Effect of Flap Selection on the Postoperative Success of Sacral Pressure Injuries: A Retrospective Analysis

Pressure injuries (PIs) are defined as tissue injuries caused by unrelieved pressure over a bony prominence.^1,2 Treatment of patients with this wound type can be challenging for medical professionals, nursing staff, and health care providers worldwide. Treatment success is limited to the closure of defects and prevention of long-term relapse. This wound type is commonly seen in the sacral area; anatomic location, presence of additional comorbidities, bed confinement status, malnutrition, immobility, lack of sensation, and inadequacy of social care services are factors that can negatively affect reconstruction success. In the literature, various flap options have been identified to reconstruct sacral PIs, but studies that compare these options using postoperative results are limited.^3-8

In this study, the authors conducted a retrospective evaluation of postoperative complications, clinical outcomes, and PI recurrence during 14 months of follow-up for patients who had received musculocutaneous (MC), fasciocutaneous (FC), or perforator (P) flaps for a sacral PI.

The study protocol was arranged in accordance with the ethical rules of the 1975 Declaration of Helsinki.  During 2002 to 2016, patients who underwent surgery for sacral PIs were retrospectively analyzed. Patients who had stage 3 and/or stage 4 PIs were included in the current study. Patients using immunosuppressive medication or having poor family social support were excluded. A total of 63 patients whose sacral PIs were reconstructed using FC (group 1), MC (group 2), and P (group 3) flaps were included in this study. Patients were analyzed in terms of their demographic features, wound etiology, defect size, postoperative complications, daily drainage volume, recurrence rate, hospital length of stay, and the statistical differences between the groups.

Statistical analysis

Whether the distributions of continuous variables were normal was determined using Kolmogorov–Smirnov test. Levene test was used for evaluating the homogeneity in variances. Continuous variables are presented as mean ± standard deviation or median (min–max), where applicable. The number of cases and percentages were used for the categorical data, and the mean differences between the groups were evaluated using one-way analysis of variance. The Kruskal–Wallis test was applied for comparisons of data that were not normally distributed. When the value of P from the one-way analysis of variance or Kruskal–Wallis test statistics was found to be statistically significant, a post hoc Tukey honestly significant difference or Conover multiple comparison tests were used to identify which group(s) differed. When there were 2 independent groups, the comparisons between the groups were performed using a Student’s t test or Mann–Whitney U test, where appropriate. The categorical data were evaluated by a Pearson’s χ² test, Fisher exact test, or likelihood ratio, where applicable. Data analysis was performed using IBM SPSS Statistics 17.0 (IBM Corp) software. A value of P less than .05 was considered to be statistically significant. However, for all possible multiple comparisons, the Bonferroni adjustment was applied to control for a type 1 error.

A total of 63 patients with sacral PIs were included in the study. The mean age was 45 years (range, 3–83 years); 39 patients were male and 24 were female. Sacral area reconstruction was performed with FC flaps (group 1) in 28 patients, MC flaps (group 2) in 13 patients, and gluteal artery P flaps (group 3) in 22 patients. Pressure injuries occurred because of paralysis in 35 patients and for other reasons in 28 patients. The statistical analysis of the demographic and clinical properties according to each of the groups is outlined in Table 1.

There were no significant differences between the groups regarding mean age and sex distribution (P >.05). In addition, there were no significant differences between the groups according to the number of patients with paralysis, immobility, lack of sensation, body mass index, nutritional status, nicotine-opioid use, and insurance status (P >.05).

There was a statistically significant difference between the groups regarding mean drain removal time (P <.001). Specifically, the drains in group 3 were removed earlier than in groups 1 and 2 (P <.001 and P <.001, respectively). There was no statistically significant difference between groups 1 and 2 (P >.05).

There was a significant difference between the groups regarding mean length of hospital stay (P =.004). In group 2, the length of hospital stay was longer than that in groups 1 and 3 (P =.047 and P <.001, respectively). Moreover, in group 1, the length of hospital stay was longer than group 3 (P =.017).

The mean follow-up time was statistically similar between the 3 groups (P >.05). There was a statistically significant difference between the groups regarding daily drainage amount (P <.001). In group 2, the mean drainage amount was significantly higher than in groups 1 and 3 (P <.001 and P <.001, respectively). Group 1 also had a significantly higher mean drainage volume than group 3 (P =.0031; Figure 1).

There was a statistically significant difference between the groups regarding incidence of recurrence, with groups 1 and 2 exhibiting a significantly higher recurrence rate than group 3 (P =.025 and P =.050, respectively). No statistical difference was detected between groups 1 and 2 (P >.05) (Figure 2).

No statistical difference was found between the groups regarding major necrosis, Fournier's gangrene, infection, fat necrosis, or marginal necrosis (P >.05).

There was a statistically significant difference between the groups regarding wound dehiscence (P =.02). Dehiscence was more frequently seen in group 2 than group 3 (P =.019). There was no significant difference between groups 1 and 2 and groups 1 and 3 (P >.05) (Figure 3).

No statistical difference was detected between the groups regarding major and minor complications (P >.05). Similarly, with mean defect size and rate of exitus, there was no statistical difference between the groups (P >.05).

In each group, among the patients younger than 50 years and older than 50 years, there was no significant difference in dehiscence or minor, major, or total complication incidence according to the Bonferroni adjustment (P >.05).

In each group, among the patients who had paralysis and those who did not, there was no significant difference regarding dehiscence, minor, major, and total complication incidences according to the Bonferroni adjustment (P >.05).

There was no statistical correlation between the mean drain removal time and dehiscence, or minor, major and/or total complication incidence according to the Bonferroni adjustment in each group (P >.05).

There was no statistical correlation between the mean length of hospital stay and dehiscence, or minor, major, and/or the total complication incidence according to the Bonferroni adjustment in each group (P >.05).

Statistically, no correlation was present between the mean defect size and dehiscence, or minor, major, and/or total complication incidence according to the Bonferroni adjustment in each group (P >.05). However, the patients in group 2 (MC) who had at least 1 complication had a significantly greater defect size than those without complications (P =.008) (Table 2).

There was no correlation between the daily drainage amount and dehiscence in each group according to the Bonferroni adjustment (P >.05). No correlation was detected between the daily drainage amount and minor and/or major complications according to the Bonferroni adjustment (P >.05). The patients in group 3 with minor and/or major complications had a significantly higher mean daily drainage than those without complications (P =.002).

Sacral defects caused by PIs usually make reconstruction difficult because of anatomic location and other patient-related factors. The flaps used in PI reconstruction serve the same purpose, but the use of muscle remains controversial.^9-16 Additionally, data on the comparison of flap choice and the long-term success rates are limited.^17-20 All reconstructive techniques have particular advantages and disadvantages. Furthermore, a clinical comparison of the success rates of these techniques can serve as a guide to the surgeon. Although the current study was retrospective in nature, the variables that could affect the results were minimized as much as possible by including only sacral PIs.

Drain removal occurred later in patients who underwent reconstruction using the MC flaps (group 2). It was considered that this finding might be due to the wide dissection and dissection in the vascular tissue planes for flap elevation.

The mean length of hospital stay was found to be longer in the FC and MC groups (group 1 and 2, respectively) when compared with the P group (group 3). This may result from P flap dissection usually being performed in a limited area compared with MC and FC counterparts. As a result, early drain removal in this group might play a role in this observation.

Daily drainage volume was lowest in the P flap group (group 3). It was considered that this finding might be due to the fine and delicate nature of the perforator dissection and strict microhemostasis.

When recurrent cases were compared between each of the 3 groups, recurrence was found to be lower in the P flap group (group 3). This may be because perforator dissection is a time-consuming process and requires microsurgical techniques, which lengthen the surgical time in most cases. Therefore, this reconstruction might not be preferred in patients in poor condition. Although the 3 groups exhibited no differences regarding the number of patients with paralysis, the lowest percentage of patients with paralysis were found in the P flap group. This explains why a low rate of recurrence can be accepted as a normal result in the P flap group.

Similar recurrence rates were observed in the FC (group 1) and MC (group 2) groups, and this finding is consistent with data in the literature.^21-24

In a study by Thiessen et al,⁹ the authors determined that PIs reconstructed with FC or MC flaps exhibited no differences regarding the presence of hematoma, seroma, wound dehiscence, or secondary procedures.

There were no differences found between the FC and MC groups regarding wound dehiscence and necrosis, which is in line with other studies. However, wound dehiscence was determined to be significantly lower in the P flap group (group 3) when compared with the MC group (group 2). Therefore, the long pedicle length of the P flap allows for a tension-free and easy adaptation of the flap to the defect and might have caused this difference. Additionally, the high rate of dehiscence in MC flaps can be explained by the limited arc of rotation due to the muscular components.

In the literature, MC flaps were accepted as a standard treatment modality for many years due to their robust blood supply and resistance to infection.^25-27 In some studies, MC flaps were found to be superior to FC flaps in managing infected wounds^25-27; other studies demonstrated no such superiority.^28,29 In the current study, no significant difference was found between the 3 groups.

From the authors’ experience, the main determining factor in infection rates is related to the total bursectomy procedure rather than flap choice.

In the multivariate analysis, a significant relationship was found between the defect size and rate of complications in the MC flap group (group 2). It was considered that this might be due to the limited arc of rotation inherent to MC flaps when compared with the FC and P flap groups.

The study performed by Thiessen et al⁹ is the most similar to the current study. However, in this study, the P and FC flaps, which were used for the sacral area reconstruction, were ranked within the same group.¹¹ The P and FC flaps exhibit significant differences regarding the blood supply, tension at the closure, and the width of the dissection area.

In the authors’ opinion, because of the aforementioned differences, P and FC flaps should be evaluated separately. Therefore, the current study differs from other similar studies in the literature.

The limitation of the study is that it is not a prospective randomized controlled study; rather, it is retrospective. In addition, it is very difficult to standardize patients according to groups in wound studies, and postoperative patient care conditions may vary. This situation also makes it difficult to evaluate recurrence in these patient groups.

Patient-, wound-, and surgeon-related factors are adequate for the flap choice in sacral PI reconstruction. Per the authors’ clinical experience, P flaps are preferred as the first choice in appropriate patients because they are associated with shorter mean length of hospital stay, reduced daily drainage, and decreased recurrence rate. The current study will be a guide for future randomized controlled studies.

Authors: Serhat Şibar, MD¹; Kemal Findikcioglu, MD¹; Kirdar Guney, MD²; Serhan Tuncer, MD¹; and Suhan Ayhan, MD¹

Affiliations: ¹Gazi University Hospital, Department of Plastic, Reconstructive and Aesthetic Surgery, Ankara, Turkey; ²Renee Clinic, Istanbul, Turkey

Correspondence: Serhat Şibar, MD, Gazi University Hospital, Department of Plastic Reconstructive and Aesthetic Surgery, 14th floor, Besevler, Ankara, Turkey; serhatsibar@hotmail.com

Disclosure: The authors disclose no financial or other conflicts of interest.