Factors Affecting Wound Healing in Individuals With Pressure Ulcers: A Retrospective Study

Owing to the number and severity of concomitant factors, pressure ulcers remain a significant problem. A retrospective study of data from adult patients with a pressure ulcer was conducted to identify factors that may affect their healing. Data from patients who were hospitalized between January 1, 2011, and December 31, 2015, in a private Turkish university hospital who had a Stage 2, Stage 3, Stage 4, or unstageable pressure ulcer that was assessed using the Bates-Jensen Wound Assessment Tool (BWAT) were abstracted.The following variables were examined: demographic characteristics (gender, age, hospital unit, duration of hospitalization), health status and disease data (vital signs, mobility, nutrition, diagnosis, chronic diseases, medication), laboratory values (albumin, hemoglobin, blood glucose), and pressure ulcer characteristics (stage, location, healing status, duration) and pressure ulcer risk status as determined by patient Braden Scale score. Seventy-eight (78) patient records were identified. Patient mean age was 70.8 ± 13.47 years, and length of hospitalization was on average 32.52 ± 27.2 days. Most ulcers (62; 79.5%) were Stage 2 and located in the sacral area (59; 75.6%). Thirty-four (34) patients (43.6%) were discharged and 44 (56.4%) died. At the time of discharge or death, 65.4% of the ulcers had not healed. Patients whose wounds were healed were significantly more likely to have higher hemoglobin and mean arterial pressure, better mobility, received oral nutrition, and discharged from the hospital than patients whose ulcers did not heal. The results suggest that these variables, including Braden Scale and BWAT scores, might be considered when developing a treatment plan of care. Additional studies examining risk factors for nonhealing pressure ulcers, including studies with large samples to facilitate multivariate analyses, are needed.

According to reviews of the literature^1,2 and a systematic review,³ pressure ulcers are associated with adverse physical, social, psychosocial, and economic effects. Pressure ulcers increase the duration of hospitalization; cause loss of labor, activity, and income; and have a negative impact on body image and self-esteem.^1-3 In April 2016, the National Pressure Ulcer Advisory Panel⁴ (NPUAP) suggested the term pressure injury should replace pressure ulcer.

Clinical, prospective, and/or prevalence studies^5-12 have found pressure ulcer frequency varies between 1.6% and 26.8% in acute and long-term care environments. In a study¹³ of patients hospitalized in 209 general medicine clinics in Turkey, the prevalence of pressure ulcers was 5.8% for patients ages 65 years and older; another Turkish study⁸ among 404 patients found a prevalence rate of 10.4%. The NPUAP¹⁴ suggests further prevalence and incidence studies of pressure injuries should be performed. 

Many risk factors for pressure ulcers have been identified in descriptive and prevalence studies.^14-18 Intrinsic factors include age, weight, ethnicity, comorbidities, tissue perfusion (such as hemoglobin, blood pressure, and chemical sedation), hypothermia, albumin level, and blood sugar and patient conditions such as sensory disorders, coma, and incontinence. Extrinsic factors include the source of the pressure, pressure duration, moisture, activities of daily life, repositioning type and frequency, friction, and shearing. The Braden Scale was developed by Bergstrom and Braden¹⁹ in 1988 to determine the risk for pressure ulcers; this instrument utilizes many of these risk factors. A validity study for this tool was performed in Turkey in 1997 by Oğuz and Olgun,²⁰ which was repeated in 1998 by Pınar and Oğuz.²¹ The Braden Scale has 6 subscales: sensory perception, mobility, activity, moisture, nutrition, and friction. The first 5 of these subscales are scored on a 1–4 scale; the friction subscale is scored on a 1–3 scale. The total score is expressed on a scale of 6 to 23.¹⁹ Lower scores indicate increased risk of pressure ulcers.^20,21

Pressure ulcers or injury can be an important indicator of the quality of care.²² According to the NPUAP consensus study,²³ many risk factors for pressure ulcers can be prevented or minimized, but complex, nonmodifiable intrinsic and extrinsic factors risk factors such as terminal illness, septic shock, body edema, immobility, and nutrition problems may render some pressure ulcers unavoidable. Therefore, pressure ulcers continue to be a significant problem despite the development of preventive programs.

It is important to evaluate the wound and its healing process to develop a proper and effective approach to wound healing.²⁴ The Bates-Jensen Wound Assessment Tool (BWAT) is among the most valid tools used for the evaluation of pressure ulcers; its use is recommended by the NPUAP.^14-18,22-26 This tool enables the comprehensive evaluation of pressure injuries and facilitates monitoring of the wound healing process.²⁷ Developed by Bates-Jensen,²⁸ this evaluation tool quantitatively defines the physiologically significant properties of the wound and comprehensively evaluates its anatomic properties. The BWAT includes 13 variables that numerically evaluate the status of the ulcer utilizing scores from 1 to 60; an additional section defines the shape and placement of the wound. Higher scores indicate a worse wound state: 1 indicates healthy tissue, 5 indicates wound regeneration, and 60 indicates wound degeneration.²⁹ The reliability coefficient of the scale was found to be 0.91 and the Spearman’s correlation coefficient 0.99.^28,30 A validity-reliability study on the BWAT was performed in Turkey in 2011 by Karahan et al.²⁷

The fact that risk factors such as hemodynamic instability, head-of-bed elevation, septic shock, burns, and terminal illness cannot be ameliorated in some patients with a pressure ulcer indicates the need for additional research regarding assessment. Still, the BWAT enables comprehensive, continuous, objective evaluation of the wound, so it is thought that information obtained from patients evaluated with this tool will contribute to the knowledge of the wound healing process and help determine the factors that affect it. The BWAT also can be used as a guide for planning patient care and have a positive impact on the outcomes of care.^{14,18,22,24,26,27}

The objective of this study was to identify factors that may affect wound healing in individuals with pressure ulcers at a private university hospital. 

Study design and setting. A retrospective study design was used involving data from all adult patients with pressure ulcers admitted to a private university hospital in Turkey whose ulcers were evaluated using the BWAT. The 346-bed hospital includes 79 intensive care unit and 267 inpatient clinic beds. A total of 300 nurses work at the hospital. Patients who need major surgery and other complicated procedures/treatments are admitted to this hospital. Nurses take direct responsibility for the prevention, evaluation, and management of Stage 1 and Stage 2 pressure injuries. Stage 3, Stage 4, or unstageable pressure injuries are treated by nurses and physicians. In this study, patients with Stage 1 pressure ulcers were excluded. 

Pressure ulcer prevention program and participants. A pressure ulcer prevention and care program was developed in 2011 at the study hospital. This program included theoretical and clinical training regarding the definition, process, staging, assessment, prevention, and treatment of pressure ulcers and the use of the BWAT and new nursing intervention forms for prevention and management of pressure ulcers. Before the program was introduced, wound assessment and care were not performed in a systematic and objective manner. This program was designed using recent literature- and evidence-based practices. The forms were developed to guide nursing care for the prevention and management of pressure ulcers, and the BWAT was implemented to evaluate ulcers that developed. In-service training was provided to nurses regarding the prevention of pressure ulcers and wound management within the program. 

In the clinical area, wound care nurses demonstrated the program content for nursing staff. Wound care nurses evaluated staff nurses during wound assessment and provision of care and provided help when staff nurses could not determine stage of the pressure ulcer or appropriate wound management.

Nurses were instructed on the BWAT for assessment. Wound care nurses explained program protocols during orientation training. Interventions for individual precautions and care were planned and applied for each patient by nursing staff. The program was controlled by the manager/supervisor and wound care nurses. Quality-assured pressure ulcer statistics were obtained, and feedback was provided to nursing staff about these data.

Data collection. A data collection form was developed based on the relevant literature.^{1,10,14,15,22,29} It comprised 15 questions that included demographic characteristics (gender, age, unit, duration of hospitalization), health status and disease data (vital signs, mobility, nutrition, diagnosis, chronic diseases, medication, healing process), laboratory values that might affect pressure ulcer healing (albumin, hemoglobin, blood glucose), pressure ulcer-related characteristics (stage, location, healing status, duration), and patient risk status.

The Braden Scale score, BWAT, and nursing notes also were used to collect data about the pressure ulcer, mobilization, and nutrition. Nurses assessed pressure ulcer risk daily using the Braden Scale. Nursing interventions for wound assessment and management were recorded directly on nursing observation forms for Stage 1 pressure ulcers; the BWAT was used daily for Stage 2, Stage 3, Stage 4, and unstageable pressure ulcers. Data related to patient prognosis was obtained from patient files. All Braden and BWAT scores in the patient files were recorded in the data collection forms for each patient, and mean scores were calculated for both the Braden Scale scores and BWAT. Limited mobility in bed or no mobility was taken into consideration when assessing mobilization problems; mobilization assessment is part of the nursing form and includes 3 options (mobile without assistance, mobile with assistance, and limited mobile in bed/no mobile). Meal consumption and Braden items regarding nutrition were recorded on nursing forms and assessed. If a patient could not sustain oral nutrition, a dietitian was consulted according to the physician’s orders. Nutrition assessment tools were not routinely used for all patients; therefore, the nutrition status data was obtained from nursing and physician notes.

Study procedure. Braden and BWAT scores and questionnaire data were acquired from patient records included in the sample group. Data on Stage 2, Stage 3, Stage 4, and unstageable pressure injuries were obtained via the BWAT, and data on Stage 1 pressure ulcers and wound healing were obtained from nursing observation and quality forms. In patients with more than 1 pressure ulcer, the ulcer with the highest stage (worst ulcer) was analyzed.

Data collection and analysis. Data from all adult patients with a pressure ulcer treated between January 1, 2011, and 31 December 31, 2015 were retrieved for analysis; however, the program is ongoing. After the data were transferred to SPSS version 17.0 (SPSS for Windows, Chicago, IL), independent variables (age, gender, unit, mobility, nutrition, laboratory values, vital signs, healing status, Braden and BWAT scores, and BWAT items) were compared according to wound healing status (healed or not healed pressure ulcer). The total BWAT and Braden scores obtained from the patient files were analyzed for each patient in addition to the first, highest (the worst situation of the wound), and last assessment of the BWAT item scores for each patient. Descriptive statistics (percent, mean, standard deviation [SD]) and chi-squared, Student’s t test, Mann Whitney U test, and Pearson’s correlation coefficient were used for comparison with independent variables and pressure ulcer healing status. A P value <.05 indicated statistical significance.

One hundred, three (103) patients had a Stage 2, Stage 3, Stage 4, or unstageable pressure ulcer. BWAT data could not be located in the records of 25, leaving a study sample of 78 patients. Of these, 41 (52.6%) were female; mean age of all participating patients was 70.8 ± 13.47 (range 22–91); and 47 (60.3%) were admitted to intensive care units and 39 (50%) were admitted to surgical intensive care units and wards. Admitting diagnoses included malignancy (19); cardiovascular (18), respiratory system (11), gastrointestinal (10), nervous (9), and urinary system (7) disease; or other health problems (4). Patients had a mean 2.17 ± 1.32 (range 0–6) chronic diseases, used 4.58 ± 2.83 (range 0–15) prescription drugs, and 41 (52.6%) used drugs that could affect wound healing (ie, immunosuppressants [4], anticoagulants [31], steroids [9], or chemotherapeutic agents [2]). Nutrition was provided orally (32, 41%), enterally (24, 30.8%), and parenterally (22, 28.2%). Forty (40) patients (51.3%) had mobility problems. Sixty-two (62) patients (79.5%) had Stage 2 pressure ulcers, 10 (12.8%) had Stage 3, 3 (3.8%) had Stage 4, and 3 (3.8%) had an unstageable ulcer. The majority of the pressure ulcers (59; 75.6%) were located in the sacral region; 9 patients (11.5%) had a pressure ulcer in more than 1 area. Ten (10) patients (12.8%) developed an infection in their pressure ulcer. The total hospitalization period of the patients was 32.52 ± 27.2 (range 4–143) days; 34 (43.6%) were discharged after healing and 44 (56.4%) died. In 51 patients (65.4%), the pressure ulcer did not heal during the hospitalization period. 

The number of patients whose wound healed was lower among patients who had mobility issues (9, 22.5%) than in patients who did not have mobility limitations (18, 47.4%). The number of patients whose wound healed was lower in the parenteral (5, 22.7%) and enteral feeding groups (4, 16.7%) compared to patients who were on an oral diet (18, 56.3%); also, fewer pressure ulcers healed in patients in the deceased group (4, 9.1%) compared to the healed/discharged group (23, 67.6%) (P <.05). A statistically significant difference could not be determined between patient healing status according to gender, age, and the hospital units where they received care (P >.05) (see Table 1). The number of chronic diseases, hospitalization duration, and the use of drugs that could affect wound healing also were not significantly associated with the healing status of the wound (P >.05). 

Patients whose pressure ulcer healed had higher hemoglobin levels (10.38 ± 1.72 g/dL) and higher mean arterial pressure (MAP) (87.51 ± 10.10 mm Hg) compared with values of the patients who did not heal (9.14 ± 1.10 and 74.34 ± 12.59, respectively; P <.05). The differences between the blood sugar, albumin, body temperature, and oxygen saturation (SaO2) values were not statistically significant between the healed and nonhealed groups (see Table 2). 

The mean Braden Scale score of all patients was 13.58 ± 2.95 (range 6–20); 73 (93.6%) were at risk for pressure ulcer development. Thirty (30, 38.5%) had high risk (≤12), 20 (25.6%) had moderate risk (13–14), 23 (29.5%) had mild risk (15–18), and 5 (6.4%) had a low risk score. The mean Braden Scale score was higher in individuals whose pressure ulcer healed (14.77 ± 2.06) compared to patients whose ulcer did not heal (13.38 ± 3.30) (P <.05) (see Table 3). 

The mean total BWAT score was 22.8 ± 5.72 (range 11.7–40). The mean BWAT total score for patients whose pressure ulcer healed (18.36 ± 3.59) was lower than that of patients whose ulcer did not heal (25.20 ± 5.21) (P <.05) (see Table 3). 

A weak negative correlation was noted between mean Braden Scale score and mean BWAT score: mean BWAT score increased as mean Braden score decreased (r=-.368, P = .001) (data not shown). In other words, a higher risk for pressure ulcer development was correlated with a more extensive or slower-healing wound. Wound healing status was compared according to the first, highest, and last BWAT total scores. Statistically significant differences were noted among all groups. The first mean BWAT total score of patients whose pressure ulcer healed (19.22 ± 4.25) was lower than that of patients whose ulcer did not heal (23.60 ± 6.39) (P <.001), the middle mean BWAT total score of the wound healing group (21.25 ± 4.88) was lower than that of the nonhealing group (28.98 ± 6.69) P <.001), and the last BWAT total score of the wound healing group (16.33 ± 4.79) was lower than that of the nonhealing group (27.47 ± 6.30) (P <.001) (see Table 4). 

Average BWAT item scores during hospitalization were compared according to pressure injury healing status. The scores were lower for all item scores except for undermining, peripheral tissue edema, and peripheral tissue induration for patients whose pressure ulcer healed compared with patients whose pressure ulcer did not heal (P <.05) (see Table 5). 

Pressure ulcers can be an indicator of nursing quality.³¹ However, not all pressure ulcers can be prevented^14,18,32; the factors that may have precipitated their development also may affect healing. The purpose of this retrospective study was to examine factors affecting pressure ulcer healing. 

Age. Elderly patients are at greater risk of developing a pressure ulcer due to many factors.^12,33 The age range of patients included in pressure ulcer studies cited ranged from 49 to 70 years.^17,34-41 The mean age of the patients in the current study was slightly higher (70.8 years). This may be because the study was performed at a university hospital where elderly patients present with greater care needs and more complex clinical situations. In this study, no difference with respect to wound healing was observed between patients younger or older than 65 years of age. 

Ulcer stage. In this study, wound healing occurred in 34.6% of the patients during hospitalization; 38.7% of patients had Stage 2, 30.0% of patients had Stage 3, and 0% of patients had Stage 4 or unstageable pressure ulcers. In their retrospective study (N = 209), Aygör et al¹³ reported wound healing occurred in only 2 of 12 patients before discharge. In a retrospective study by Smit et al,⁴² more than half of the 76 patients hospitalized at the authors’ medical intensive care unit experienced pressure ulcer healing and were discharged. In a retrospective study by Park⁴³ involving 155 patients with Stage 2 pressure ulcers in a Korean acute care hospital, 71.5% of patients healed.

In this study, a Stage 2 pressure ulcer was the most frequent type and the type most likely to heal. Previous studies, including prevalence, retrospective, and quasi-experimental studies^{8,13,34,38,42,44} have reported similar results, with the frequency of Stage 2 pressure ulcer reported as 45% to 89%. In a prospective study by Günes⁴⁵ involving 72 patients who had pressure ulcers, 49% were Stage 2, 47% were Stage 3, and 4% were Stage 4. Only 23% of pressure ulcers healed during the study period (8 weeks) and all were Stage 2. According to a retrospective study (N = 265) by Ikechukwu et al,⁴⁶ Stage 4 ulcers were less likely to heal than all other stages.

Gender and injury location. In the current study, the authors detected no gender predominance in patients with pressure ulcers. Similar results have been determined in other studies.^13,17,35,40 Pressure ulcers frequently occur in regions of bony prominence.¹⁴ In this study, pressure ulcers occurred mostly in the sacrococcygeal region (75.6%). Similar results have been obtained in other pressure ulcer prevalence studies, with rates varying between 12% and 70%.^{8,11,12,38,44,47} This is thought to be due to the fact that patients are positioned mostly in the supine position in some hospitals and cannot reposition themselves for various reasons.

Diagnoses, length of hospitalization, and discharge. The results of previous studies have differed in terms of the diagnosis, facility, and average hospitalization time of patients.^{17,34,35,38,41,42} In general, according to a retrospective logistic regression analysis⁴⁸ patients with a pressure ulcer tend to be hospitalized for a longer period compared with those that do not have a pressure ulcer. In the present study, the authors detected no statistically significant relationship between these variables and pressure ulcer healing. 

Mobility. According to retrospective and quasi-experimental studies,^17,18,38 repositioning is effective for the prevention of pressure ulcers other than in cases where decreases in capillary action thwart repositioning efforts.⁴⁸ The importance of repositioning is known in clinical practice. However, evidence is insufficient to support an optimal protocol for repositioning,⁴⁹ and a review of the literature⁵⁰ showed repositioning may be contraindicated for some patients. A retrospective study⁵¹ determined that immobile patients (N = 37) with a heel pressure ulcer and a Braden score ≤18 were at risk for developing further pressure ulcers. A prospective study⁴¹ of 369 patients showed a pressure ulcer developed in 37.8% patients who could not be repositioned, whereas  21.9% of the patients who could be repositioned developed a pressure ulcer. A retrospective study⁴³ (N = 155) found repositioning was a significant factor of healing versus nonhealing. In the present study, pressure ulcers healed in 22.5% of patients who had mobility problems during hospitalization, while pressure ulcers healed in 47.4% of patients who did not have mobility problems.

Although 44 patients died, the authors believed that based on their data their complex situations such as terminal illness, hemodynamic instability, chronic diseases or malignancy affected wound healing. 

 Nutrition. Insufficient nutrition has an adverse effect on wound healing.^14,18,39,52 According to Levine et al,⁴⁹ nutritional optimization has a beneficial effect on pressure injury healing. The European Society for Clinical Nutrition and Metabolism⁵² guideline recommends use of oral nutrition if possible. According to a systematic review by Langer and Fink,⁵³ there is no clear evidence of benefit associated with parenteral or enteral nutrition for the treatment of pressure ulcers. In the current study, the number of patients who healed was significantly higher in the oral nutrition than in the enteral and parenteral nutrition group. Although not statistically significant, average blood glucose levels were lower and albumin and hemoglobin levels were higher in patients whose ulcer healed compared with the nonhealing group. Similar results were obtained in a retrospective study⁴³ (N = 155) where hemoglobin was not associated with pressure ulcer healing. On the other hand, a guideline¹⁴ and a review⁴⁹ have shown hemoglobin levels will affect tissue oxygenation and delay wound healing.^14,49 In a prospective study (N = 326), Sung and Park⁵⁴ found pressure ulcer healing was improved when serum albumin level was higher than 2.8 g/dL. 

A decrease in blood pressure may disrupt tissue perfusion and have a negative impact on wound healing.^14,17,18 The MAP value of the patients in the present study was 78.90 ± 13.31 mm Hg. The normal MAP value range is 70 mm Hg to 110 mm Hg; therefore, the MAP values in patients with pressure ulcers were closer to the lower limits of the normal range in the present study. In this study, the MAP value of the patients who healed was significantly higher than that of the group that did not heal. The retrospective study by Park⁴³ among 155 Stage 2 pressure ulcer patients found that the mean total MAP value was 87.7 ± 11.2 — 88.6 ± 10.9 and 85.4 ± 11.8 for the wound healing and no wound healing groups, respectively (P <.05). The prospective study by Sung and Park⁵⁴ found pressure ulcer healing improved when the MAP was higher. 

Braden score. The Braden Scale is one of the most valid tools for determining the risk of pressure ulcer. The mean Braden score of the patients in the current study was 13.58 ± 2.95; the score varied between 11.9 and 17 in other studies.^{13,36,38,41,42,55} In a retrospective cohort study,³⁴ a pressure ulcer developed in >60% of patients with a Braden score 6 to 9 (very high risk) during hospitalization and in 9% of the patients with a score of 10 to 12. The total Braden Scale score was determined to be highly predictive of pressure ulcer development. In a retrospective study,⁴³ Braden scores were significantly different between patients with Stage 2 pressure ulcers that did and did not heal. In the current study, 38.5% of the patients had a high or very high risk score; patients whose wounds healed had higher scores (lower risk) than patients whose wounds did not heal (P <.05). These results led the authors to affirm the Braden Scale may be an effective tool for predicting PU healing outcomes. 

BWAT. The mean total BWAT score of the patients was 22.8 ± 5.72; the scores of patients with healing wounds were significantly lower than scores for patients with no wound healing. A limited number of studies have used the BWAT. In a retrospective cohort study (N = 24), Bates-Jensen et al⁵⁶ determined the mean BWAT total score to be 33.63 ± 8.44 for pressure ulcer patients with a spinal cord injury. In the current study, the BWAT item scores of patients whose pressure ulcer healed were lower than patients whose pressure ulcer did not heal except for 3 items (undermining, peripheral tissue edema, and peripheral tissue induration). These results also confirmed that the BWAT is a valid tool for evaluating pressure injuries. 

The present study has some important limitations. First, the number of patients was limited because the nurses could not evaluate every patient using the BWAT. In addition, some variables such as height, weight, and prealbumin level could not be obtained from the patients’ medical records. The data analysis in this study was affected by the fact that more than half of patients died during the study period and most of them were in the nonhealed group. Employee and environmental or administrative factors were not examined in this study; the authors focused on patient-related factors. Additional prospective and larger scale studies examining all factors that may affect pressure ulcer healing, as well their relationship to each other, are needed. 

A retrospective study of Turkish patients’ pressure ulcer data indicated limited mobility, insufficient oral nutrition, low hemoglobin levels, low MAP, and discharge status may affect pressure ulcer healing potential. Health care providers should consider these factors when designing and implementing a comprehensive pressure ulcer treatment plan. 

This study was approved by Baskent University Institutional review Board (Project No: KA 15/250).

Dr. Karahan is an Associate Professor; and Ms. Abbasoğlu, Dr. Işık, and Dr. Çevik are lecturers, Baskent University, Health Sciences Faculty, Department of Nursing, Ankara, Turkey. Ms. Saltan is a wound care nurse, Baskent University Ankara Hospital, Ankara, Turkey. Dr. Elbas is a professor, Baskent University. Ms Yalılı is a research assistant, Health Sciences University, Nursing Faculty, İstanbul, Turkey.

Please address correspondence to: Azize Karahan, PhD, RN, Baskent University Health Sciences Faculty, Department of Nursing, Ankara, Turkey email: azize03@yahoo.com or kazize@baskent.edu.tr.