Factors related to successful mesh salvage with negative pressure wound therapy: a retrospective cohort study Authors: Alejandro González-Muñoz, Juan Carlos Vallejo-Soto, Juan Diego Barragán-Pinilla, Antonio Pesce, Camilo Ramírez-Giraldo Published: 2024 ABSTRACT BackgroundMesh salvage using negative pressure wound therapy (NPWT) in cases of mesh infection following hernia repair has emerged as an alternative to early mesh removal. However, the factors related to the success or failure of mesh salvage with NPWT remain unclear.MethodsThis retrospective cohort study included 61 patients with mesh infections after hernia repair treated with NPWT between 2018 and 2024. We analyzed demographic, clinical, and surgical variables, as well as the bacterial spectrum and antimicrobial susceptibility. A binary logistic regression model was used to identify factors associated with NPWT failure, defined as the need for mesh removal.ResultsMesh salvage was successful in 80.3% of cases. Active smoking was significantly associated with NPWT failure (OR = 7.82, CI 95% 1.05–64.8; p = 0.044). Other factors, such as age, body mass index, Charlson comorbidity index, mesh type, and mesh position, were not significantly related to failure. Most infections were caused by Staphylococcus aureus (24.6%) and Escherichia coli (22.9%).ConclusionsNPWT is an effective method for salvaging infected meshes, with a high success rate. Active smoking was identified as a risk factor for NPWT failure, highlighting the need for early identification of patients who may benefit from alternative approaches. Further studies are required to develop predictive models for NPWT success in mesh salvage. Introduction Currently, for the repair of both primary and incisional abdominal wall defects, the use of mesh implants is recommended as part of the surgical technique to reduce recurrence rates [1, 2]. However, meshes use can lead to certain complications, including mesh infection, visceral adhesion, enteral fistula, chronic pain, and even systemic immune disease [3, 4]. Mesh infection, in particular is a potentially devastating complication [5]. Where infection occurs, targeted antibiotic treatment should be initiated based on the identified causative microorganism, and any collection should be drained. Historically, early mesh removal was recommended, which carried risks of hernia recurrence, enterotomies, and fistula formation; with a need for mesh removal reported in up to 72.7% of cases [6, 7]. More recently, mesh salvage therapy using negative pressure wound therapy (NPWT) has been suggested, as it helps to avoid the complications associated with mesh removal [5, 8]. In infected wounds treated with NPWT, studies have shown an increased in cytokines such as IL-1β and IL-8, along with a reduction in TNF-α, an increase in neutrophil counts in the wounds, and a reduction in bacterial load [9]. There is also evidence that NPWT can significantly alter the morphology and spatial distribution of Staphylococcus aureus colonies and reduce bacterial proliferation in microcolonies within 48 h of starting therapy [10]. Despite these findings, evidence on bacterial clearance with NPWT remains limited, emphasizing the necessity of thorough wound debridement [11, 12]. Regardless of the underlying mechanism, NPWT has shown efficacy in treating mesh infections and avoiding removal, with a reported success rate of approximately 85% [8, 13]. Nevertheless, current evidence is still limited, and the factors contributing to treatment success remain unclear. Understanding these factors is crucial to identifying early on who may benefit from NPWT and, conversely, who may require prompt mesh removal. Given these considerations, the objective of this work is to explore the factors associated with successful mesh salvage using NPWT in cases of mesh infection following hernia repair. Methods Study design A retrospective observational cohort study was conducted. All patients who were admitted to the institution with surgical site infections involving mesh following hernia repair between 2018 and 2024 were identified. Our endpoints were the success rate of NPWT for mesh salvage and the factors associated with failure in mesh salvage using NPWT. All variables were collected in an anonymous database. Upon admission to the institution, patients provided written informed consent for the use of their clinical information in research. The study was reviewed and approved by our institution’s ethics committee (number CEISH-2024008). We followed the STROBE guidelines in reporting this study [14]. Patients Patients who underwent hernia repair at any anatomical location on the abdominal wall, were diagnosed with mesh infection during follow-up, and were admitted for mesh salvage using NPWT were included in the study. Upon admission, empirical antibiotic treatment was initiated, followed by targeted therapy based on culture results. Patients who underwent hernia repair surgery or began infection treatment at another institution were excluded. The following data were analyzed: patient demographics, body mass index (BMI), ASA Physical Status Classification, presence of comorbidities, Charlson comorbidity index, smoking habit, hernia type (incisional or primary), mesh position, mesh type, time from index surgery to the development of surgical site infection, NPWT dressing change intervals, isolated microorganisms with their susceptibility profiles, and length of hospital stay. Follow-up was conducted until the first postoperative visit, scheduled within 15 days of hospital discharge. However, if any complications occurred, patients were advised to visit the emergency department, either before or after their first postoperative appointment. Surgical procedure The dissection was extended to the anatomical plane containing the mesh. A culture sample of the infected surgical wound was taken. Debridement and irrigation of the wound, including the mesh, were performed with sterile saline solution until clean tissue was macroscopically evident. Subsequently, hemostasis was confirmed to be adequate, after which NPWT (3M™ V.A.C.® Granufoam™ Dressing, 3M™ V.A.C.® Drape, and 3M™ SensaT.R.A.C.™ Pad Tubing) was applied at a continuous pressure of 100–125 mmHg. Dressings were changed every 5 days or sooner if system dysfunction occurred. Therapy was discontinued once the mesh was partially integrated into the tissue and the wound was clean; in cases of persistent purulent secretion and lack of mesh integration, therapy was considered unsuccessful and subsequently discontinued (Fig. 1).Fig. 1Inguinal hernia repair with surgical site infection, the wound was irrigated, and a negative pressure system was applied in an attempt to salvage the mesh Inguinal hernia repair with surgical site infection, the wound was irrigated, and a negative pressure system was applied in an attempt to salvage the mesh When mesh salvage with NPWT was unsuccessful, the mesh was partially or completely removed, depending on its integration into the tissue [15]. Statistical analysis A description was made with demographic, clinical and surgical variables. Categorical variables were described as proportions and continuous variables as medians with their respective interquartile range (IQR). A bivariate analysis was performed with a χ2 value on categorical variables and with the Mann–Whitney test on continuous variables in order to compare differences between the variables according to whether mesh was salvaged or mesh was removed. A binary logistic regression was performed to explore the factors associated with mesh removal (failure of NPWT). A p-value of < 0.05 was considered statistically significant. All analyses were carried out using R (version 2023.12.1 + 402). Results This study included 61 patients. The success rate (mesh salvaged) of NPWT was 80.3%. Median patient age was 70.0 years (IQR 61.0–74.0) and were predominantly female (68.9%). Table 1 shows demographic, clinical and surgical characteristics according to whether mesh was salvaged or mesh was removed. No statistically significant differences were found in the bivariate analysis.Table 1Demographic, clinical and surgical characteristics according to whether mesh was salvaged or mesh was removedN (%)Mesh salvaged (n = 49)Mesh removed (n = 12)p valueAge(median)(IQR) (years)70.0 (61.0–74.0)71.0 (61.0–76.0)67.5 (58.0–72.50)0.473*Sex0.220 Female42 (68.9)36 (73.5)6 (50.0) Male19 (31.1)13 (26.5)6 (50.0)Body mass index(median)(IQR) (kg/m2)26.5 (23.8–31.0)26.5 (24.0–30.8)26.0 (23.5–32.2)0.885*ASA0.137 I10 (16.4)6 (12.2)4 (33.3) II27 (44.3)21 (42.9)6 (50.0) III20 (32.8)19 (38.8)1 (8.3) IV4(6.5)3 (6.1)1(8.3)Co-morbidity Arterial hypertension42 (68.9)36 (73.5)6 (50.0)0.220 Diabetes mellitus17 (27.9)14 (28.6)3 (25.0)1.000 Chronic obstructive pulmonary disease7 (11.5)6 (12.2)1 (8.3)1.000 Chronic kidney disease6 (9.8)5 (10.2)1 (8.3)1.000 Cardiovascular disease17 (27.9)12 (24.5)5 (41.7)0.406 Immunomodulatory therapy3 (4.9)1 (2.0)2 (16.7)0.175Charlson comorbidity index(median)(IQR)(points)2.0 (1.0–3.0)2.0 (1.0–3.0)2.0 (1.0–3.0)0.839*Current smoker7 (11.5)4 (8.2)3 (25.0)0.256Hernia type1.000 Primary20 (32.8)16 (32.7)4 (33.3) Incisional41 (67.2)33 (67.3)8 (66.7)Mesh position0.496 Onlay23 (37.7)20 (40.8)3 (25.0) Sublay38 (62.3)29 (59.2)9 (75.0)Mesh type0.730 Macroporous polypropylene55 (90.2)45 (91.8)10 (83.3) Other6 (9.8)4 (8.2)2 (16.7)Time from surgery to infection (median)(IQR) (days)20.0 (10.0–25.0)20.0 (12.0–25.0)11.0 (6.2–28.7)0.378*NPWT dressing change (median) (IQR) (#)3.0 (2.0–5.0)3.0 (2.0–4.0)3.5 (1.0–5.2)0.985*Cultures0.337 Positive47 (77.1)36 (73.5)11 (91.7) Negative14(22.9)13 (26.5)1 (8.3)Polymicrobial0.568 No47 (77.1)39 (79.6)8 (66.7) Yes14 (22.9)10 (20.4)4 (33.3)Bacterial susceptibility0.343 Susceptible16 (26.2)13 (26.5)3 (25.0) Resistant31 (50.8)23 (46.9)8 (66.7)Staphylococcus aureus infection1.000 No46 (75.4)37 (75.5)9 (75.0) Yes15 (24.6)12 (24.5)3 (25.0)Hospital stay (median)(IQR) (days)14.0 (8.0–23.0)12.0 (7.0–20.0)18.5 (12.0–32.0)0.182*p values were obtained using the chi-squared test*p values were obtained using the Mann–Whitney test Table 1Demographic, clinical and surgical characteristics according to whether mesh was salvaged or mesh was removedN (%)Mesh salvaged (n = 49)Mesh removed (n = 12)p valueAge(median)(IQR) (years)70.0 (61.0–74.0)71.0 (61.0–76.0)67.5 (58.0–72.50)0.473*Sex0.220 Female42 (68.9)36 (73.5)6 (50.0) Male19 (31.1)13 (26.5)6 (50.0)Body mass index(median)(IQR) (kg/m2)26.5 (23.8–31.0)26.5 (24.0–30.8)26.0 (23.5–32.2)0.885*ASA0.137 I10 (16.4)6 (12.2)4 (33.3) II27 (44.3)21 (42.9)6 (50.0) III20 (32.8)19 (38.8)1 (8.3) IV4(6.5)3 (6.1)1(8.3)Co-morbidity Arterial hypertension42 (68.9)36 (73.5)6 (50.0)0.220 Diabetes mellitus17 (27.9)14 (28.6)3 (25.0)1.000 Chronic obstructive pulmonary disease7 (11.5)6 (12.2)1 (8.3)1.000 Chronic kidney disease6 (9.8)5 (10.2)1 (8.3)1.000 Cardiovascular disease17 (27.9)12 (24.5)5 (41.7)0.406 Immunomodulatory therapy3 (4.9)1 (2.0)2 (16.7)0.175Charlson comorbidity index(median)(IQR)(points)2.0 (1.0–3.0)2.0 (1.0–3.0)2.0 (1.0–3.0)0.839*Current smoker7 (11.5)4 (8.2)3 (25.0)0.256Hernia type1.000 Primary20 (32.8)16 (32.7)4 (33.3) Incisional41 (67.2)33 (67.3)8 (66.7)Mesh position0.496 Onlay23 (37.7)20 (40.8)3 (25.0) Sublay38 (62.3)29 (59.2)9 (75.0)Mesh type0.730 Macroporous polypropylene55 (90.2)45 (91.8)10 (83.3) Other6 (9.8)4 (8.2)2 (16.7)Time from surgery to infection (median)(IQR) (days)20.0 (10.0–25.0)20.0 (12.0–25.0)11.0 (6.2–28.7)0.378*NPWT dressing change (median) (IQR) (#)3.0 (2.0–5.0)3.0 (2.0–4.0)3.5 (1.0–5.2)0.985*Cultures0.337 Positive47 (77.1)36 (73.5)11 (91.7) Negative14(22.9)13 (26.5)1 (8.3)Polymicrobial0.568 No47 (77.1)39 (79.6)8 (66.7) Yes14 (22.9)10 (20.4)4 (33.3)Bacterial susceptibility0.343 Susceptible16 (26.2)13 (26.5)3 (25.0) Resistant31 (50.8)23 (46.9)8 (66.7)Staphylococcus aureus infection1.000 No46 (75.4)37 (75.5)9 (75.0) Yes15 (24.6)12 (24.5)3 (25.0)Hospital stay (median)(IQR) (days)14.0 (8.0–23.0)12.0 (7.0–20.0)18.5 (12.0–32.0)0.182*p values were obtained using the chi-squared test*p values were obtained using the Mann–Whitney test Demographic, clinical and surgical characteristics according to whether mesh was salvaged or mesh was removed Charlson comorbidity index (median)(IQR)(points) p values were obtained using the chi-squared test *p values were obtained using the Mann–Whitney test A higher proportion of patients had incisional hernias (67.2%), while the remaining cases were primary hernias, including inguinal, umbilical, or epigastric hernias. The mesh location in most cases was described as sublay, which includes retromuscular, preperitoneal, retro-oblique, interoblique, among others. In no case was the mesh located intraperitoneally or inlay [16]. The type of mesh used was mostly macroporous polypropylene (90.2%); in three cases (4.9%), it was microporous polypropylene, in two cases (3.3%), it was multifilament polyester, and in one case (1.6%), it was polypropylene/ePTFE. The time from the index surgery (hernia repair) to the presentation of the surgical site infection was 20.0 days (IQR 10.0–25.0). A median of 3.0 (IQR 2.0–5.0) wound debridements and NPWT changes were required, with no statistically significant differences between those whose mesh was salvaged and those whose mesh was removed (p = 0.985). Fourteen of the cultures were negative for microbiological isolation, and in 14 cultures, more than one microorganism grew. The most frequently isolated bacteria from the surgical wound cultures was Staphylococcus aureus (24.6% of cases), of which three cases were methicillin-resistant and none were resistant to vancomycin. The next most frequently isolated microorganism was Escherichia coli, present in 22.9% of cases, of which three were extended-spectrum beta-lactamase producers and three were carbapenemase producers. Proteus mirabilis was isolated in 11 cases (18.0%), of which 3 were carbapenemase producers. Other isolated microorganisms included Klebsiella pneumoniae (6 cases), Enterobacter cloacae (6 cases), Enterococcus spp. (6 cases), Morganella morganii (2 cases), Serratia marcescens (1 case), Pseudomonas aeruginosa (1 case), Staphylococcus epidermidis (1 case), Streptococcus spp. (2 cases), Cutibacteriumavidum (1 case), Aeromonas hydrophila (1 case), and Citrobacter amalonaticus (1 case) (Fig. 2).Fig. 2Bacterial spectrum Bacterial spectrum We performed a binary logistic regression model to assess the factors related to NPWT failure. The model showed that being an active smoker is a risk factor for NPWT failure in salvaging the mesh and avoiding its removal in cases of infection (OR = 7.82 CI 95% 1.05–64.8); the other variables included in the model did not show this relationship (Table 2).Table 2Binary logistic regression model to identify factors related to NPWT failureOR (CI 95%)p valueAge0.95 (0.88–1.01)0.146Body mass index0.97 (0.85–1.12)0.700Charlson comorbidity index1.14 (0.73–1.77)0.567Current smoker7.82 (1.05–64.80)0.044Macroporous polypropylene0.31 (0.03–3.15)0.290Sublay mesh location2.58 (0.53–15.60)0.260Primary hernia1.05 (0.18–6.02)0.958Culture positive2.31 (0.29–49.20)0.482Polymicrobial3.62 (0.59–27.30)0.176Bold values indicate statistically significant p values (p < 0.05) Table 2Binary logistic regression model to identify factors related to NPWT failureOR (CI 95%)p valueAge0.95 (0.88–1.01)0.146Body mass index0.97 (0.85–1.12)0.700Charlson comorbidity index1.14 (0.73–1.77)0.567Current smoker7.82 (1.05–64.80)0.044Macroporous polypropylene0.31 (0.03–3.15)0.290Sublay mesh location2.58 (0.53–15.60)0.260Primary hernia1.05 (0.18–6.02)0.958Culture positive2.31 (0.29–49.20)0.482Polymicrobial3.62 (0.59–27.30)0.176Bold values indicate statistically significant p values (p < 0.05) Binary logistic regression model to identify factors related to NPWT failure Bold values indicate statistically significant p values (p < 0.05) Discussion Our study demonstrated an 80.3%, success rate with NPWT for mesh salvage, consistent with findings in the literature. In the systematic review by Bueno-Lledó et al. [8] a success rate of 85.2% was reported, while Junsheng Li et al. [3] reported a success rate of 76.2%. We found that active smoking was associated with NPWT failure in mesh salvage cases. This may be due to tobacco’s well-documented negative impact on wound healing through several mechanisms. In addition to ensuring a clean surgical field, NPWT is expected to promote granulation tissue formation, facilitating mesh integration with surrounding tissue. However, tobacco reduces fibroblast proliferation and induces vasoconstriction, which can lead to local tissue ischemia [17–19]. The macroporous polypropylene mesh type has been associated with a higher success rate for NPWT mesh salvage, likely due to its improved tissue integration [20]. In our study, we found a success rate of 81.8% with macroporous polypropylene meshes compared to 66.6% with other mesh types, with no statistically significant differences. Notably, most meshes used in our cohort (90.2%) were macroporous polypropylene. On the other hand, another factor described in the success of NPWT for salvage is the position of the mesh. It has been found that extraperitoneal mesh locations have a higher success rate for salvage with NPWT [8]. In this study, all meshes were placed extraperitoneally, although their sublay positions (retromuscular, preperitoneal, interoblique, or other) were not further categorized. No significant differences were found in success rates between onlay and sublay placements. In our study, Enterobacteriaceae were the most prevalent bacteria identified, similar to findings in other studies, although with a lower rate of polymicrobial isolations, which reached 22.9% [21]. Additionally, Staphylococcus aureus was isolated in 24.6% of cases; this microorganism is commonly reported as a leading causative agent and has been associated with higher failure rates for mesh salvage [21]. In our study, however, infection with Staphylococcus aureus did not affect the NPWT success rate for mesh salvage. NPWT is a very highly valuable therapy and should be the treatment of choice for patients with mesh infection following hernia repair to promote mesh salvage and avoid complications related to mesh removal. It is essential to develop predictive models for NPWT salvage failure to guide early mesh removal decisions and prevent unnecessary NPWT treatment, which incurs additional costs due to required supplies, dressing changes, wound debridement, and prolonged hospital stays. Regarding the limitations of this study, we acknowledge the retrospective design, small sample size, and limited follow-up. Limited medium- and long-term follow-up data in the NPWT group are particularly notable, as monitoring hernia recurrence rates is crucial for a comprehensive assessment of long-term outcomes in this group. The results with this small sample should be interpreted with caution, and it is essential that the results are validated in larger studies. Conclusions NPWT is an effective method for salvaging infected meshes, with a high success rate. Active smoking was identified as a risk factor for NPWT failure, highlighting the need for early identification of patients who may benefit from alternative approaches. Further studies are required to develop predictive models for NPWT success in mesh salvage.