Prophylactic Negative Pressure Wound Therapy Reduces Superficial Surgical Site Infection Risk of Emergency Surgery Patients: Results of a Multicenter Randomised Prospective Clinical Trial
Funding: The authors received no specific funding for this work.
ABSTRACT
Despite modern aseptic precautions, surgical site infection remains a significant problem. Although the benefits of negative pressure wound therapy in the treatment of chronic wounds are well established, high-level evidence is still lacking on the potential role of negative pressure in the prevention of surgical site infections. We conducted a multicenter, randomised, prospective trial of closed incision vacuum therapy. A total of 90 general surgery patients undergoing emergency laparotomy were enrolled and randomised, 45 cases in the treatment group and 45 cases in the control group. Our aim was to show a significant difference in the rate of surgical site infection between the two groups. In the study group, laparotomy wounds were treated with a single 5-day course of prophylactic vacuum therapy, whereas the control group underwent conventional postoperative wound management with sterile gauze dressings. Ten of the 45 patients in the study group developed a surgical site infection compared to 20 of 45 in the control group (22.2% vs. 44.4% p = 0.025). Upon further analysis, the proportion of superficial SSIs was found to be significantly lower in the ciNPWT group (40% vs. 20% p = 0.038), whereas the difference in deep SSI rates was not statistically significant (4.4% vs. 4.4% p = 1.0). In conclusion, negative pressure wound therapy is not only an effective way to heal chronic wounds, but it's prophylactic use may reduce the overall rate of surgical site infections.
Trial Registration: Clinicaltrials.gov: NCT03716687
Summary
- Multicenter, randomized prospective trial in high-risk general surgery patients.
- ciNPWT significantly reduced superficial and overall SSI vs standard dressings.
- No significant difference in deep SSI rates between groups.
1 Introduction
In the early days of surgery, wound infection after surgery was generally regarded as an inevitable phenomenon, a necessary part of the healing process. Later in the 19th and 20th centuries, with the discoveries of Louis Pasteur and Joseph Lister, asepsis and aseptic precautions began to spread throughout the world, dramatically reducing morbidity and mortality in surgical patients. However, surgical site infection (SSI) remains a significant problem today and continues to drive the need for modern aseptic solutions.
In general, approximately 5% of patients undergoing surgery will develop an SSI [1]. This rate can be as high as 7%–32% in emergency laparotomies for acute abdominal disease [2]. It is undoubtedly a serious problem because patients with SSI usually experience a poor quality of life with limitations in physical, social and psychological functioning [3]. The healthcare professional-patient relationship is also negatively affected [3]. The postoperative length of stay can be 7–11 days longer, which means higher costs and thus a financial impact on the healthcare system [4].
SSIs can be classified according to the depth of tissue involved: superficial incisional SSI, which involves only the skin and subcutaneous tissue of the incision; deep incisional SSI, which involves the deep soft tissues of the incision (e.g., fascia and muscle layers); and visceral SSI, which involves any part of the body deeper than the fascia and muscle layers that is opened or manipulated during surgery [5]. Risk factors (or prognostic factors) for developing a wound infection after surgery include patient-related factors (diabetes, smoking, obesity, low albumin level, PAD, anaemia, age, gender, deprivation), disease-related factors (dignity, sepsis, immunosuppression) and surgery-related factors (wound class, reoperation(s), duration of surgery, length of incision, emergency/elective, blood loss) [4].
Implementing various aseptic strategies may reduce the risk of SSI, but few are supported by randomised trials. For example, avoiding razors for hair removal (4.4% with razors vs. 2.5% with clippers), using chlorhexidine gluconate and alcohol-based skin preparations (4.0% with chlorhexidine gluconate plus alcohol vs. 6.5% with povidone-iodine plus alcohol), maintaining normothermia with active warming such as warmed intravenous fluids or skin warming (4.7% with active warming vs. 13% without) and perioperative glycaemic control (9.4% with glucose < 150 mg/dL vs. 16% with glucose > 150 mg/dL) may all reduce the risk of wound infection, although it is important to emphasise that even with strict implementation of the above strategies, SSI remains a serious complication and a significant cause of perioperative morbidity [4]. Finally, surgical site infections (SSIs) represent a significant burden on healthcare budgets, a fact supported by several official reports [6]. Negative Pressure Wound Therapy (NPWT) has been around in its modern form since the 1990s [6]. It's based on the sub-atmospheric pressure technique, where an open-cell foam dressing is placed in the wound cavity and a sub-atmospheric pressure of around −120 Hgmm is applied either continuously or intermittently.
NPWT improves wound healing in a number of ways: it promotes the formation of granulation tissue, continuously removes debris and oedema, improves vascularisation and the cellular and humoral immune response, and provides the moist environment essential for optimal wound healing [7]. The theoretical background of prophylactic NPWT (or closed incision NPWT-ciNPWT) is that negative pressure can effectively cause vertical drainage of wound exudate, any fluid collections, preventing superinfection of the collection and providing all the additional local effects mentioned above (oedema, blood perfusion, antibiotic concentration, local immune response, etc.) [8].
Several case studies and case series have been published in recent years suggesting the efficacy of ciNPWT in the prevention of SSI in various types of wounds [9]. Among these studies were retrospective and prospective investigations, as well as meta-analyses [10-12]. However, the indication spectrum for ciNPWT is poorly defined for cost effectiveness reasons, and a high level of evidence to support this indication is still lacking. The majority of the available studies involved non-general surgical patients, and the cases were seldom of an emergency nature [10]. In a retrospective study evaluating the role of ciNPWT in the prevention of SSI in vascular surgery patients, the incidence of SSI was significantly lower in the treatment arm [13]. A randomised, controlled trial investigated the potential prophylactic effect of prophylactic NPWT on orthopaedic patients and obtained promising results [14], whereas another prospective study focusing on patients after breast reconstruction also resulted in lower incidence of complications in the NPWT group [15]. The abovementioned studies presume the positive effect of NPWT on the prevention of surgical site infections; however, a knowledge gap still remains regarding the utility of ciNPWT in general surgical patients undergoing emergency procedures.
The aim of the present study is to evaluate the potential benefit of prophylactic NPWT in the prevention of surgical site infections in high-risk abdominal surgical wounds.
2 Materials and Methods
2.1 Study Design
We conducted a prospective, randomised, multicenter study of high-risk surgical patients. Equipoise of data from hospitals with different levels of progression and different geographical areas was maintained by equal numbers of cases accepted from each center. Based on previous single-center results published at a conference (LINK Congress 2019), an enrolment of 300 cases (150 ciNPWT-150 control) was estimated to demonstrate a 40% reduction in the overall SSI rate.
2.2 13 Centers in Hungary Were Approved to Participate in the Study
Based on previous single-center results published at the congress, an enrolment of 300 cases (150 ciNPWT-150 control) was estimated to demonstrate a reduction in the overall SSI rate by 40% (13% exposure-discordant pair, one-sided alpha risk 5%, power 0.6) with a significance of p less than 0.05 [16].
2.3 Randomisation
At each participating center, the senior surgeon in charge determined patient eligibility. Enrolled cases underwent software-assisted randomisation to either the experimental group (ciNPWT) or the control group. Randomisation took place immediately after surgery or no later than the following morning (for overnight laparotomies).
2.4 Inclusion Criteria
Any patient (male or female) between the ages of 18 and 80 whose abdominal wound was classified as ‘high risk’ (SSI risk at least 3 times higher than the normal rate (6%–8%), class III-IV) could be included in the study. SSI risk could be assessed using a validated risk calculator or based on the surgeon's clinical judgement. The recommended SSI risk calculator was: https://www.ohri.ca/SSI_risk_index/Default.aspx.
2.5 Exclusion Criteria
Patients who refused to give a consent, patients requiring open abdominal wound management, patients with disseminated malignancy (peritoneal carcinomatosis, abdominal wall malignancy), patients with planned re-laparotomy within 5 days, pregnant women, patients with life expectancy of 3 months or less were excluded from the study.
2.6 Experimental Arm
In the experimental arm, wounds were dressed with a layer of silver-containing impregnated mesh (polyamide fibres coated with elemental silver and an ointment-impregnated layer based on triglycerides) placed directly on the primary closed wound, a layer of alcohol-soaked foam (made of hydrophilic polyvinyl alcohol (PVA) and is pre-moistened with sterile water) and a self-adhesive sealing film of sufficient size. Finally, the portable vacuum device was connected with a disposable canister system. An electronically regulated vacuum pump removes air from the sub-dressing wound environment, creating controlled negative pressure that promotes wound healing through enhanced perfusion, exudate removal, and stimulation of granulation tissue. A negative pressure of −90 Hgmm was routinely applied, but this could be increased to −120 Hgmm in the case of high output wounds. If the patient complained of wound pain, the pressure was reduced to −70 Hgmm. Preventive NPWT was defined as one dressing for 5 days without change or manipulation. According to the study protocol, ciNPWT had to be applied at the end of the emergency procedure or by the following morning at the latest in the case of late-night operations.
2.7 Control Arm
The control arm wounds were treated as to the local routine: sterile gauze or highly absorbent surgical dressing with or without subcuticular drainage.
2.8 Wound Assessment
At the end of the 5-day treatment period, a blinded dedicated surgeon removed the entire wound dressing and assessed the surgical wounds. Follow-up was performed at 5 and 30 days.
The primary endpoint was surgical site infection requiring re-opening (suture removal) and open wound management (Clavien-Dindo 2+) within 30 days. The secondary endpoint was the rate of full-thickness abdominal wall dehiscence requiring re-operation.
2.9 Data Recording
Study data were collected and managed using REDCap electronic data capture tools [17, 18].
2.10 Ethical Background
The study was conducted in accordance with the Declaration of Helsinki and with the approval of the National Scientific and Ethical Committee and the National Institute of Pharmacy and Nutrition (OGYÉI/15347-5/2018). The study was registered on Clinicaltrials.gov: NCT03716687.
2.11 Statistical Analysis
Descriptive statistics of discrete variables were calculated and reported as counts and percentages, continuous variables as means ± standard deviation in case of Gaussian distribution and median, minimum (min), maximum (max) and interquartile range (IQR) when the distribution was skewed. Continuous variables were compared using Student's t-test or Mann–Whitney U-test, as appropriate. Categorical variables were compared using χ2 tests. A p value less than 0.05 was regarded as significant in each test.
3 Results
13 centers in Hungary were approved to participate in the study, but in the end only 8 surgical centers were able to enrol patients. A total of 125 cases were enrolled (62 ciNPWT, 63 control), of which 81% were emergency cases. The final number of patients evaluated was 90 (45 ciNPWT cases, 45 control cases).
3.1 Patient Characteristics and Equipoise in the Experimental and Control Groups
Of these 90 patients, 47 were male (26 in the ciNPWT group and 21 in the control group) and 43 were female (19 in the ciNPWT group and 24 in the control group). Of the baseline parameters recorded, age (60.96 ± 15.3 vs. 67.67 ± 14.3 p = 0.034) and ASA score (p = 0.006) were significantly different between the two groups (Table 1). In the multivariate logistic regression analysis, neither ASA score (OR = 1.324; 95% CI: 0.704–2.493; p = 0.384) nor age (OR = 1.016; 95% CI: 0.981–1.053; p = 0.364) were significantly associated with the development of SSI. Patients were treated with antibiotic prophylaxis (44.4% and 37.8% p = 0.52) or antibiotic therapy (53.3% and 46.7% p = 0.527) in the ciNPWT and control groups respectively (Table 1). Incision type, abdominal or skin closure technique did not differ significantly (Table 3).
| Outcome | ciNPWT group (n = 45) | Control group (n = 45) | p |
|---|---|---|---|
| Age (mean ± SD) | 60.96 ± 15.3 | 67.67 ± 14.3 | 0.034 |
| Sex | 0.291 | ||
| Male (%) | 26 (57.8) | 21 (46.7) | |
| Female (%) | 19 (422) | 24 (53.3) | |
| BMI (kg/m2) (mean ± SD) | 26.64 ± 5.76 | 27.38 ± 5.59 | 0.537 |
| ASA | 0.006 | ||
| I. | 13 (30.2) | 1 (2.3) | |
| II. | 20 (46.5) | 27 (61.4) | |
| III. | 8 (18.6) | 13 (29.5) | |
| IV. | 1 (2.3) | 3 (6.8) | |
| V. | 1 (2.3) | 0 (0) | |
| Not reported | 2 (4.6) | 1 (2.3) | |
| Smoking (%) | 17 (37.8) | 15 (33.3) | 0.565 |
| Alcohol (%) | 4 (8.9) | 9 (20) | 0.134 |
| Peripherial artery disease (PAD) (%) | 5 (11.1) | 8 (17.8) | 0.368 |
| COPD (%) | 2 (4.4) | 4 (8.9) | 0.398 |
| Diabetes (%) | 8 (15.1) | 7 (15.2) | 0.753 |
| Type I. (%) | 2 (4.4) | 3 (6.7) | |
| Type II. (%) | 4 (8.9) | 4 (8.9) | |
| Untreated (%) | 1 (1.9) | 0 (0) | |
| Chronic renal failure (%) | 3 (6.7) | 3 (6.7) | 1.0 |
| Chirrosis (%) | 1 (2.3) | 1 (2.3) | 1.0 |
| IBD (%) | 1 (2.3) | 1 (2.3) | 0.603 |
| Metastatic malignant disease (%) | 1 (2.3) | 2 (4.6) | 0.557 |
| Active chemotherapy (%) | 2 (4.4) | 2 (4.4) | 1.0 |
| Steroid therapy (%) | 1 (2.2) | 4 (8.9) | 0.167 |
| Sepsis | 15 (33.3) | 17 (37.8) | 0.527 |
| Op. time (median; min–max; IQR) | 90; 40–215; 51 | 90; 30–193; 68 | 0.282 |
| Laboratory findings | |||
| Albumin (g/l) (mean ± SD) | 36.03 ± 8.51 | 34.53 ± 8.32 | 0.433 |
| Haemoglobin (g/l) (median; min–max; IQR) | 124; 93–162; 25 | 131; 85–164; 32 | 0.256 |
| Antibiotics | |||
| Profilactic | 20 (44.4) | 17 (37.8) | 0.52 |
| Therapy | 24 (53.3) | 21 (46.7) | 0.527 |
- Note: Statistically significant p-values (p < 0.05) are indicated in bold.
| Outcome | ciNPWT group (n = 45) | Control group (n = 45) | p |
|---|---|---|---|
| No SSI (%) | 35 (77.8) | 25 (55.6) | 0.025 |
| SSI (%) | 10 (22.2) | 20 (44.4) | 0.025 |
| Superficial SSI (%) | 9 (20) | 18 (40) | 0.038 |
| Deep SSI (%) | 2 (4.4) | 2 (4.4) | 1.0 |
| Abdominal dehiscence (%) | 1 (2.2) | 2 (4.4) | 0.513 |
| Mortality | 7 (15.6) | 5 (11.1) | 0.535 |
- Note: Statistically significant p-values (p < 0.05) are indicated in bold.
3.2 Primary and Secondary Outcomes
A total of 10 (22.2%) surgical site infections were recorded in the ciNPWT group (Table 2). Of the SSI cases, 9 (20%) were superficial, 2 (4.4%) were deep and there was only 1 (2.2%) case of abdominal wall dehiscence. In the control group, there were 20 (44.4%) cases of surgical site infection (p = 0.025). The subgroup of superficial surgical site infection included 18 (40%) patients (p = 0.038), 2 (4.4%) patients developed deep SSI (p = 1.0), whereas there were also 2 cases of abdominal wall dehiscence (4.4%) (p = 0.513). The data on ICU stay are presented in Table 4. No statistical significance was found for deep SSI (2 and 2 p = 0.885) and abdominal dehiscence (1 and 2 p = 0.506) (Table 2).
4 Discussion
There are numerous studies that have successfully confirmed the clinical benefits of ciNPWT; however, in general surgery, the vast majority of studies have focused on elective surgery.
Zaidi et al. performed a retrospective review at a single institution and evaluated 181 patients. Of these 181 patients, 69 received ciNPWT and 112 were treated with gauze dressings. According to the method statement, only high-risk patients undergoing midline laparotomy were included in the study. To be considered high risk, the patient had to have a BMI greater than 35 or meet at least two of the following criteria: malignancy, history of smoking, immunosuppression, malnutrition, atherosclerosis or emergency surgery. Two patients in the ciNPWT group (2.9%) and 23 (20.5%) in the control group developed SSI (p < 0.0009). However, emergency surgery was not a complusory factor in this study. Although this retrospective cohort analysis showed a significant reduction in the risk of SSI, there is a high risk of selection bias [19].
Boland and colleagues conducted a systematic review and meta-analysis of randomised controlled trials of prophylactic negative pressure wound therapy in laparotomy wounds. Five RCTs were included and a total of 931 patients were studied (467 ciNPWT and 464 control cases), of which only 44 were emergency laparotomies (4.7%). The overall SSI rate was 18.6% (n = 87/467) in the ciNPWT group and 23.9% (n = 111/464) in the control group, a difference that reached statistical significance. An important and notable conclusion of the study is the relatively low number of emergency cases (44/467 patients). In their conclusion, the authors encourage the recruitment of more emergency cases, as emergency itself is a known high-risk factor for SSI [10].
In our study, we focused on the effect of ciNPWT only after emergency laparotomy in general surgery patients. It was a multicenter study with several participating centers all over Hungary, but since it was not funded, the effort invested by the heads and working groups of each center was very heterogeneous. Therefore, the recruitment dynamics lagged far behind our expectations and the case numbers based on the previous power calculation. This is the reason for the relatively small number of cases included in the study.
In a similar study, Garg et al. addressed the same issue. They randomised 50 patients into a study arm (ciNPWT; n = 25) and a control arm (dry gauze dressing; n = 25). According to their results, 12% of the experimental group developed an SSI, compared to 32% of the control group. Although there was a clear trend towards benefit in the experimental arm, this difference was not statistically significant (p = 0.08). (The only positive finding from Garg's study was the lower frequency of dressing changes in the ciNPWT group, which was not surprising given the routine 5–7 day run-off time in any NPWT setting) [12]. In contrast, our study included twice as many participants, which was sufficient to demonstrate a statistically significant reduction in the incidence of both total and superficial surgical site infections (SSIs). The success of our study was partly due to the exceptionally high SSI risk of the enrolled cases. Lakhani et al. published a meta-analysis including 7 studies (4 prospective and 3 retrospective) and a total of 1199 patients (ciNPWT: 566, control: 633) undergoing emergency laparotomy. The results showed that the use of ciNPWT was associated with a significantly lower incidence of SSI, making it a promising method in the prevention of surgical site infections. (OR = 0.43, 95% CI: 0.30–0.62, p < 0.001) [11]. The use of antibiotic prophylaxis or therapy to prevent SSIs is already well established and widespread. We recorded whether the patient was treated with antibiotic prophylaxis or therapy, and the calculations showed no significant difference between the treatment and control groups. These results rule out any possible confounding effect of antibiotics on the results of our study.
The role of abdominal wall suturing techniques after emergency surgery in the development of surgical site infections has not been clarified. Polychronidis et al. performed a randomised controlled trial (CONTINT) to compare the two most preferred techniques of abdominal closure (interrupted and all layer continuous sutures) in terms of early (abdominal rupture) and late (postoperative hernia) complications. They reported that there was no significant difference between the two arms [20]. We obtained the same results: different types of sutures were categorised in our study, but without a statistically significant discrepancy between the treatment (ciNPWT) and control arms (Table 3). Although our cases were not categorised according to suture length, Millbourn et al. found an interesting correlation: after performing a randomised controlled trial in which 737 patients were enrolled, they reported that the rate of surgical site infections (and in addition to abdominal dehiscence, superficial and deep SSI were included) was higher in the long suture group than in the short suture group (10.2% vs. 5.2% p = 0.02). This was interpreted as a result of the greater amount of subcutaneous necrotic tissue in the long suture group [21].
| Wound classification (%) | 0.354 | ||
| I/A clean | 1 (2.2) | 0 (0) | |
| II/B clean-contaminated | 3 (6.7) | 7 (15.6) | |
| III/C contaminated | 26 (57.8) | 21 (46.7) | |
| IV/D infected | 15 (33.3) | 17 (37.8) | |
| Skin incision (cm) (median; min–max; IQR) | 20; 8–30; 8 | 20; 12–45; 11 | 0.119 |
| Incision type (%) | 0.794 | ||
| Median | 38 (84.4) | 39 (86.7) | |
| Paramedian | 0 (0) | 1 (2.2) | |
| Subcostal | 3 (6.7) | 3 (6.7) | |
| Transverse | 2 (4.4) | 1 (2.2) | |
| Other | 2 (7.5) | 1 (2.2) | |
| Abdominal closure (%) | 0.503 | ||
| Simple interrupted | 30 (66.7) | 28 (62.2) | |
| Continuous | 6 (13.3) | 11 (24.4) | |
| Continous-interlocking | 3 (6.7) | 3 (6.7) | |
| Other | 2 (4.4) | 2 (4.4) | |
| Not specified | 4 (8.9) | 1 (2.2) | |
| Skin closure technique | 0.573 | ||
| Simple interrupted | 38 (84.4) | 41 (91.1) | |
| Running | 1 (2.2) | 1 (2.2) | |
| Skin clips | 6 (13.3) | 3 (6.7) |
According to the literature, the mortality rate of emergency abdominal surgery can be as high as 20%. In a Danish prospective study where no ciNPWT was used, patients under 70 years of age had a mortality rate of 12.4%, whereas among patients older than 70, the 30-day mortality was 22% [22]. Given this, our results (15.6% vs. 11.1% in the ciNPWT and in the control group respectively, p = 0.535) might be considered quite normal, with no statistically significant difference between the two study arms. However, we acknowledge the exceptionally high-risk nature of our patient population. This might be one of the reasons for the high risk of SSI, although we cannot state with certainty that the mortality rate is a direct consequence of the SSI rate.
An alternative method to NPWT and closed incision NPWT along the traditional NPWT system (including the machine and the cannister for fluid collection) is single use NPWT (sNPWT). Some of these are cannister-free, and although the pressure is continuous and not adjustable, its smaller size and ease of portability give it relevance among negative pressure systems. In the case of SSI prevention, small chronic wounds or skin graft protection, sNPWT may be the appropriate choice for a discharged patient in otherwise good general health. During the COVID-19 era, the importance of avoiding unnecessary hospital admissions and shortening hospital stays was recognised. Banasiewicz et al. summarised their views and reported a possible treatment algorithm for patients with chronic wounds and high-risk patients undergoing urgent surgery. They also highlighted the importance of telemedicine and patient education [23].
According to the Centers for Disease Control and Prevention's (CDC) 2024 report, surgical site infection is the most common and costly form of healthcare-associated infection. The annual cost is estimated at $3.3 billion, or about $20.000 per admission, with an average length of additional stay of 9.7 days. Preventive procedures (including ciNPWT) may seem expensive, but in the long run they are likely to be more cost-effective than treating SSI. Nherera et al. were able to confirm this assumption by conducting a prospective observational study of 2621 cardiac surgery patients after coronary artery bypass grafting (CABG) surgery. In the study arm, patients were treated with single use incisional NPWT (sNPWT), whereas the control arm received standard care. The estimated mean cost per patient was 19.986 Eur in the sNPWT group and 20.572 Eur in the standard care group, resulting in a cost saving of 586 Eur per patient. At the same time, the proportion of wounds that healed without complications was 92.5% in the study group and 75% in the control group, a significant difference (p = 0.03) [24]. Although our study design included a cost-effectiveness analysis, we were unable to draw a definitive conclusion on this aspect of ciNPWT due to incomplete data collection. The difference of length of ICU stay between the two groups was not statistically significant in our study. ICU admission was not commonly required, particularly in the NPWT group, where 75% of patients did not spend any time in the ICU (Table 4). In the future, it may be recommended to further study, measure, and report on the overall cost-effectiveness of this ciNPWT on SSI prevention to encourage healthcare decision makers to liberally indicate this method.
| Outcome | ciNPWT group (n = 45) | Control group (n = 45) | p |
|---|---|---|---|
| Length of hospital stay (days) (median; min–max; IQR) | 8; 4–52; 11 | 10.5; 3–40; 8 | 0.289 |
| Length of hospital stay (days) (mean ± SD) | 13.42 ± 15.35 | 12.31 ± 7.73 | |
| Intensive care (days) (median; min–max; IQR) | 0; 0–25; 0 | 0; 0–28; 4 | 0.153 |
| Intensive care (days) (mean ± SD) | 1.96 ± 5.2 | 2.67 ± 5.62 |
4.1 Limitations of the Study
This study has a number of limitations. First, the number of enrolled patients remained significantly below expectations. This shortfall was partly attributable to the COVID-19 pandemic, which substantially hindered patient recruitment. Contributing centers did not receive financial support for participation, and many were already overburdened during the pandemic, leading to declining cooperation over time. Consequently, only 90 patients were enrolled, instead of the originally planned 300. Yet, presumably because of the inclusion of patients with a particularly high risk of SSI, the study also produced statistically significant results with the lower number of patients included.
Although both the ASA scores and patients' age differed significantly between the two arms in the study, neither was found to be an independent predictor of surgical site infection (SSI) in the multivariate logistic regression analysis. This suggests that baseline imbalances in these variables did not confound the primary outcome.
Another limitation is the lack of standardised timing for the application of ciNPWT following emergency surgery. (Surgeons were allowed to set-up NPWT dressing either at the end of the operation or latest in the morning after an overnight emergency surgery). Although this variability may introduce some confounding, it is worth noting that the delay before application in our cohort was limited to a few hours. Similarly, in a prospective study investigating NPWT in the emergency setting, the device was applied within 3 days, though the timing was not strictly regulated there either [25].
5 Conclusion
The effectiveness of NPWT in healing chronic wounds has been well established for decades. On the other hand, the use of closed incisional NPWT to prevent surgical site infections is not well understood and its indications are still controversial. The existing literature is very heterogeneous and only a small percentage of the available studies deal with general surgical patients in an emergency setting. To the best of our knowledge, this is the first multicenter randomised clinical trial in this area to demonstrate the efficacy of ciNPWT in the prevention of SSI in the emergency surgical field.
Further RCTs are needed to investigate the benefit of ciNPWT in lower SSI risk groups. Targeted cost-effectiveness studies in different healthcare settings are needed to support the routine indication of this technique in abdominal surgery.
Acknowledgements
The authors have nothing to report.
Ethics Statement
The study was conducted in accordance with the Declaration of Helsinki. The study was approved by the National Scientific and Ethical Committee and the National Institute of Pharmacy and Nutrition (OGYÉI/15347-5/2018).
Conflicts of Interest
The authors declare no conflicts of interest.
Open Research
Data Availability Statement
The data that support the findings of this study are not publicly available due to patient confidentiality and institutional restrictions, but may be available from the corresponding author upon reasonable request and subject to ethical approval.
