participants reported some negative features of the cloths, including patients feeling a cold sensation when they were applied to their skin and risk of cloth wastage when used on patients who were discharged earlier than expected.

You know, we have patients that might just come in 1 or 2 days and they’re gone again so it’s a lot of wastage as well because you can only use it for one patient, you know. (FG6)

FG 6 voiced concerns over the lack of visibility of barrier properties of the cloths and the inability to see where one has applied the cloth due to its transparent appearance on the skin, preferring creams instead:

But you can’t see with something that’s clear. You can’t see how thick it is… and if you change the pad again you can still see if there’s cream, or you need to add more.

However, FG participants believed that if cloths were more visible on application, they would prefer the cloth to layering creams.

Related factors. Nurses emphasized the importance of being educated in 2 areas: how to use the cloths appropriately to ensure they were used optimally and how to identify IAD.

So if I could have a bank of photographs that I could identify and run a PowerPoint presentation of those pictures, because we felt that from the evaluation that we understand what it is and how it comes about and how we can help prevent it, but we wanted to be more confident with identifying it on the ward by having a range of pictures and photographs and examples. (FG5)

Some participants experienced confusion as to how to identify and distinguish between skin conditions, such as incorrectly identifying PIs, emphasizing a need for better education. This was a difficult issue to address because nurses often worked different shifts on different days.

Discussion

The current study, guided by an implementation science framework,¹⁴ evaluated the impact of initiatives on the prevalence of IAD and incontinence practices across high-risk wards in a LHD. Compared to pre-intervention implementation audit results, IAD and hospital-acquired PI rates were significantly lower across the LHD and incontinence management practices had improved. Substantially fewer incontinence products were used in the post as compared to the pre-implementation audit. More specifically, bed protection products were used sparingly and the majority of patients with incontinence pads had the correct size pad based on the level of soiling. Continence management guidelines and a cluster randomized controlled trial^22,28 showed incorrect pad size results in poor absorbency and leakage of urine and feces, compromising patient skin integrity and predisposing to the development of IAD and PIs.^8,19 In their guidelines on the use of incontinence products, Cottenden et al²² recommended that pad size (ie, absorbency) should be selected in accordance with the volume of the flow to ensure urine and feces are contained and no leakage occurs.

One (1) major practice change was the use of a 3-in-1 barrier cream cloth for patients with incontinence to cleanse, moisturize, and protect patients’ skin after soiling. The nurses in this study provided positive feedback on the product’s effect on time efficiency and the improved skin conditions of patients with incontinence, particularly those with IAD. Many nurses preferred barrier cream cloths to the traditional steps of cleansing, moisturizing, and using a barrier cream because they could combine all the steps by using only 1 cloth.

Critical to the success of the prevention of IAD and any patient intervention is compliance with cleansing and moisturizing methods. To prevent IAD, clinicians should diligently adhere to the best practice principles of following a 3-step process of cleansing, moisturizing, and providing a barrier. However, it can be difficult to monitor and enforce such practices when they are implemented across several wards and changing teams of staff. Reducing the number of steps involved may help reinforce best practice adherence, something that is emphasized in the current guidelines.¹⁷ Other studies^23,29 also have found benefits in using a 3-in-1 barrier cream cloth (with a 3% dimethicone formula) as a solution for IAD. In a randomized controlled trial²³ of 141 nursing home residents in Belgium to compare a 3-in-1 cloth (Comfort Shield Barrier Cream Cloth; Sage Products Inc, Cary, IN) with standard care (water and neutral pH soap) to prevent and treat IAD, the 3-in-1 washcloth was found to provide a significant reduction in IAD prevalence (by 63.7%; P = .003). Similarly, a prospective, descriptive quality improvement study²⁹ of a 3-in-1 barrier cloth, performed in 2 acute care neurology units in a trauma center in the United States, found none of the patients with incontinence who used the 3-in-1 barrier cloth (25/46) developed sacral IAD or a PI. Following the current study and per international literature, the use of the 3-in-1 barrier cloths has become standard practice in the authors’ LHD to prevent and manage IAD.

In the current study, forming a committee to address the evidence-based practice gaps coupled with a leader to drive implementation planning and regular feedback from clinicians led to key stakeholder buy-in and clinician engagement. The committee also worked closely with senior nurses on the ward, as well as NUMs, to monitor and ensure implementation of initiatives. The PARiHS framework guided the process of planning and evaluating initiatives by taking barriers as well as the context into account.^14,15 Using this approach enabled clinical nurse engagement, support from senior leaders, and adjustment of the strategy relevant to ward-specific concerns. For example, 1 ward slowly implemented the practice change over several months and encouraged interested nurses to campaign for practice change; other wards decided to rely on their clinical nurse educator to educate all staff and implement initiatives in a shorter time frame. Similar to the current study, a recent quasi-experimental, clustered pre- and post-test design study by Sving et al²⁶ using PARiHS to reduce PIs evaluated the effectiveness of a multifaceted, tailored intervention aimed at improving adherence to PI strategies across 5 wards in a tertiary hospital in Sweden. Significant improvements in PI care were demonstrated, including a higher number of at-risk patients receiving the appropriate and necessary care to prevent PIs.

Current study findings informed next steps — specifically, an economic evaluation study on interventions in the LHD to address skin care, namely to prevent IAD and PIs.³⁰ Other initiatives include the development and implementation of additional strategies to assist with translating evidence into practice, specifically implementation of the evidence-based guidelines³ to prevent IADs across other wards in the LHD and translating the recently published Ghent Global IAD Categorisation Tool (GLOBIAD).³¹ The authors believe a designated local implementation committee with motivated senior leaders to drive practice change was particularly important because it encouraged key stakeholder buy-in and clinician engagement, leading to successful outcomes similar to other studies.^32,33 The key lesson in this study was the importance of using evidenced-based research to examine the problem and identify and evaluate each implementation step.

Limitations

Unlike incidence studies, cross-sectional studies capture a snapshot of a population at a single point in time. Therefore, some IAD cases may have been missed, thereby underestimating the problem. Another limitation is that focus group data were collected with nurse participants only. The perspectives of other members of the health care team could have provided extra valuable and unique information. Lastly, it is important to acknowledge that the impact of the reduction of IAD and improved incontinence practice is likely to be greater when there are dedicated champions driving an initiative and that care is delivered in a structured way, also known as the Hawthorne effect.^34-36 However, given that this study aimed to close the gap between evidence and practice, the strategies initiated were guided by an implementation science framework (PARiHS) to ensure uptake and sustainability to overcome the Hawthorne effect.

Furthermore, prevalence audits in this study were conducted 6 to 12 months after the intervention was implemented, suggesting that the practice changes had been sustained. A follow-up study in this LHD, 2 to 5 years post-intervention, would be valuable to determine whether adherence to the intervention has continued and if the IAD rate remains low.

Conclusion

Following implementation of evidenced-based initiatives to improve the nature of incontinence practices and prevent IAD on 12 wards in 1 LHD, prevalence rates of IAD and hospital-acquired PI were significantly lower and improvements were noted in prevention of IAD and incontinence practices. Nurse FGs revealed that 1 of the new initiatives, using a 3-in-1 barrier cream cloth, minimized the number of steps in incontinence management and improved skin condition. This suggests that using an evidenced-based approach leads to less work for nurses and better patient outcomes. This study provided valuable insight into addressing a clinical problem in a health district using an implementation science framework and merits further research in wider scale translation and scalability across other settings and other health districts.

Acknowledgments

The authors acknowledge, and are most thankful for, the commitment and enthusiasm of the district IAD implementation committee and all the nursing staff at Royal Prince Alfred Hospital, Concord Repatriation General Hospital, Canterbury Hospital, and Balmain Hospital. In particular, the authors recognize the efforts of clinical nurse leaders Melissa O’Grady, RN, MN, BN, DipSci, Cert.Cont,Manage, Advanced Clinical Practitioner in Continence; Thomas Leong, RN, MAppSci; Ashleigh Dolton, RN, BN, Grad.Cert.Crit.Care; John Sheehy, RN, BN, MBA; Francesca Rowshanzadeh, RN; Catherine Leahy, RN, BN, Stomal Therapy Nurse; Megan White, RN, BAppSc, MN; Tara Finnie, RN, BN; Carl Sharp, RN; and Judy McGlynn, RN, BN, Grad.Cert. in Aged Care, Grad.Cert. Teaching and Clinical Redesign. The authors gratefully acknowledge Judith Fethney for her assistance in statistical analyses of the patient data and for her review of this manuscript. The authors also thank the former district Director of Nursing, Ms. Katharine Duffy, for her support of this study. Finally, the lead author was awarded the Clinician Researcher Scholarship, a PhD scholarship, by Sydney Research, which in part supported this study.

Abstract

Pain during negative pressure wound therapy (NPWT) has been reported in the literature. Purpose: The study was conducted to describe patients’ pain experience, pain-coping skills, and the effect of NPWT-related pain on daily life activities following abdominal surgery. Method: Using a descriptive, qualitative design, semi-structured face-to-face interviews were conducted between April 3, 2016 and December 26, 2016, in the surgical ward of a university hospital in Edirne, Turkey. Patients aged ≥18, receiving NPWT, who had at least 1 dressing change, and with no diagnosis of diabetes mellitus or neurological disease were included. Interviews were conducted at the patients’ bedside 1 day after wound debridement. All wounds were covered with the NPWT black foam dressing, and NPWT settings were -50 mm Hg to -125 mm Hg. One (1) researcher led the interviews using a voice-recorder while 2 researchers observed and took notes. Data were analyzed using Colaizzi’s phenomenological method. Results: The themes identified were: 1) pain experience, 2) pain coping, 3) pain prevention, and 4) affects daily life activity. Patients mostly reported pain during foam dressing changes and wrap removal unless the dressing change occurred while receiving anesthesia. Self-applied pain-coping strategies between dressing changes included limiting mobility, trying not to cough, applying pressure, or walking; these strategies were mostly ineffective. The results are supported by many findings from other studies investigating the effects of NPWT on patient pain. Conclusion: This study provides further insight into the patients’ wound pain experiences during NPWT and its effect on daily activities. Increased awareness about NPWT-associated pain and pain control measures as well as qualitative and controlled quantitative studies are needed. Inservice training and educational meetings should be conducted at surgical clinics to expand surgical nurse and physician knowledge and awareness of how to efficiently manage pain during NPWT treatment and related procedures. 

Negative pressure wound therapy (NPWT) is an adjunct wound treatment method. To achieve negative pressure on the wound surface, foam dressing material is inserted into the wound, covered with an adhesive wrap, and then a small hole is cut to attach the foam to a suction pump and canister with a suction tube.¹ A systematic review² has shown improved quality of life with NPWT compared with standard wound care using gauze dressing, and an observational study³ reported better quality of life than with conventional wound therapy involving wound irrigation and saline dressing is possible for patients undergoing bariatric abdominoplasty wound management when NPWT is applied. A pilot study⁴ on quality of life after NPWT (N = 21) showed NPWT benefits include fewer dressing changes and a better social life than with standard wound treatment with gauze dressing. According to a randomized, controlled trial⁵ (RCT) involving the treatment of chronic leg ulcers, NPWT-managed wounds afforded patients increased life quality by the end of treatment. 

Despite these positive outcomes, reviews of the literature^6,7 regarding patients’ experiences show varying side effects of this therapy such as pain, stress, and anxiety. A qualitative study⁸ on patient perspectives mentions other negative impacts of this treatment such as sleep disturbances and feelings of distress. Qualitative research by Fagerdahl⁹ shows patients undergoing NPWT felt stress relative to the care environment and dressing change process. Pain has been shown in an explorative qualitative study¹⁰ to be another disadvantage for patients treated with NPWT. In a retrospective study by Schimp et al,¹¹ 67% of patients had pain during foam dressing changes that required management involving oral analgesics. Pain has been shown in qualitative and descriptive studies^10,12 to occur mostly during wound debridement, dressing change, and adhesive wrap removal; as a result of sitting for a long time in the same position; or because treatment involved a high negative pressure level. According to a review of the literature,¹³ pain also negatively affects patients’ physical activities and psychological well-being. According to a prospective study,¹⁴ patients with higher acute pain (>8 pain rating) at the first 2 postoperative days experienced slower wound healing. In a retrospective study by Apostoli and Caula,¹⁵ patients had to be given a break during treatment because of severe pain during NPWT (ie, 5 of 25 patients requested to interrupt the NPWT due to pain; average pain score 6.2 ± 2.8.) Pain also can cause stress and affect patients’ daily lives. In a qualitative study, Abbotts¹⁶ noted NPWT was experienced as stressful by patients in terms of the impacts on their daily life. 

Because nurses play a pivotal role in pain assessment, it is important for them to consider the patient’s pain experience in order to increase patient adherence with treatment and improve quality of life. Factors such as pain, sleep disturbances, and immobility can affect the quality of a patient’s life; thus, psychological factors should be considered with this treatment.^17,18 In recent years, a variety of research,^4,6,13 including a case series study¹⁹ has been conducted to explore the effects of NPWT on patients’ quality of life and pain severity. However, patients’ pain-coping skills and the daily life activities that are affected warrant additional research. The aim of this study was to explore the pain experiences of patients undergoing NPWT, including their pain-coping skills, and the effects of this pain on their daily life activities during NPWT.

Methods 

Setting and participants. This descriptive, qualitative study was conducted in the surgical ward of a university hospital in the city of Edirne in the Eastern Thrace region of Turkey. This clinic has a 48-bed patient capacity, and each room has 1 or 2 beds. Data were collected between April 3, 2016, and December 26, 2016. After reviewing qualitative studies^9,10 involving NPWT in the literature, a purposeful sampling technique was used to identify patients to include in the study. Inclusion criteria stipulated participants should be >18 years of age, willing to participate in the study, receiving NPWT in the abdominal area for the first time, not diagnosed with diabetes mellitus or neurological disease, and physically and mentally able to participate in the interview; participants also had to have undergone at least 1 dressing change following wound debridement before the study interview. 

Ethical considerations. Written permission to conduct the study was obtained from the head of the General Surgery Department and the hospital directorship. The study was approved by the ethical review board of the authors’ institution. Before the interviews, patients were informed about the aim, the context of the study, and any voice recording and confidentiality issues that needed to be addressed. The patients who volunteered to participate in the study signed an informed consent form, and they were told the information obtained would be used only for scientific purposes. 

Procedure. Negative pressure was applied continually using black foam dressing at a level of -50 mm Hg to -125 mm Hg; the level of pressure was determined by the surgeon. Wound debridement was provided only during the first dressing change for all patients. 

Study design. Data were collected using semi-structured interviews designed according to the phenomenologist guidance of Colaizzi²⁰ to obtain patients’ pain experiences (see Figure 1). After the participants provided written consent, a 9-question interview ascertained personal (age, gender, education level, and chronic disease status) and NPWT characteristics (amount of pressure, type of foam, frequency and number of dressing changes, and days since therapy started). Face-to-face interviews were conducted at the patients’ bedside 1 day after the first debridement, including the dressing change in the operating room, once they agreed to participate in the study. Because most nursing procedures (eg, preoperative patient preparation, bed sheet changes, medications, and other care procedures) typically are performed in the morning and the afternoons are not as busy, the interviews were conducted in the afternoon to maintain consistency for the patients. 

Interview process. All 3 authors (each with at least 7 years of clinical nursing experience and working as nurse academicians in the surgical nursing department of a state university) attended the interviews, which were recorded. One (1) researcher led the recorded interviews, and the other 2 observed and took notes regarding facial expressions and tone of voice, asking additional questions when necessary to ensure all necessary data were obtained (eg, if the interviewer missed an answer, another researcher could ask the participant to enlarge on his/her response). During the interviews, patients were encouraged to talk freely; the interviews took place in the quiet of the patients’ rooms to encourage them to express themselves more easily and so the voice recordings would be clear.

Patient pain severity was evaluated with a numeric rating scale (NRS). Patients were asked to indicate the intensity of their current pain level on a scale of 0 (no pain) to 10 (worst pain imaginable). Patients also were asked to describe their pain experiences during wound debridement and dressing change; severity was not evaluated. 

To ensure the credibility of this study, the voice recordings were listened to twice, and the consistency of the extracted meanings and key statements were reviewed by all authors to control the accuracy of the final themes that emerged. The researchers and an independent academician with surgical nursing research experience worked together until consensus was reached on the themes. When the themes and subthemes were decided, 4 patients were invited to approve them and comment on the results of the study.

Data collection. The collected data were entered on a spreadsheet in Microsoft Word. 

Data analysis. Colaizzi’s method of phenomenological data analysis was used with the guidance of the literature.^21,22 The steps followed from Colaizzi’s descriptive phenomenological method are presented in Figure 2. Sociodemographic characteristics were recorded and noted but not compared.

Results

Patient demographics. The 12 patient participants (mean age 65.91 ± 6.21) included 6 women. All patients had at least a primary school education. The patients had been receiving NPWT for a mean of 10.07 ± 3.75 (range 5–15) days and had 2 to 3 NPWT dressing changes per week. The average number of dressing changes for all participants was 2.25 ± 0.62. Key patient characteristics are shown in Table 1. The length of voice recordings was 33 to 41 minutes, with a mean duration of 37 minutes. Mean current pain level was 4.83 ± 2.72. Four (4) patients used analgesics to cope with pain, and 6 of the patients were debrided and had their dressings changed under sedation.

Themes. From the data analysis, 4 overarching themes were identified: 1) patients’ pain experiences during therapy, 2) patients’ pain coping skills, 3) patients’ pain prevention experiences, and 4) effects of pain on patients’ daily life activities. These main themes, together with the subthemes, are summarized in Table 2. 

Theme 1:Patients’ pain experiences during therapy. The first theme dealt with 2 subthemes. For the first subtheme, pain during treatment, patients gave information on the severity of their pain, initiating/enhancing factors, and the region, duration, and characteristics of their pain experience. For the second subtheme, pain during dressing changes, patients gave information on wound debridement, foam changes, and adhesive wrap changes.

For pain during treatment, patients quantified their pain using numbers. Patients were asked to give a number between 0 (no pain) and 10 (extreme pain) to define the severity of their pain. Patients also combined quantitative and qualitative words in several different ways, and they characterized the factors that increased or decreased their pain. In the patients’ descriptions, lying down, coughing, standing up, sitting for a long time for meals, feeling cold, and the first NPWT application period increased their pain. When patients were asked about the location of their pain, they described it as being situated in different parts of the NPWT as well as spreading from the wound. They described that the pain was felt in different NPWT locations such as under the foam or around the foam area. The patients expressed different time periods for how long their pain lasted; these periods ranged from 2 seconds to 10 minutes. When speaking about pain characteristics, they described the pain as a “pulling-out” and “stabbing in and out” feeling (a shooting pain). Some of the patients associated the pain characteristics with the sounds caused by the negative pressure and movements inside the wound during NPWT (see Table 2).

Patients also reported pain during the dressing change (ie, when the foam dressing was being applied or changed, during wound debridement, and when the adhesive wraps were removed from the skin). The patients said that while watching the wound debridement, it was very painful when the surgeon put his or her hands inside the wound or touched the wound. Patients who were provided anesthesia before debridement and dressing changes reported they did not feel any pain. Patients who had not undergone anesthesia during foam changes described the experience of pain severity as including yelling and crying. The main causes of dressing change pain were swift wrap removal and the resulting traumatized skin (see Table 2). 

Theme 2: Patients’ pain-coping skills. Patient coping mechanisms with regard to pain included using self-management, positioning, and medicine (the only option for some patients); others did nothing (see Table 2).

To self-manage pain, patients said they pressed on the painful area, covered the painful area because they thought the pain was due to cold, walked, tried to forget it, tried not to cough, and lay down on the bed. They suggested wetting the adhesive wraps before pulling them off the skin during dressing changes to cope with the pain. For positioning, the patients dealt with pain by lying on their side. Four (4) patients who needed medical treatment to cope with pain stated intramuscular injection (eg, meperidine), oral analgesics, and intravenous treatment (eg, nonsteroidal anti-inflammatory drugs) were effective. Patients were unsure about whether the medical treatment was working; 3 of the patients who experienced pain said they did not do anything to cope with the pain, tried to endure, waited for it to pass, and did not inform nurses (see Table 2).

Theme 3:Patients’ pain prevention experiences. Patients’ pain prevention experiences were categorized into 3 subthemes: staying in the same position, changing position, and doing nothing. To prevent pain, some of the patients preferred to turn slowly, stay in bed constantly except to use the toilet, and avoid coughing, while others preferred to change position by walking (moving). Some stated that they did nothing because they believed they could not do anything to prevent pain (see Table 2), but 2 patients preferred to change position by walking (moving).

Theme 4: Effects of pain on patients’ daily life activities. The study found eating, breathing, sleeping, physical activities, and communication are affected by pain; therefore, these were selected as subthemes. The patients stated eating was not affected all that much by pain, but breathing was affected (ie, feeling “not able to inhale”). The patients who experienced pain said they could not sleep easily and needed analgesics and sleeping pills to fall asleep. The patients also reported on the effects of pain during physical activities; although walking decreased pain, the patients experienced pain when sitting and getting out of bed, so they did not even want to go to toilet because of that pain. When the patients experienced pain, they did not want to see anyone or talk with anybody. Because they could not tolerate listening to others talk, they believed the pain was affecting their communication activities (see Table 2).

Discussion

Theme 1: Patients’ pain experiences during therapy.

During treatment. In the present study, some patients quantified their pain in numbers while others expressed the level as little or no pain. These results are similar to the literature. A qualitative study by Andrews and Upton²³ explored the views of 50 patients undergoing NPWT and found 81% had minimal pain (0–3 points) during NPWT. In an observational study, ²⁴ 10 patients who underwent thoracic surgery with NPWT reported no pain associated with the NPWT system and dressing change. In a descriptive and follow-up study by Egemen et al,²⁵ patients who had chronic venous ulcers and were treated with NPWT for at least 6 weeks stated they did not feel enough pain to interrupt their NPWT. On the other hand, in clinical reviews of pain and wound healing, pain during wound care and dressing changes made patients feel anxious and stressed.^26,27 In the current study, some patients cried or yelled because of their pain. These vocal expressions are a response to pain that can cause more stress. On the positive side, in this study, pain did not affect the continuity of treatment. 

Patients described pain when lying down, standing up, and sitting for a long time when eating meals, and they claimed their pain disappeared occasionally when they got up and walked around. Andrews and Upton²³ found similar results; in their study, pain was caused by turning in bed, walking, and going to the toilet. Monsen et al¹⁰ investigated the experiences of patients with NPWT provided to the groin area after vascular surgery; patients stated bending over and sitting in the same position for a long time increased pain, but changing positions helped diminish the pain. Staying in the same position for a long time and changing positions also were the pain-enhancing factors mentioned by patients undergoing NPWT after abdominal surgery in the retrospective study by Nobaek et al.²⁸ 

Patients in the present study also reported the first NPWT application procedure was very painful. According to the experimental and prospective clinical studies, patients feel pain especially during the first application of NPWT in both acute and chronic wounds.^29,30 Apostoli and Caula¹⁵ reported 5/25 patients undergoing NPWT in the hospital took a break from treatment because of pain. According to patient reports, pain was less before therapy started and increased significantly after therapy had begun. In the study by Andrews and Upton,²³ 18% of patients who had undergone NPWT said the first dressing change was the most painful period of this treatment. These findings suggest patients should be supported psychologically or with analgesics during the first application of NPWT. 

In the current study, some patients experienced instantaneous pain upon NPWT application that lasted between 2 seconds and 10 minutes. In a pilot randomized study,³¹ 10 patients provided NPWT for diabetic foot wound care stated they had pain for <30 seconds when the foam collapsed and during the foam dressing removal, even though they may have had neuropathy that could have blunted pain sensation. In a RCT by Gonzalez et al³² involving 126 patients with a total 144 lower limbs ulcers, pain relief was better after 3 days of treatment and was associated with the limited systemic inflammatory response by the third day; the systemic inflammatory response was limited and edema was decreased among patients treated with a suction system at the bedside, which does not occur as part of ambulatory wound management. 

Because pain is an individual experience, pain perception may vary from person to person. For some patients, pain was described as a pulling-out, stabbing, or tingling feeling, and others described their pain characteristics with sounds like “vıykk vıykk” and “gorrk.” Similarly, Andrews and Upton²³ stated patients who had undergone NPWT interpreted the vacuum sound of the device as a pain feature and as a strange and tingling feeling. In addition, the sounds coming from the device were described as characteristic of the pain they felt when the vacuum was turned on. 

Pain during dressing change. In the current study, patients stated they experienced pain during several stages — namely, the application of suction and during the removal of dressings and adhesive wraps. A black, foam-based NPWT dressing was used on all patients in the present study; 2 experienced pain when the foam dressing was being changed. Procedural pain during dressing changes is described in the literature and appears to be related to the type of NPWT dressing.⁷ According to the literature, including consensus reports about wound-related pain, patients experience pain during dressing changes because new granulation tissue that grows into the dressing foam is torn during dressing removal, causing pain.^33,34 Several studies have addressed dressing type and pain. In a pilot study involving 20 patients, Panicker³⁵ compared gauze dressing with the foam used in topical NPWT and reported minimal pain at dressing removal with the gauze dressing compared to the medical sponge. Fraccalvieri et al⁶ researched pain feedback among 31 NPWT patients by comparing gauze and foam; less pain was reported in persons who received gauze dressing. In the prospective, multi-centered study by Hurd et al³⁶ (N = 152), 20% of participants experienced pain with gauze-based NPWT. 

A case series³ and a pilot³⁷ study on NPWT (N = 7) recommend that white foam should be used in order to reduce pain. According to the prospective study by Mendez-Eastman³⁸ in which the guidelines for using NPWT are described, it is postulated white foam has smaller pores than black foam, thus restricting the growth of granulation tissue into the foam and reducing pain upon removal. These findings highlight that dressing type is an important factor for reducing pain.

In the current study, patients had 2 to 3 dressing changes per week, with between 50 mm Hg and 125 mm Hg of negative pressure applied. This is similar to previously published studies. According to the results of a prospective study,³⁹ dressings should be changed every 2 to 5 days, and based on international consensus,⁴⁰ a therapeutic range of -40 mm Hg to -150 mm Hg is recommended.

Similarly, previous studies conducted on patients undergoing NPWT report dressing changes every 3 to 5 days and using a pressure application of between -50 mm Hg and -125 mm Hg.^36,41 In a pilot observational study by Stansby et al,⁴² 3 NPWT dressing changes were performed per week; 22% of the patients described pain when the treatment started and 31% reported pain during dressing changes. Low-pressure NPWT (75 mm Hg) was explored in a case study with 3 patients by Nease,⁴³ in which low pain levels were reported. In an experimental study by Borgquist et al⁴⁴ that explored the influence of pressure levels during NPWT, a pressure of -125 mm Hg was required to remove exudate from peripheral wound tissue in 16 wounds, but this NPWT level can cause ischemia and pain. These results indicate negative pressure levels and dressing changes should be taken into account during treatment to control pain. If possible, the pressure level may be decreased as needed for patient comfort. 

Some patients in the current study stated they were anesthetized during the debridement and dressing change procedures and did not feel any pain. Furthermore, patients who received anesthesia did not need to take any analgesics. Schimp et al¹¹ noted that during the NPWT procedure on patients with massive wounds or extensive debridement, dressing changes were performed in the operating room under anesthesia or pain medications were used at the time of dressing change. In a consecutive case series study by Saadi et al⁴⁵ that reviewed the medical records of 27 patients who had undergone intrathoracic NPWT, NPWT dressing changes were performed under general anesthesia, thus none of the patients felt pain; Begum and Papagiannopoulos²⁴ performed NPWT dressing changes on 11 patients under general anesthesia and found similar results. In the cited studies, opioid analgesics (eg, morphine or fentanyl) and topical lidocaine also were used for pain management during wound care therapy.^19,46 A retrospective study by Wolvos⁴⁷ conducted among 5 patients receiving NPWT found using topical anesthetics on the wound reduced pain during NPWT. In summary, appropriate analgesia or anesthesia modalities can be used to increase patient comfort and decrease pain levels during treatment or dressing changes in a variety of wound types. 

In the current study, some patients described feeling pain when the adhesive wraps were removed from the skin because the skin was traumatized skin and hair pulled off. In their RCT reporting 16 patients’ experiences with NPWT, Monsen et al⁴⁸ also found removing the adhesive wrap was associated with pain and caused skin damage. In the pilot study by Ousey et al,⁴ a patient said, “It was like being waxed… it pulled the hair off,” similar to current findings. This patient recommended shaving the wound area. In their pilot RCT, Gillespie et al⁴⁹ noted 69% of the 70 patients undergoing NPWT had skin complications such as bruising. The descriptive study by Kim et al⁵⁰ explored 399 nurses’ perceptions of patient pain during dressing changes and found that although nurses were aware dressing changes caused wound-related pain, no pain reduction measures were employed. To prevent skin trauma and pulling out hair, patients can be shaven if needed and water can be applied to ease removal of adhesive wraps. 

Theme 2: Patient pain-coping skills. Patients in this study suggested wetting the adhesive wraps before pulling them off the skin during dressing changes as a way to cope with the pain. Other patients dealt with pain by lying on their side. A study that investigated the perspectives of patients undergoing NPWT found patients did not change positions to prevent pain²³ and another study that examined the experiences of patients undergoing NPWT found pain disappeared with changing positions.¹⁰ According to these results, some patients may avoid movement so patient mobility levels should be considered by health professionals.

In the present study, patients used analgesics to cope with pain; these patients’ dressing changes were not performed under anesthesia. In the retrospective study by Nobaek et al,²⁸ patient-reported outcomes and long-term results of NPWT showed 6/8 patients used drugs to cope with pain. In the study by Price and Harding,¹⁷ 64% of 135 patients undergoing NPWT for chronic wounds used analgesics in order to cope with pain; in another qualitative study²³ exploring the experiences of patients undergoing NPWT, 36% of 50 patients stated analgesics should be used for pain relief. The authors recommend nurses give patients information about pain-coping strategies and use analgesics as needed to decrease pain levels and reduce stressful conditions. 

Theme 3: Patient pain prevention experiences. In the current study, while some patients preferred to move slowly, stay in bed except to use the toilet, and avoid coughing to prevent pain, others preferred to change position by walking. In their critical analysis, Eriksson et al⁵¹ examined patients’ pain experiences after surgery and found some patients performed self-care by using distraction to avoid pain; some stated they did not want to disturb health care professionals while they were working. Similarly, 1 patient in the current study stated, “I don’t do many things to relieve the pain. I don’t tell the nurse about it.” (S, 67). Some patients tried to cope using self-care methods such as pressing on the wound, trying to forget the pain, not coughing, and lying down in bed in order to prevent pain. However, some of these pain prevention methods may cause unwanted effects. If patients try to not cough, this may retain secretions; if they try to maintain their position, immobilization may cause harmful results, all underscoring that patients should be encouraged to tell nurses about their discomfort and obtain professional help with pain management. 

Theme 4: Effects of pain on patient daily life activities. The patients in this study experienced pain when sitting and getting out of bed; 1 patient did not want to go to the toilet because of that pain. In a multicenter RCT,⁵² 132 patients reported NPWT negatively affected their normal activities and mobility. According to descriptive and prospective studies^53,54 related to pain-relieving strategies, patients were not satisfied with their nurses’ pain-relieving strategies, and patients whose pain was not adequately controlled were not able to perform activities such as moving, breathing effectively, and coughing. The current study similarly showed patients had difficulty breathing, coughing, sleeping, moving, and communicating due to pain. According to a meta-analysis by Smetana et al,⁵⁵ the general rate of pulmonary complications is between 7.7% and 19.7% after abdominal surgeries. In the current study, 1 patient reported, “I can’t cough because I don’t want to have or cause pain, so I can’t expel my mucus and my breath is diminished” (B, 49). Therefore, nurses should teach patients (especially those undergoing abdominal NPWT) about deep-breathing exercises and, if appropriate, coughing exercises to prevent pulmonary complications. 

Patients in the current study who experienced pain said they could sleep only if they were provided analgesics and sleeping pills. In the questionnaire survey study by Upton and Andrews,⁵⁶ 33% of patients had sleep disturbances during NPWT because of pain, worrying about the treatment process, and sleeping in an uncomfortable position because of the device. In the RCT by Armstrong et al⁵² on the effectiveness of NPWT devices conducted among patients with noninfected and nonischemic diabetic and venous wounds treated in various settings, 25% patients reported interrupted sleeping periods because of NPWT, so nursing activities should focus on patients’ quality of sleep and provision of interventions if needed. 

The present study also found that some patients did not want to communicate because of their pain. A qualitative study by Yılmaz and Gürler⁵³ that investigated the pain control expectations of postoperative patients found patients recommended visitor restriction and that nurses should be more considerate; for example, because communication with a patient in pain requires specific skills, short sentences can be used during communication and support can be gathered from patient caregivers. In a review by Kösgeroglu and Ünver⁵⁷ about patient education on NPWT, emotional support was underlined as an important nursing care intervention.

Limitations

This study has some limitations to be considered. First, the sample came from a single center, so the results cannot be compared with patients from other facilities. Second, the findings are based on individual interviews with hospitalized patients undergoing NPWT applied to the abdominal area; the findings do not include patients undergoing NPWT in different anatomical locations or in patients trying to function at home. Also, the comparisons of current study results to those in the literature are not limited to abdominal wounds. Third, the small number of patients (12) limits the study’s generalizability to the full population using these devices (ie, all patients undergoing NPWT). 

Implications for Nursing Education, Practice, and Research

NPWT is a globally recognized and accepted wound care therapy. In accordance with the literature, experimental studies provide evidence of the benefits of NPWT for some indications, and patient quality of life has been shown to improve by the end of therapy. Although NPWT has many positive outcomes, patients may experience varying levels of pain during the wound debridement procedure — for example, while removing the adhesive wrap from the skin, during the dressing change process, and while NPWT is in place. Pain may negatively affect patients’ physical and psychological well-being. Because pain causes stress and affects patients’ daily life, a more indepth understanding of patients’ pain experience during NPWT is needed. This may be facilitated by future qualitative studies, a methodology that may help to identify how pain influences patients’ daily life activities. Further RCTs (standard wound care versus NPWT) also are needed to investigate the long-term effects of different wound treatment methods on patients’ pain experiences and daily life activity changes. Because nurses play a vital role in pain assessment and management, it is important to consider patients’ pain experience to increase their compliance with treatment and to improve their quality of life. In accordance with consensus documents regarding wounds and pain,^33,34,41 new campaigns and instructional meetings should be held to raise awareness among all health care providers (eg, physicians, surgical nurses, physical therapists) about pain during the NPWT procedure. Following this, inservice training and educational meetings should be conducted at surgical clinics to expand surgical nurses’ and physicians’ knowledge and awareness of how to efficiently manage pain during NPWT-treatment and related procedures.

Conclusion

A descriptive, qualitative study aimed to describe patients’ pain experience, pain-coping skills, and the effect of NPWT-related pain on daily life activities following abdominal surgery. Data from the interviews of 12 patients were analyzed and 4 overarching themes were identified: pain experience, pain coping, pain prevention, and effects on daily life activity. Pain was mostly reported during foam dressing changes and wrap removal. Limiting mobility, trying not to cough, or walking were the self-applied pain-coping strategies of patients although they were mostly ineffective. This study provided further insight into the pain experience of patients during the NPWT process. The findings showed pain experienced during NPWT negatively affected patients and their daily life activities. As pain affected patients’ daily life activities such as breathing, sleeping, communicating, and mobility, nurses need to consider patient pain levels together with their daily life activities.

Acknowledgment

The authors thank to the patients for participating in this study and providing their time.