Synthetic versus Biologic Mesh in Single-Stage Repair of Complex Abdominal Wall Defects in a Contaminated Field

Abstract

Background:

Synthetic meshes have been used with varying rates of success in a contaminated setting, although their use is not widely accepted because of concerns for infection. A biologic mesh (BM) is assumed to be more resistant to infection than a synthetic mesh; however, sparse clinical data support this theory. The hypothesis for this study: Uncoated polypropylene synthetic mesh (USM) can be used to obtain a durable repair in the setting of a contaminated abdominal wall reconstruction (AWR) in a single-stage procedure with comparable infectious outcomes to a biologic mesh repair.

Patients and Methods:

We performed a retrospective chart review on contaminated AWR, comparing 34 BM with 24 USM with infection as the primary outcome of interest. Secondary outcomes were re-admission and re-operation. We also investigated the microbial isolates that were cultured.

Results:

Mesh groups were similar in their demographics, duration of surgery, previous mesh, surgical site class, and source of contamination. Length of stay was 4 d longer in BM, p = 0.01. Overall infection rate was 50% for BM vs. 29.2% for USM, p = 0.18. Treatment in case of infection was similar across both groups. Gram positive bacteria comprised 39% of BM microbiology vs. 63% for USM. Re-admission rate was 52.9% for BM versus 45.8% for USM, p > 0.5. The BMs re-admitted for surgical site infection/abdominal abscess were 38.9% versus 55.6% for USM. No USM were re-admitted for seroma versus 33.3% of BM, p = 0.06. Re-operation rate was 26.5% for BM versus 33.3% for USM, p > 0.5. Procedures performed at re-operation were similar between groups. Regression analysis did not demonstrate an association between mesh type and our outcomes.

Conclusion:

Our results show that synthetic meshes are not inferior to biologic meshes in contaminated AWR. This is important in view of the tremendous cost disparity between these two products and the questionable ability of biologic mesh to offer a durable hernia repair.

Large abdominal wall hernias are a difficult clinical problem fraught with complications and have a high rate of recurrence. When a source of infection is added to this already difficult clinical situation, the optimal treatment becomes very controversial. Some argue for a single-stage repair, where the hernia is repaired and the source of contamination is removed in one surgical procedure. A mesh is used in the vast majority of these cases to reinforce the abdominal wall because it has been shown to lower recurrence rates from 54% to 16% [1 –3]. The challenge of managing contaminated ventral hernias is that synthetic mesh materials are perceived as contraindicated [4 –6]. As a result, some might argue for a multi-staged reconstructive approach with delayed definitive repair. The prolonged convalescence and the complexity of the large abdominal wall defect created have prompted surgeons to investigate options for a single-staged approach to repairing contaminated abdominal wall hernias.

Biologic grafts reportedly promote cellular infiltration, neovascularization, and potentially regenerate into native tissue that might provide substantial advantages over synthetic materials in the setting of contamination [7,8]. Theoretically, they are more resistant to infection compared to synthetic mesh, although limited data support this theory [9]. Synthetic meshes have been used with varying rates of success in a contaminated setting, although their use is not widely accepted because of concerns for infection. Further, their use in clean-contaminated and contaminated ventral hernias is associated with a recurrence rate as low at 7% [5].

Few studies have compared uncoated synthetic mesh (USM) with biologic mesh (BM) in an unclean abdominal wall repair [8,10 –12]. We performed a retrospective chart review on contaminated complex incisional hernia repairs, comparing BMs with USMs with infection as the primary outcome of interest. We hypothesize that polypropylene USM can be used to obtain a durable repair in the setting of a contaminated abdominal wall hernia in a single-stage procedure with comparable infectious outcomes to a BM repair.

Patients and Methods

After acquiring Institutional Review Board approval, we conducted a chart review of patients undergoing ventral hernia repair at our institution. We queried the operative records for the keywords ventral hernia, incisional hernia, and abdominal wall reconstruction from January 1, 2013 through November 30, 2015. Inclusion criteria were defined as a midline hernia of at least 100 cm² or multiple defects that added up to a 100 cm², necessitating a form of component separation, and an identified source of contamination (gastrointestinal operation, infected mesh, fistula, stoma, etc.). Surgical site class was either clean-contaminated, contaminated, or dirty as defined by the U.S. Center for Disease Control and Prevention (CDC) surgical site class. Hernia grade was determined according to the Ventral Hernia Working Group (VHWG) hernia grade classification, whereby they were either grade III or IV based on our inclusion criteria.

Patients were excluded if follow-up was less than 5 months, if they were in a clinical trial for hernia repair, if the mesh was bridging, the abdomen was left open, or if we identified a false result from the keyword search. Patients who died during their initial hospitalization were excluded as well. We screened 608 patients, of whom 69 met these criteria.

We captured a number of variables: Age, gender, body mass index (BMI), American Society of Anesthesiologists (ASA) class, hypertension, diabetes mellitus, smokers, chronic obstructive pulmonary disease (COPD), operation date and duration, type of mesh used and its location, previous mesh, reason for contamination, surgical site classification, length of stay, and days until re-admission, re-operation, and infection if any. Mesh infection was defined as any treatment with oral or intravenous antibiotics with or without culture data. Re-admission diagnosis and procedure at re-operation were recorded, as well as the most invasive intervention in case of infection and culture data if present. Smokers were defined as having smoked within the past year.

We compared the BM with the USM for the demographics and outcomes listed above. Pearson chi square test, Fisher exact test, and Mann–Whitney U test were performed where appropriate. A binary logistic regression model was used to assess whether age, gender, BMI, operative duration, surgical site class, smoking, diabetes mellitus, hospital length of stay, whether a previous mesh was present, and mesh type and location were associated with infections, re-admissions, and re-operations. Non-inferiority testing was performed to show whether the USM were at least not worse than BM in resisting infection. The level of significance was set at 5%.

Results

Our patient population was distributed mainly between BM and USM, 34 and 24, respectively. Of the 34 BM, nine were human dermis (FlexHD,^® Musculoskeletal Transplant Foundation, Edison, NJ), two were porcine small intestinal submucosa (Biodesign,^® Cook Medical, Bloomington, IN), and the rest were porcine dermis (two XenMatrix,™ [Bard, Warwick, RI], two XCM [DePuy Synthes, West Chester, PA], two Fortiva,^® [RTI Surgical, Alachua, FL], 15 Strattice,™ [Acelity, San Antonio, TX], and two not recorded). Of the USM, 19 were lightweight polypropylene and five were heavyweight polypropylene. When comparing BM with USM, the two groups were similar with regard to age, gender, BMI, smoking, hypertension, diabetes mellitus, and COPD. Patients were generally at a similar medical status as evidenced by their ASA class, because the mode was 3 for both groups.

Clean-contaminated cases comprised 43.7% of our surgical procedures, contaminated were 35.6%, and dirty/infected accounted for 20.7%. Sources of contamination are presented in Table 1, along with mean hospital length of stay and operative duration, previous mesh, surgical site class, reason for contamination, and mesh placement location.

Table 1.

Group Characteristics with Regard to Surgical Factors

	Biologic (34)	Uncoated synthetic (24)	p
Length of stay (d)	11.09 (SEM: 2.2)	7 (SEM: 1)	0.01
Duration of operation (h)	5.48 (SEM: 0.26)	5.66 (SEM: 0.48)	>0.5
Previous mesh	15–44.1%	11–45.8%	>0.5
Surgical site class			>0.5
• Clean-contaminated	17–50%	10–41.7%	-
• Contaminated	10–29.4%	8–33.3%	-
• Dirty	7–20.6%	6–25%	-
Contamination			>0.7
• GI surgery	16–47.1%	9–37.5%	-
• Infected mesh	7–20.6%	6–25%	-
• Fistula	6–17.6%	6–25%	-
• Stoma	3–8.8%	1–4.2%	-
• Other	2–5.9%	2–8.3%	-
Mesh location			<0.001
• IPOM	18–53%	0%	<0.001
• Onlay	6–17.6%	3–12.5%	>0.5
• Retrorectus	10–29.4%	21–87.5%	<0.001

SEM = standard error of the mean; GI = gastrointestinal; IPOM = intraperitoneal onlay mesh.

Hernia grade was similar between groups, with BM being 21% (7/34) grade IV versus 29% (7/24) for USM, p = 0.54. Median follow-up was 11.3 mo for BM versus 7.9 mo for USM, p = 0.18. Length of stay was different between groups; patients who had a BM implanted spent 4 extra days on our wards. Mesh location was significantly different between groups: 87.5% of USM were placed in the retrorectus space versus 29.4% for BM; none of the USM were intraperitoneal onlay mesh (IPOM) repairs whereas 53% of BM were IPOM repairs, p < 0.001.

The BM observed an infection rate of 50% versus 29.2% for USM, p = 0.18 (Table 2A). When the meshes became infected, most occurred within 30 d (65% of BM infections vs. 57% of USM infections). The 30-day overall infection rates were 32.3% for BM and 16.7% for USM, p > 0.5. Treatment for patients with infection was not significantly different between mesh groups. Subsequent surgical procedure was required for 42.9% of patients with infections who received USM versus 29.4% for BM, p > 0.5. Re-admission rates were 52.9% for BM and 45.8% for USM, p > 0.5.

Table 2.

Outcome Comparison between Groups

A. Infections	Biologic (34)	Uncoated synthetic (24)	p
Infection rate	17–50%	7–29.2%	0.18
30-d infection	11–32.3%	4–16.7%	>0.5
Mean time (d)	50.9 (SEM: 14.2)	34 (SEM: 9.4)	>0.5
Treatment			>0.5
• PO antibiotics	3–17.6%	2–28.6%	-
• IV antibiotics	2–11.8%	1–14.3%	-
• IR drain	7–41.2%	1–14.3%	-
• Surgical procedure	5–29.4%	3–42.9%	-

B. Re-admissions	Biologic (34)	Uncoated synthetic (24)	p
Re-admission rate	18–52.9%	11–45.8%	>0.5
Mesh-related re-admission	16–88.9%	10–90.9%	>0.5
30-d re-admission	8–42.1%	6–54.5%	>0.5
Mean time (d)	115.9 (SEM: 49.9)	36.4 (SEM: 9.4)	0.47
Re-admission diagnoses^*
• Seroma	6–33.3%	0%	0.06
• Surgical site infection	4–22.2%	5–55.6%	0.24
• Abdominal abscess	3–16.7%	0%	0.27
• Bowel obstruction	1–5.6%	1–9.1%	>0.5
• EC fistula	1–5.6%	0%	>0.5
• Nausea/vomiting	0%	3–33.3%	0.05
• Other	3–16.7%	1–9.1%	>0.7

C. Re-operations	Biologic (34)	Uncoated synthetic (24)	p
Re-operation rate	9–26.5%	8–33.3%	>0.5
30-d re-operation	3–8.8%	0	0.21
90-d re-operation	3–8.8%	4–16.7%	0.64
Mean time (d)	290 (SEM: 102)	142 (SEM: 34)	>0.5
Procedures at re-operation			0.66
• Wound debridement	4–44.4%	3–37.5%	-
• Hernia repair	2–22.2%	3–37.5%	-
• Bowel obstruction	1–11.1%	1–12.5%	-
• EC fistula takedown	1–11.1%	0%	-
• Excision of infected mesh	1–11.1%	0%	-
• Seroma evacuation	0%	1–12.5%	-

Other diagnoses were pneumonia, ileus, bladder obstruction, and colitis with anastomotic leak.

EC = enterocutaneous; SEM = standard error of the mean; PO = by mouth; IV = intravenous; IR = interventional radiology.

Re-admission diagnoses for the USM group were mainly surgical site infection (SSI) (55.6%) and nausea and vomiting (33.3%), whereas the BM group had a more evenly distributed array of re-admission diagnoses (Table 2B). Rates of re-operation and the procedures performed were similar between groups (Table 2C). Re-admission and re-operation rates at 30 d post-operative were similar between mesh groups. Further, 90 d re-operation between groups was also noted to be similar. Distribution of infection, mesh-related re-admission, and re-operation according to surgical site class are detailed in Table 3.

Table 3.

Outcomes by Surgical Site Class

	Infection	Mesh-related re-admission	Re-operation
Clean-contaminated	11–45.8%	12–46.1%	8–47.1%
Contaminated	11–45.8%	10–38.5%	2–11.8%
Infected	2–8.4%	4–15.4%	7–41.2%
Total	24	21	15

Column percentages are shown.

We retrieved culture data on 71% of patients in whom an infection developed (17/24), and their microbiology is presented in Table 4. Of the total microorganisms cultured, gram positive bacteria comprised 39% (11/28) of BM microbiology versus 63% (5/8) for USM. Similarly, gram negative bacteria totaled 50% for BM versus 25% for USM. With regard to surgical site class, gram positive bacteria grew in 40% (10/25) of clean-contaminated surgical procedures versus 44% (4/9) of contaminated ones and 100% of dirty/infected cases (2/2). Gram negative bacteria were present in 52% of clean-contaminated surgical sites versus 33% of contaminated incisions.

Table 4.

Microorganisms Cultured from Those in Whom Infectious Complications Developed

A. Distribution by mesh type	Biologic (12)	Uncoated synthetic (5)
Enterococcus	4–14.3%	1–12.5%
Bacteroides	4–14.3%	0
MSSA	3–10.7%	3–37.5%
MRSA	3–10.7%	0
Proteus	3–10.7%	0
Yeast	3–10.7%	1–12.5%
Pseudomonas	2–7.1%	1–12.5%
E. coli	2–7.1%	0
Morganella	1–3.6%	0
Citrobacter	1–3.6%	0
Klebsiella	1–3.6%	0
Actinomyces	1–3.6%	0
Enterobacter	0	1–12.5%
GPC NOS	0	1–12.5%
Total microorganisms	28	8
Total gram +	11/28–39%	5/8–63%
Total gram −	14/28–50%	2/8–25%

B. Distribution by surgical site class	Clean-contaminated	Contaminated	Dirty/infected
Enterococcus	4	1	0
Bacteroides	3	1	0
MSSA	3	1	2
MRSA	2	1	0
Proteus	3	0	0
Yeast	2	2	0
Pseudomonas	2	1	0
E. coli	2	0	0
Morganella	1	0	0
Citrobacter	1	0	0
Klebsiella	0	1	0
Actinomyces	1	0	0
Enterobacter	1	0	0
GPC NOS	0	1	0
Total microorganisms	25	9	2
Total gram +	10/25–40%	4/9–44%	2/2–100%
Total gram −	13/25–52%	3/9–33%	0

MSSA = methicillin-sensitive Staphylococcus aureus; MRSA = methicillin-resistant S. aureus; E. coli = Escherichia coli; GPC NOS = gram-positive cocci not otherwise specified.

Percentage of total isolates per group.

Logistic regression analysis did not demonstrate any association between mesh type (biologic or uncoated synthetic) and the development of infection, re-admission, or re-operation with p values of 0.73, 0.68, and 0.47, respectively. Mesh location (IPOM vs. onlay vs. retrorectus), surgical site class (clean-contaminated vs. contaminated vs. dirty), and all other remaining factors in the model demonstrated no significant associations.

Primus and Harris [13] reviewed the performance of biologic mesh in a contaminated setting and found that SSI ranged from zero to 60%, with a follow-up of 6–18 mo. Ferzoco [14] found infection rates of 20%–75% for unclean surgeries in which a biologic mesh was used. Assuming BM as an existing effective treatment with an infection rate of 50%, the non-inferiority margin was specified as an absolute difference of 15%, which corresponds to the margin odds ratio of 1.86. To have 80% power with a 15% error margin, 24 patients were needed in each group. The odds ratio for infection in USM compared with BM was 0.41 (95% confidence interval, 0.14–1.25), which is well below the non-inferiority margin of 1.86, establishing the non-inferiority of USM to BM in resisting infection.

Discussion

We compared the performance of BM with USM in complex incisional hernia repairs in which the surgical site class was clean-contaminated, contaminated, or dirty/infected. Our main outcome of interest was the development of infection; however, we also looked at rates of re-admission and re-operation. We found that the choice of mesh, whether uncoated synthetic or biologic, did not affect our outcomes even when adjusting for confounders in a regression model. These findings challenge the general doctrine that BM is preferred in the setting of contamination.

Patient follow-up ranged from 5 mo to 2.5 y, and although we included patients from January 2013, our practice did not start implanting USM in the retrorectus position until halfway through that year. The biologic grafts used in our practice were mostly human acellular dermal matrix (FlexHD) (46%) and porcine acellular dermal matrix (Strattice) (26%). In the USM group, 19 were lightweight polypropylene and five were heavyweight polypropylene.

Patients fitted with a BM had a longer length of stay by up to 4 d, which may be attributed to selection bias. Surgeons may be more likely to place a BM in patients with more co-morbidities, although our data suggests that the two groups were similar with regard to co-morbidities. Another possible explanation is that the difference in length of stay could be attributed to the mesh location. The BMs were placed mostly in the IPOM location and had direct contact with the abdominal viscera, which could promote ileus and in turn longer lengths of stay.

Mesh placement was different between BM and USM because of their inherent properties. The USMs usually are not placed intraperitoneally because of the risk of adherence and erosion into the bowel and are placed preferentially in the retrorectus space. The BMs on the other hand are believed to be safe in the intraperitoneal location and are implanted largely there given the ease of placement in this location.

Infection: BM

The infection rate for patients who were treated with a BM was 50% overall (17/34). Of the 17 BMs that became infected, 64.7% (11/17) occurred within 30 d of operation. Most studies addressing BM were conducted either in a clean setting or had no mention of surgical site class. Primus and Harris [13] reviewed the performance of BM in a contaminated setting. The SSI ranged from zero to 60%, with recurrences in zero to 50% of patients with a follow-up of 6–18 mo. All studies included in this review conclude that BM is indicated under those circumstances; however, Primus and Harris [13] contested that their conclusions are wrong. They point out inconsistencies in the data and caution drawing solid conclusions and recommendations based on poor evidence.

In a systematic review, Ferzoco [14] found infection rates of 20%–75% for unclean operations in which a BM was used, with a zero to 50% rate of recurrence. With mesh explantation rates as low as zero to 23%, he claims that it is justifiable to implant a BM in a contaminated setting, as it can be salvaged in case of infection. In another systematic review by Slater et al. [15], the infection rate for contaminated and dirty wounds was 19.3% and recurrence was 23.1%. Although these authors found these rates to be high, BM was most often salvageable in case of infection, and mesh explantation rate was 4.9%. Itani et al. [16] studied the performance of Strattice in a single-stage repair of clean-contaminated and contaminated hernias in 85 patients. Their reported rate of surgical site occurrence (SSO) was 60%, SSI was 30%, and recurrence at 2 y follow-up was 28%, with an overall success rate of 70%.

Infection: USM

Infection rates for USM in our population was 29.2% overall (7/24) with 57.1% of them (4/7) occurring within 30 d of operation. Because the Ventral Hernia Working Group (VHWG) recommends against implanting a synthetic mesh in patients who are at high risk of infection and the general tendency is in accordance with these recommendations, not many studies have examined the use of synthetic mesh in unclean fields. [4,17].

In 100 patients undergoing retrorectus repair with a synthetic mesh in unclean wounds, Carbonell et al. [5] found SSO to be 26.2% in clean-contaminated and 34% in contaminated fields. The SSI at 30 d was 7.1% and 19%, respectively, with 7% recurrence regardless of surgical site class. Overall, superficial SSI was 25.8%, which is similar to the rate we found in our study. Another study by Souza and Dumanian [18] on 87 polypropylene mesh placements (50% grade 2, 28% grade 3) supported their use in grades 1, 2, and 3 hernias. With no comment on infection rate, they suggest a recurrence of 5.6%. Two other studies [19,20] support the use of synthetic mesh in an unclean setting; however, they were for abdominal wall reinforcement in the absence of a large hernia.

Infection: BM versus USM

A direct comparison of mesh outcomes is warranted in each clinical setting to determine which type is preferred. In a clean field, one study [21] showed a SSI rate of 20% for BM versus 35% for USM used for incisional hernia repairs. The rate of seroma formation was 32.5% versus 15%, respectively, but without statistical significance. The study by Fischer et al. [10] compared 45 patients with a polypropylene mesh in the retrorectus space to 27 patients with a biologic underlay repair in grade 2 hernias, showing that they have similar rates of infection, re-admission, and re-operations. A recent meta-analysis by Darehzereshki et al. [8] showed a SSI rate of 10.9% in biologic meshes versus 36.5% in synthetics with p < 0.00001. Eight articles compared biologic with non-biologic meshes totaling 889 patients, 328 with BM and 561 with non-biologic meshes. They did not, however, comment on surgical site class. They also state that there was a significant amount of variation in defining outcomes among studies making it difficult to draw solid conclusions.

The only mention of BM versus USM placement in a single-stage repair of contaminated open incisional hernias in a single study was in the comment by Brahmbhatt et al. [9] on the article by Carbonell et al. [5]. Brahmbhatt cited his own experience, not revealing infection rates but rather mesh explantation and hernia recurrence [9]. Because he directly compared synthetics with biologics in a contaminated setting, the latter had zero occurrences of either whereas synthetics had 5.1% and 7.7%, respectively. The authors questioned the ability of a lightweight polypropylene mesh to tolerate bacterial load and support a lasting hernia repair.

Infection: Choice of mesh

Our current study provides new evidence on incisional hernia repair in unclean conditions. Taking into account patient demographics and surgical factors, the choice of mesh in our population did not appear to impact rates of infection. We believe the reason is the inherent properties of a macroporous lightweight polypropylene mesh. Studies have demonstrated effective clearance of bacteria and more favorable outcomes in previously contraindicated settings [5]. The large-pore synthetic materials seem to perform better than thick non-porous biologic implants, with better tissue integration when placed deep to the well-vascularized rectus muscle, and with similar rates of bacterial clearance [18,22,23]. Patients report higher satisfaction and less chronic pain and foreign body sensation with large-pore synthetics [24].

In case of infection, the literature cites that BMs are better salvaged (avoid explantation) than synthetics [7,9,14,15,25]. In our patient population, in case of infection, 42.9% of synthetics required re-operation versus 29.4% of biologics, p > 0.5. Because we lacked power in this aspect, conclusions cannot be drawn. The ability to salvage lightweight polypropylene mesh, however, has been demonstrated with negative pressure therapy; all retromuscular meshes were completely salvaged after a median of five dressing changes in one study, and 11 of 12 patients were treated successfully in another [26,27].

Of the 24 patients in whom an infection developed, culture data were available for 17 (Table 4A). The BMs grew gram positive micro-organisms in 40% of total isolates compared with gram negatives at 52%. Synthetic meshes on the other hand grew 63% gram positive bacteria as opposed to 25% gram negatives. One would expect gram negative micro-organisms to be cultured in contaminated cases, reflecting gastrointestinal flora.

Because their presence in BM was double that of USM, one could suggest that the inherent properties of these mesh types makes them susceptible to different micro-organisms. With synthetic meshes growing gram positive bacteria in 63% of isolates, one could argue that their biggest threat comes from the outside world rather than the gastrointestinal tract, making them ideal in the setting of contaminated surgical fields. This could potentially relate to the polymicrobial nature of BM cultures at 58% compared with only 20% in USM, p = 0.29. Evidence is needed to support this claim, however, because we only had culture data on eight infections in the USM group and 17 in the BM group.

Re-admissions, re-operations, recurrence

Re-admission rates related to mesh complications were 88.9% for BMs compared with 90.9% in the synthetic group (p > 0.5). Roughly just more than half of overall re-admissions within each mesh group occurred within the first 30 d, where infections are expected to be the most common. This was evidenced by our data, because 57%–65% of infections occurred within the initial 30 d. In one study, re-admission rates in a clean setting were 37.5% for BM versus 45% for USM and 29.6% versus 24.4% in another, respectively [10,21].

These studies also describe re-operation rates at 35% for both in one and 3.7% versus 8.9% in the second. These differences were not significant, and they cannot be related to our study because of surgical site class discrepancy. We noted, however, that 50% of re-operations in USM cases are within 90 d of operation compared with 33.3% for BMs. These data suggest that if a synthetic mesh is bound to fail and require a surgical procedure, it will likely fail early within 3 mo. Our study, however, lacks adequate long-term follow-up to capture the late sequelae of both BM and USM.

Our chart review goes as far back as January 2013, a time where our hernia clinic was not yet established. Few patients have more than 2 y of follow-up. Nevertheless, five patients (two BM and three USM) were re-operated on for hernia repair between 7 mo and 20 mo from the index operation. The source of contamination at initial surgery was an infected mesh in four of them; however, none of the re-operations were because of infection.

Recurrence-free survival for single-stage repair of contaminated incisional hernia with BM was 92% at 1 y and 77% at 2 y [28]. Because our patient cohort is larger in the last year of the study with more than half of our population operated on in 2015 and only 25% in 2013, we lack sufficient follow-up to actually compare recurrence rates between USM and BM. With the data that we have captured, however, recurrence for BM was 5.9% compared with 12.5% for USM. Other studies cite recurrence rates ranging from zero to 50% in any surgical site class (5,6,8–10,12–16,18,21].

Costs

A fundamental difference between biologic and synthetic meshes other than material is their financial burden. The price of BM ranges between $8.60 and $22.00 per cm² with the average mesh used for abdominal wall reconstruction costing more than $10,000 compared with most USMs costing less than $300 [29]. Reynolds et al. [30] performed a cost data analysis on 415 patients undergoing open ventral hernia repair showing that synthetic meshes have an average net profit of $60, whereas BM present a negative median contribution margin at $4560 with a net loss of $8370 per case.

This underscores the importance of patient selection for the cost-effective use of BMs. The surgeon is often faced with assessing risk-benefit ratio while taking into account the increased cost of biologic grafts, the options consisting of a single-stage repair with a BM, a single-stage repair with a synthetic mesh, or a multi-stage procedure with control of contamination first. Decreased morbidity would justify an increased cost; however, there is a need for data to inform appropriate patient selection.

Limitations

As in any retrospective chart review, our study is subject to selection bias, incomplete medical records, and inconsistent data reporting. It is possible that selection bias allowed for the use of synthetic mesh in more straightforward cases. Our patient characteristics, however, were similar in the two groups, as were the indications for operation and surgical site distribution. Further, our procedures were conducted by a group of general and reconstructive surgeons whose practices may vary. This lack of uniformity in procedures decreases our internal validity.

In addition, very few patients have more than 2 y of follow-up because our study period begins before our hernia clinic was established. This could lead to less infections to be detected, particularly in the synthetic mesh group and their late infections. We had a few instances of nearly significant results, increasing our chances of having a type B error; we believe a larger cohort of patients is needed to further solidify our conclusions and draw new ones.

Conclusion

Our study stands out as the first to compare BM with USM in a single-stage repair in a contaminated surgical field. The results challenge the VHWG recommendations on incisional hernia repair in grades 3 and 4 hernias that recommend against the use of synthetic meshes in surgical sites where a risk for infection exists. Our results show that USM are not inferior to BM in contaminated incisional hernia repairs. This is important in view of the tremendous cost disparity between the two types of products and the questionable ability of BM to offer a durable hernia repair. Our study sets the ground for a randomized controlled trial to evaluate the performance of USM versus BM in a contaminated setting.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Luijendijk

, Hop

, van den Tol

, et al. A comparison of suture repair with mesh repair for incisional hernia. N Engl J Med, 2000; 343:392–398.

Burger

, Luijendijk

, Hop

, et al. Long-term follow-up of a randomized controlled trial of suture versus mesh repair of incisional hernia. Ann Surg, 2004; 240:578–583.

den Hartog

, Dur

, Tuinebreijer

, Kreis

. Open surgical procedures for incisional hernias. Cochrane Database Syst Rev, 2008:CD006438.

Ventral Hernia Working Group, Breuing

, Butler

, et al. Incisional ventral hernias: Review of the literature and recommendations regarding the grading and technique of repair. Surgery, 2010; 148:544–558.

Carbonell

, Criss

, Cobb

, et al. Outcomes of synthetic mesh in contaminated ventral hernia repairs. J Am Coll Surg, 2013; 217:991–998.

Krpata

, Blatnik

, Novitsky

, Rosen

. Evaluation of high-risk, comorbid patients undergoing open ventral hernia repair with synthetic mesh. Surgery, 2013; 153:120–125.

Cevasco

, Itani

. Ventral hernia repair with synthetic, composite, and biologic mesh: Characteristics, indications, and infection profile. Surg Infect (Larchmt), 2012; 13:209–215.

Darehzereshki

, Goldfarb

, Zehetner

, et al. Biologic versus nonbiologic mesh in ventral hernia repair: A systematic review and meta-analysis. World J Surg, 2014; 38:40–50.

Brahmbhatt

, Martindale

, Liang

. Jumping the gun? Evaluating the evidence for synthetic mesh in contaminated hernia repairs. J Am Coll Surg, 2014; 218:498–499.

10.

Fischer

, Basta

, Mirzabeigi

, Kovach

3rd . A comparison of outcomes and cost in VHWG grade II hernias between Rives-Stoppa synthetic mesh hernia repair versus underlay biologic mesh repair. Hernia, 2014; 18:781–789.

11.

, Salvay

, Paul

, et al. Soft polypropylene mesh, but not cadaveric dermis, significantly improves outcomes in midline hernia repairs using the components separation technique. Plast Reconstr Surg, 2009; 124:836–847.

12.

Sailes

, Walls

, Guelig

, et al. Synthetic and biological mesh in component separation: A 10-year single institution review. Ann Plast Surg, 2010; 64:696–698.

13.

Primus

, Harris

. A critical review of biologic mesh use in ventral hernia repairs under contaminated conditions. Hernia, 2013; 17:21–30.

14.

Ferzoco

. A systematic review of outcomes following repair of complex ventral incisional hernias with biologic mesh. Int Surg, 2013; 98:399–408.

15.

Slater

, van der Kolk

, Hendriks

, et al. Biologic grafts for ventral hernia repair: A systematic review. Am J Surg, 2013; 205:220–230.

16.

Itani

, Rosen

, Vargo

, et al. Prospective study of single-stage repair of contaminated hernias using a biologic porcine tissue matrix: The RICH Study. Surgery, 2012; 152:498–505.

17.

Mariette

, Wind

, Micelli Lupinacci

, et al. Practice patterns in complex ventral hernia repair and place of biological grafts: A national survey among French digestive academic surgeons. J Visc Surg, 2014; 151:9–16.

18.

Souza

, Dumanian

. Routine use of bioprosthetic mesh is not necessary: A retrospective review of 100 consecutive cases of intra-abdominal midweight polypropylene mesh for ventral hernia repair. Surgery, 2013; 153:393–399.

19.

Birolini

, Utiyama

, Rodrigues

Jr , Birolini

. Elective colonic operation and prosthetic repair of incisional hernia: Does contamination contraindicate abdominal wall prosthesis use?. J Am Coll Surg, 2000; 191:366–372.

20.

Janes

, Cengiz

, Israelsson

. Experiences with a prophylactic mesh in 93 consecutive ostomies. World J Surg, 2010; 34:1637–1640.

21.

Liang

, Berger

, Nguyen

, et al. Outcomes with porcine acellular dermal matrix versus synthetic mesh and suture in complicated open ventral hernia repair. Surg Infect (Larchmt), 2014; 15:506–512.

22.

Harth

, Broome

, Jacobs

, et al. Bacterial clearance of biologic grafts used in hernia repair: An experimental study. Surg Endosc, 2011; 25:2224–2229.

23.

Blatnik

, Krpata

, Jacobs

, et al. In vivo analysis of the morphologic characteristics of synthetic mesh to resist MRSA adherence. J Gastrointest Surg. 2012; 16:2139–2144.

24.

Cobb

, Carbonell

, Kalbaugh

, et al. Infection risk of open placement of intraperitoneal composite mesh. Am Surg, 2009; 75:762–767.

25.

Alicuben

, DeMeester

. Onlay ventral hernia repairs using porcine non-cross-linked dermal biologic mesh. Hernia, 2014; 18:705–712.

26.

Berrevoet

, Vanlander

, Sainz-Barriga

, et al. Infected large pore meshes may be salvaged by topical negative pressure therapy. Hernia, 2013; 17:67–73.

27.

Meagher

, Clarke Moloney

, Grace

. Conservative management of mesh-site infection in hernia repair surgery: A case series. Hernia, 2015; 19:231–237.

28.

Rosen

, Krpata

, Ermlich

, Blatnik

. A 5-year clinical experience with single-staged repairs of infected and contaminated abdominal wall defects utilizing biologic mesh. Ann Surg, 2013; 257:991–996.

29.

Harris

. Biologic Prosthetics & Beyond. UCSF Postgraduate Course in General Surgery: San Francisco, CA. 2013.

30.

Reynolds

, Davenport

, Korosec

, Roth

. Financial implications of ventral hernia repair: A hospital cost analysis. J Gastrointest Surg, 2013; 17:159–166.