Use of negative pressure wound therapy on locoregional flaps: a case–control study

Since its introduction and popularisation by Argenta et al.¹ and Morykwas et al.² negative pressure wound therapy (NPWT) has revolutionised the management of various types of wounds. The exact mechanisms of action of negative pressure are often related to both a mechanical (macro-strain or macro-deformation) and biological (micro-strain or micro-deformation) tissue response, which stimulates angiogenesis and granulation tissue formation.^3,4,5 In our modern surgical practice at a tertiary healthcare institution, NPWT has an indispensable role in the treatment of complex wounds, such as diabetic foot wounds, complex lower extremity wounds needing reconstruction or in wounds immediately post-skin grafting, mirroring the current evidence present in clinical literature.^6,7,8

In recent years, further research has gradually expanded the clinical use of NPWT; we have experienced NPWT being used with various foam interfaces, combined with instillation of fluids or medication, or with other advanced wound products.^3,4,5,9 Of particular interest, is the extension of NPWT use to closed incision management. In particular, there is a growing body of evidence showing efficacy of incisional NPWT for high-risk wounds, in terms of reducing risks of surgical site complications.^10,11,12,13

In the field of reconstructive flap surgery, we often encounter complex wound environments with challenging wound and patient risk factors (i.e. comorbidities), which require vascularised flap reconstruction to resurface the wound and allow healing to occur. Postoperative dressings for such complex flap reconstructions often use traditional wound dressing methods; the main considerations for such dressings are that they:

• Should exert minimal pressure on the flap and its pedicle
• Should give the flap adequate ‘space’ to accommodate postoperative flap swelling
• Keeps the wounds clean and dry to allow primary wound healing to occur
• Prevents contamination from the external environment
• Allows easy access for clinical flap monitoring.

With conventional dressings, frequent and laborious dressing changes are required to ensure that these tenets are upheld during the postoperative recovery phase.

Having observed the clinical benefits of NPWT on high-risk wounds, many plastic reconstructive surgeons had attempted to extend the benefits of NPWT on flaps in the immediate postoperative setting. There have been some reports that showed promising results when NPWT was used in muscle flaps^{14,15,16,17,18,19} and skin flaps.^{20,21,22,23,24} Most reports are limited to retrospective case series, without a meaningful comparison with the current standard of care. Therefore, the authors seek to affirm these findings, and hypothesise that NPWT dressing is safe and efficacious for use on flaps immediately in the postoperative period.

Methods

A retrospective review was carried out based on all flap reconstructions performed in a single tertiary centre in Singapore from November 2017 to January 2020. Patients' electronic medical records were reviewed to extract perioperative information. Patient demographic data, perioperative details and morbidity outcomes were collected and deidentified before conducting statistical analysis.

The study cohort was divided into two groups:

• NPWT group
• Control group (did not receive NPWT).

In the NPWT group, VAC dressing (KCI, US), Prevena Dressing (KCI, US), or PICO Dressing (Smith+Nephew, UK) was applied after flap reconstruction and removed after 4–7 days. Where applicable, the NPWT pressure setting was set at 75mmHg of continuous suction. After completion of NPWT, the flap was dressed in the same manner as the control group.

In the control group, conventional sterile dressings, without negative pressure suction, were used, such as Mepilex Border Post-op (Mölnlycke Healthcare AB, Sweden) and Dermabond Prineo (Ethicon, Johnson & Johnson, US) (Table 1). All patients in both groups had to follow a standardised strict postoperative nursing instruction with bed rest and proper positioning.

Table 1.

Study groups

NPWT—negative pressure wound therapy

In our series of locoregional flaps, perforator flaps were all fasciocutaneous in nature, with perforators identified during surgery. Local skin flaps comprised of local flaps that were raised adjacent to the wound defect in a fasciocutaneous manner, without identifying any perforators during surgery (for example, a gluteal rotation flap for sacral sore reconstruction).

Primary endpoint

The primary endpoint of this study was to investigate if the use of NPWT after flap reconstruction in the immediate postoperative period would improve the incidence of flap complications and wound healing complications. Severity of flap complications was classified according to three grades:¹⁸

• Grade 1: minor superficial flap necrosis, defined as ≤10% of the flap surface area and which does not require surgical revision or secondary procedure
• Grade 2: partial flap necrosis of >10% of the flap surface area, or any flap necrosis that requires surgical revision or secondary procedure
• Grade 3: complete flap failure with entire flap loss.

Wound healing outcomes

Wound healing outcomes included surgical site infection (SSI) and wound dehiscence; we looked at risk factors for SSIs according to the Surgical Site Infection Risk Score (SSIRS).²⁵ Other secondary outcomes studied included the duration of hospital stay and time to full functional recovery, which was defined as when all stitches were removed, the wound was fully exposed and direct weight-bearing on the flap (where applicable) was allowed.

Ethical approval and patient consent

This study was conducted in conformity with the World Medical Association Declaration of Helsinki. Ethics review and approval were obtained with our Institutional Review Board (IRB) (Reference: KTPH/2020/00124). Patients' consent for publication of their photographs was obtained. Waiver of consent was granted by our IRB for the study team to use deidentified data for the purposes of this study.

Cost analysis

Cost analysis was performed by consulting our hospital's finance department on the cost of staff time, procedure fees and consumable dressing materials which are typically billed to the patient. We chose to look at the financial implications of the first seven postoperative days as this was where the dressing regimen would differ. This amount was calculated as an estimate as we were unable to extract the exact financial information relating to dressing change alone.

Statistical analysis

Univariate analysis was conducted using t-test for normally distributed continuous variables, and Wilcoxon two-sample test for non-parametric data. Chi-squared test was used for categorical variables. Multivariate analysis was performed using binary logistic regression. Data analysis was carried out using SPSS (SPSS Inc., US). Results were considered to be statistically significant with a p-value of ≤0.05.

Results

Of the 138 patients who underwent flap reconstruction, 37 who had free flap reconstructions were excluded, and 101 patients who underwent locoregional flap reconstructions were included. (The 37 patients who underwent free flap reconstruction were studied separately as part of an enhanced recovery protocol.) The included patients (n=101) were divided into either the NPWT group (n=51) or the control group (n=50).

In the NPWT group: 37 (72.5%) patients had NPWT dressing over closed incision of a skin-type flap (for example, a pedicled perforator flap); 13 (25.5%) had NPWT dressing on a pedicled muscle-only flap with skin graft, and one (2.0%) patient had NPWT dressing on a pedicled fascial flap.

In the control group: 20 (40.0%) patients had Mepilex Border Post-op dressing; 16 (32.0%) had the Dermabond Prineo dressing, and the remaining 14 (28.0%) patients had other simple traditional dressings such as gauze with an intervening non-contact layer dressing (Table 1). Both groups used similar types of dressings after the first seven postoperative days.

Patient age ranged from 19–86 years (mean: 49.4 years); 64 (63.4%) patients were male and 37 (36.6%) were female. Risk factors that may have impacted on postoperative complications included: patient factors (smoking, body mass index); comorbidities (history of diabetes, metastatic cancer, chronic steroid use); and operative characteristics (surgical urgency, American Society of Anesthesiologists (ASA) physical status class, wound class, operation duration) were collected and compared between both groups (Table 2).

Table 2.

Descriptive statistics of study cohort

BMI—body mass index; NPWT—negative pressure wound therapy; SD—standard deviation; ASA—American Society of Anesthesiologists classification of physical health;

*

Solid tumour at >1 site; acute lymphoblastic leukaemia, acute myeloid leukaemia or stage 4 lymphoma

Demographic and operative characteristics were comparable in both groups, using a univariate analysis for all covariates. In terms of patient and surgical risk factors, there was no significant difference between study groups (Table 3).

Table 3.

Univariate analysis between NPWT group and control group

NPWT—negative pressure wound therapy; ASA—American Society of Anesthesiologists classification of physical health;

*

Solid tumour at >1 site; acute lymphoblastic leukaemia, acute myeloid leukaemia or stage 4 lymphoma

The indication for flap reconstruction included: trauma (n=45, 44.6%); infection (n=22, 21.2%); tumour with post-excision defect (n=18, 17.8%); hard-to-heal (chronic) wound (n=13, 12.8%); and burn (n=3, 3.0%). Location of wound coverage included: lower limb (n=49, 48.5%); upper limb (n=13, 12.9%); abdomen and chest (n=19, 18.8%); and perineum and gluteal (n=20, 19.8%). The various types of flap reconstruction performed included: muscle and myocutaneous flaps (n=26, 25.7%); perforator flaps (n=26, 25.7%); local skin flaps (n=48, 47.5%); and one pedicled fascial flap (n=1, 1.0%). There was a higher proportion of traumatic wounds requiring flap reconstruction within the NPWT group, with a higher proportion of flap reconstructions being in the region of the lower extremity (Table 4).

Table 4.

Indication for flap coverage, area of reconstruction and type of flap reconstruction performed

NPWT—negative pressure wound therapy

The primary outcomes of our study were analysed and summarised in Table 5. There was no significant difference in terms of flap complication rate between the two groups. There were 16 (15.8%) cases of partial flap necrosis; eight in each group. Both groups had five cases of Grade 1 flap necrosis and three cases of Grade 2 flap necrosis. There was no complete flap necrosis (or Grade 3 flap necrosis) observed in either group, and there was no significant difference in both Grade 1 and Grade 2 flap necrosis between the groups (Table 5).

Table 5.

Postoperative complications and outcomes by treatment group

*

Flap necrosis grading: Grade 1: Minor flap necrosis ≤10% of the flap surface area, and does not require a surgical revision or secondary procedure; Grade 2: partial flap necrosis >10% of the flap surface area, or any flap necrosis that requires surgical revision or secondary procedure; Grade 3: complete flap failure with entire flap loss18;

SD—standard deviation

Postoperative wound infection and/or wound dehiscence was identified in 13 (12.9%) patients (NPWT group=6; Control group=7). No significant difference was observed between the two groups. The median time from operation to full functional recovery was 22 days, and the average duration of hospital stay after operation was 11.1 days (range: 0–56 days). There was no significant difference concerning time to full functional recovery and duration of hospitalisation between the two groups (Table 5).

A multivariate analysis was also performed using a binary logistic regression model. As an independent variable, the use of NPWT did not have a significant impact on the outcomes of flap complication (p=0.376) or wound healing complication (p=0.775). The ‘indication for coverage’ was statistically significant in terms of affecting the outcome of wound infection and wound dehiscence; this would correlate especially for traumatic or infected wounds needing flap reconstruction (Table 6).

Table 6.

Multivariate analysis with flap complications of flap necrosis and wound infection/dehiscence as dependent variables

RR—relative risk; Sig—significance; ASA—American Society of Anesthesiologists classification of physical health

When comparing the estimated projected cost of using NPWT dressings versus traditional dressings for a duration of one week, NPWT dressings cost in the range of $207–415 USD versus $370–981 USD for traditional dressings. This cost difference was mainly attributed to the decreased need for regular dressing change in the first seven postoperative days when using NPWT dressing over flaps, even though the NPWT dressings were more expensive comparatively. This monetary saving was mainly derived from staff time and cost savings, because of less need for regular dressing change.

An example of the use of NPWT is given in Fig 1 which shows the case of a 57-year-old male patient with an exposed left knee joint, after multiple surgical debridements as a result of septic arthritis. A medial genicular artery perforator-based flap was performed to reconstruct the knee defect, and incisional NPWT was applied immediately after surgery (Fig 1).

Discussion

In open wounds, NPWT helps to promote a better wound healing environment by reducing oedema, removing wound effluent, promoting perfusion and increasing granulation tissue formation.¹¹ The evidence that supports the clinical benefits of NPWT in diabetic ulcers, venous leg ulcers and partial foot amputations is very convincing.⁵ In addition, the consensus recommendation for closed incisional negative pressure therapy by Willy et al.,¹¹ gives clinical guidance for the use of NPWT on high-risk closed incisions.

However, when it comes to applying NPWT directly onto a flap immediately post-surgery, there are additional challenges that have to be considered. On top of managing the usual surgical site complications, such as wound infection, dehiscence and pain etc., one should be mindful that the application of NPWT on a flap in an immediate postoperative setting should not compromise the perfusion or viability of a flap. In-depth knowledge of flap physiology and perfusion of skin and soft tissue flaps allows the reconstructive surgeon to understand that exerting direct pressure with external dressings must be done with caution, to avoid compromising flap perfusion.²⁶ The basis of this stems from previous studies by Kairinos et al., which show that NPWT ironically exerted a net positive pressure directly on the wound,^27,28 and that NPWT did not show an increase in tissue perfusion.²⁹ Given such findings, some surgeons question the benefits or even fear the deleterious effects of NPWT on the perfusion of flaps.^27,28 It is because of these considerations that we would prefer the lowered pressure setting of 75mmHg for NPWT over flaps, rather than 125mmHg on open wounds or over skin grafts. This preference resonates with some of the previous published case series of use of NPWT on flaps.^17,30,31

Overall, most published studies on the use of NPWT on flaps have shown there are more benefits than harm. Though difficult to quantify, several authors have noticed decreased flap oedema after removal of NPWT compared with what was usually observed in flaps with traditional dressing.^14,19 Chim et al.¹⁹ attempted to quantify such a reduction in swelling for muscle flaps with direct measurements. The downstream effect of reduced flap oedema is that it may translate to reduced surgical drain output, earlier drain removal and quicker recovery of the patients' wounds. This was also observed in our case series, even though it was not proven to be statistically significant. In addition, NPWT allows for assisted egress of excess post-surgical fluid between closure sutures without maceration of the wound edge. Another advantage for the patient is that there is a lower frequency of dressing change, given that the NPWT dressings can last for 5–7 days. Lastly, NPWT that is applied on the flap allows for a good seal to exclude the external environment, preventing wound contamination and allowing primary wound healing to occur before exposure to the environment.

Despite its benefits, the use of NPWT for flaps should still be applied with caution. Inappropriate application of NPWT dressing over skin flaps may result in maceration of the wound edge. Excessive tension/pressure while applying the NPWT dressing may translate to excessive pressure transmitted to the flap itself, which may precipitate flap necrosis. An adequate window should be created in the NPWT dressing for clinical flap monitoring—a fine balance has to be established between applying the opaque foam dressing over the flap with good suction seal versus exposing enough flap to aid its monitoring. In Fig 2, we demonstrate the typical result achieved with the use of Prevena dressing on a medial genicular artery perforator-based flap on a 35-year-old patient who had an open tibial fracture with a resultant complex skin and soft tissue defect at the distal third of his shin. This factor can be mitigated somewhat by using an implantable Doppler or near infra-red spectroscopy, where applicable.^14,32 Our findings of proposed risks/benefits of using NPWT dressings over flaps are summarised in Table 7.

Table 7.

Summary of proposed benefits/risks of negative pressure wound therapy (NPWT) on flaps

With regards to primary endpoints, there was no complete flap loss, with an overall partial flap necrosis of 15.8%, and only 5.9% of flaps required minor revision surgery. Wound infection and/or wound dehiscence rates were at 12.9%. Other outcome measures included the median time to full functional recovery at 22 days, and the mean duration of hospitalisation was 11.1 days. There was no statistically significant differences between the NPWT group and control group. With both flap necrosis and wound healing complications as dependent variables, multivariate analysis did not yield any statistical significance, except for ‘indication for coverage’ for wound infection/dehiscence. This should prompt a further subgroup analysis in the future, when there is a larger number of cases. Overall, we would conclude from this study that NPWT over flaps is not inferior compared with conventional dressings; perhaps larger numbers of patients would be required to prove better outcomes. Other studies have also concluded that NPWT may facilitate free flap transfer with few side-effects and help to rescue flaps with threatened viability or venous congestion, particularly due to swelling.^33,34

Certain intangible benefits, which are difficult to quantify and analyse in a scientific manner, are often keenly felt by patients and clinicians, such as convenience and reduced dressing change frequency. The authors also found that flap oedema reduced more quickly with the use of NPWT. The authors have summarised their own findings in Table 7 of the benefits/risks of using NPWT on flaps. Unfortunately, many of these benefits are not directly measurable and in this study did not translate directly to statistically significant outcomes.

Having said that, the authors feel that it would be ideal if NPWT could be placed on all flaps in the immediate postoperative period (around the first 5–7 days post-surgery); however, this would not always be possible in certain scenarios. Firstly, for some flaps, for example in areas around the scrotal, perineal and perianal regions, it would be challenging to achieve a negative pressure seal, hence, NPWT was not attempted for the flaps around these areas. Secondly, some flap reconstructions for small defects (for example, perforator island flaps for pilonidal sinus defects) were carried out as ambulatory day surgery and patients were discharged within 24 hours from time of operation, and so NPWT was not feasible in these settings. Although we did not include the 37 free flap cases within our study, we do routinely use NPWT on complication-threatened free flaps during secondary free flap surgery.

Limitations

There are limitations to our study. Due to its retrospective nature, it is susceptible to recall or information bias. Also, given our overall preference for NPWT dressings over flaps, there would be selection bias given the differing clinical circumstances. Some examples where NPWT dressing was not preferred included when patients were planned for discharge from hospital 1–2 days after surgery, or when the flap reconstruction was conducted in the perineal or perianal region, where NPWT dressing seal would be difficult to achieve and maintain. Lastly, our small sample population did not reveal a statistically favourable outcome for NPWT dressing, and future follow-up studies would be required to better discern outcomes, including a cost–benefit analysis.

Conclusion

In this study, the appropriate use of NPWT over flaps was shown to be safe, was not inferior to conventional dressing, and can be recommended as a useful adjunct in the immediate postoperative period. The main benefits of the use of NPWT over flaps include better exudate management, oedema reduction and potential cost savings. Though not statistically significant in this study, further studies with greater numbers of patients are required to ascertain if NPWT dressing on flaps would provide an overall benefit in the immediate postoperative period.

Reflective questions

• What is the ideal dressing type and regimen for a locoregional flap?
• What are some of the attributes or qualities that the ideal wound dressing should possess, in cases of post-reconstructive locoregional flaps?
• While most clinicians are familiar with the use of negative pressure wound therapy (NPWT) on wounds, what is the evidence for NPWT use on locoregional flaps?
• What are some practical pointers to which attention should be paid when starting to use NPWT on locoregional flaps, in order to optimise results?