Poly-4-hydroxybutyrate Mesh for Ventral Hernia Repairs: A Single-Surgeon Experience

 

Ventral hernias are a very common problem facing patients, with about 350,000 ventral hernia repairs (VHRs) performed annually in the US.¹ These hernias are defects in the abdominal wall fascia through which abdominal contents can herniate, resulting in significant morbidity and reductions in quality of life.¹ Surgical repair of hernias is the mainstay of treatment and works to relieve symptoms and prevent future complications. 

Optimal repair of ventral hernias requires both primary closure of the fascial defect with sutures and reinforcement of the abdominal wall with a piece of mesh.² Mesh is required due to unacceptably high rates of hernia recurrence with suture-only repair.^3-5 Three types of mesh are currently available: synthetic, biologic, and biosynthetic.

Synthetic mesh is highly effective at providing long-term support to the abdominal wall and preventing recurrence; however, the long-term presence of an implanted foreign body predisposes patients to increased morbidity from wound and mesh infections.^6,7 Patients at increased risk for postoperative wound complications are therefore poor candidates for synthetic mesh.^8,9 Biologic mesh was designed as a temporary alternative to synthetic mesh, providing a reinforcement to the abdominal wall that is resistant to infections and does not remain in the body.¹⁰ Although biologic mesh successfully reduces the risk for long-term mesh infection and need for subsequent mesh removal, this alternative is expensive and studies have found it to be associated with high hernia recurrence rates.^11-14 

Biosynthetic mesh was developed to provide the benefits of biologic mesh at a lower cost and with a longer, more predictable time to resorption to reduce recurrence rates.⁶ Phasix mesh (CR Bard Inc) is a biosynthetic mesh of poly-4-hydroxybutyrate (P4HB) introduced in 2015 that resorbs over a period of 12 to 18 months and is replaced by host collagen.¹⁵ P4HB hydrolyzes at a known constant rate into carbon dioxide and water.¹⁵ In contrast to biologic mesh, this inherent degradation rate is not influenced by changes in pH, infection, or contact with enteric contents.¹⁶ Therefore, augmentation of the abdominal wall with P4HB provides a longer-term, more predictable model of reinforcement than biologic mesh.¹⁷

There are currently limited data regarding the use of biosynthetic mesh and its outcomes in clinical practice. Early data demonstrate acceptable postoperative outcomes and recurrence rates; however, further investigation is warranted.^6,18-24 In this study, we aim to analyze outcomes of a single surgeon’s experience using P4HB  mesh for ventral hernia repairs and to identify predictors of postoperative outcomes.

Study Design

We performed a retrospective study of adult (≥18 years of age) patients who underwent open ventral hernia repair with retrorectus or onlay P4HB mesh by the senior author from 2014 to 2020. Patients were identified from the electronic medical record and relevant data were collected, including demographics, risk factors, operative characteristics, postoperative outcomes, and follow-up data. Patients with less than 12 months of follow-up data and those who received other mesh materials in addition to P4HB were excluded. This study was approved by the Institutional Review Board.

The study population was divided into multiple subgroups for comparison. Patients were divided by history of prior ventral hernia repair, history of prior mesh infection, and mesh plane (onlay vs retrorectus). Demographics, risk factors, operative characteristics, and outcomes were compared across subgroups. Preoperative, intraoperative, and postoperative predictors of surgical site occurrences (SSOs), readmissions, reoperations, and recurrence were analyzed as well. SSO was defined per Haskins et al and included cellulitis, surgical site infection, seroma, hematoma, nonhealing incisional wound, wound dehiscence, infected mesh, and enterocutaneous fistula.²⁵ Recurrence was defined as a clinically evident recurrent ventral defect identified on physical examination or by follow-up imaging. Patients were not routinely scanned in follow-up unless there was a clinical indication to do so, such as suspicion of recurrence or concern for another intra-abdominal process.

Statistical Analysis

Data for the overall population and each subgroup are presented with summary statistics including medians, interquartile ranges, and proportions. Categorical variables were compared using a Fischer exact test or χ² test as appropriate. Continuous variables were compared using a Wilcoxon rank sum test. Stepwise multivariate logistic regression was performed using a tolerated significance of α < 0.2 to identify predictors of SSOs, readmissions, reoperation, and recurrence. Statistical significance in all analyses was defined as P ≤ .05. All analyses were performed using Stata 16.1 (StataCorp).

Overall Population

During the study period, 169 patients were identified who underwent open VHR with P4HB mesh and met inclusion criteria. Median follow-up time was 15 months (interquartile range [IQR], 12-18). Baseline characteristics, risk factors, operative details, and postoperative outcomes can be found in Table 1. 

Patients were mostly middle-aged (mean, 55 years; range 47-63), female (62.1%), obese (61.0%), and white (76.9%). Common medical risk factors included hypertension (61.0%), diabetes (29.0%), hyperlipidemia (36.7%), current tobacco use (21.3%), and immunosuppression (23.7%). Ostomies were present in 24.3% of patients. A history of prior hernia surgery (47.9%) and of prior mesh infection (18.3%) were common. Most patients had defects classified as American Society of Anesthesiologists (ASA) class III (61.5%) and Ventral Hernia Working Group (VHWG) grades 2 and 3 (46.2% and 30.2%, respectively). Wounds in most patients were classified as clean (39.7%) or contaminated (29.6%).

Median hernia defect size was 200 cm² (IQR, 120-375 cm²). At the time of hernia repair, the majority of patients underwent a concurrent abdominal procedure (76.3%), most commonly lysis of adhesions (58.0%). Onlay mesh was used in 53.2% of cases, retrorectus mesh was used in 44.4% of cases, and both onlay and retrorectus mesh were used in 2.4% of cases. Component separation occurred in 56.2% of cases. The most common component separation techniques (CST) used were bilateral transversus abdominis releases (TAR) (35.5%) and bilateral external oblique releases (EOR) (16.5%).

Postoperatively, median length of stay was 5 days (IQR, 4-7 days). SSOs occurred in 21.9% of patients, which were most commonly surgical site infections (9.5%), nonhealing wounds (8.3%), and fascial dehiscence (5.9%). Readmissions occurred in 21.3% of patients and reoperations in 17.8% of patients. Patients experienced low rates of hernia recurrence (4.7%).

Subgroup Analysis: History of Prior Ventral Hernia Repair

The subgroup analysis data for patients with and without a history of prior VHR can be found in Table 2. Patients with a history of prior VHR had significantly higher body mass index (BMI) (33.2 [range, 28.2-38.4] vs 31.1 [range, 24.7-35.8] kg/m²; P = .045) and lower rates of ostomies (17.3% vs 30.7%; P = .042). Patients with a history of VHR were more likely to be VHWG grade 4 (22.2% vs 2.3%; P = .001) and have dirty-infected wound (wound class 4: 21.0% vs 2.3%; P = .001). Other baseline characteristics and risk factors were comparable. Defect sizes, rates of concurrent abdominal procedures, and rates of component separation were similar across groups. Postoperatively, there were no differences in outcomes across groups including rates of recurrence.

Subgroup Analysis: History of Prior Mesh Infection

The subgroup analysis data for patients with and without a history of prior mesh infection can be found in Table 2. Patients with a history of prior mesh infection were more likely to have had prior hernia surgery (97.0% vs 39.0%; P < .001). Baseline characteristics and risk factors were otherwise comparable. There were no differences in defect sizes or other operative details. Postoperatively, patients with a history of prior mesh infection were more likely to have a mesh infection (6.5% vs 0.7%; P = .029). Other postoperative outcomes were comparable.

Subgroup Analysis: Onlay vs Retrorectus Mesh

The subgroup analysis data for patients who received either onlay or retrorectus mesh can be found in Table 3. There were no significant differences in baseline characteristics or relevant risk factors across groups, including rates of prior hernia surgery and prior mesh infection. Defect sizes were similar across groups. Component separation was more commonly used in patients who underwent retrorectus mesh placement (88.0% vs 27.8%; P < .001). Postoperative outcomes were comparable.

Multivariate Analyses: Postoperative Outcomes

The results of stepwise multivariate logistic regressions to identify predictors of SSOs, readmissions, reoperations, and recurrence can be found in Table 4. SSOs are significantly more likely in patients with obesity (odds ratio [OR] = 2.49; P = .042) and chronic obstructive pulmonary disease (COPD) (OR = 2.95; P = .029).

Readmission was significantly more likely in patients with a history of prior VHR (OR = 3.76; P = .036) and those with postoperative SSOs (OR = 77.01; P < .001). Readmission was less likely in patients who underwent component separation (OR = 0.26; P = .033) and those with a history of previous mesh infection (OR = 0.11; P = .020). 

Reoperations were significantly more likely to be required in patients with SSOs (OR = 64.73; P < .001) and hypertension (OR = 6.96; P = .012). Reoperations were less likely in patients with a history of mesh infection (OR = 0.17; P = .044) and those who underwent component separation (OR = 0.28; P = .049). Other risk factors, including a history of previous hernia surgery, were not found to be significant predictors of need for reoperation.

Hernia recurrence was significantly more likely in patients with a history of hypertension (OR = 13.64; P = .046), immunosuppression (OR = 42.57; P = .004), and previous hernia surgery (OR = 20.20; P = .014). Recurrence was less likely among patients with diabetes (OR = 0.05; P = .043) and those who underwent concurrent abdominal procedures (OR = 0.07; P = .026). 

Ventral hernia repair with synthetic mesh has been the gold standard for preventing hernia recurrence since 2004; however, recurrence remains a burden with rates exceeding 30%, and synthetic mesh is associated with complications including infection and mesh migration.^3,26 Biologic meshes are an expensive alternative to synthetic mesh that have been shown to effectively reduce the morbidity associated with complications common to synthetic mesh. Wide variability in outcomes has been reported across studies on biologic mesh; however, a meta-analysis found recurrence rates similar between biologic and synthetic mesh in potentially contaminated fields.^11,27,28 

Biosynthetic P4HB (Phasix) mesh is a new addition to surgeons’ armamentarium that seeks to provide longer-term support to the abdominal wall compared with biologic meshes while reducing complication rates compared with synthetic mesh.⁶ The 12- to 18-month resorption time of P4HB mesh allows for adequate time to maximize wound healing strength before loss of mechanical strength, which is lost completely at 12 months.^15,29 This allows the autologous myofascial abdominal wall to heal and gain tensile strength over time while being reinforced with P4HB mesh during its slow resorption. In this way, biosynthetic mesh leverages the ability of the living, myofascial abdominal wall to heal in a protected manner and, ultimately, allows full load transference from the mesh to the native myofascial abdominal wall once it has gained adequate tensile strength. This is increasingly important among patients with a history of prior VHRs given the scar tissue from each repair has 80% of the tensile strength of the prior tissue.³⁰ This strength reduction has a compounding effect in patients with multiple prior repairs, significantly increasing risk of recurrence.³¹ There is currently limited data on outcomes associated with P4HB mesh; therefore, further investigation is warranted.^6,18-20

In this study, we analyzed patients with a minimum follow-up time of 12 months, when the P4HB mesh is no longer providing mechanical support. We found that recurrence rates were low (4.7%) at a median follow-up of 15 months, despite nearly half of patients having undergone prior repairs and having significant comorbidities. Multivariate analysis identified prior VHR as a significant independent risk factor for recurrence; however, rates among those with history of VHR remained low (7.4%) and comparable with those without a history of VHR. These recurrence rates are consistent with multiple other studies, including two by Christopher et al that found 5.9% recurrence at a median follow-up of 20 months³² and 12.7% at 43 months,²⁰ and 1 study by van Rooijen et al that showed 11.0% recurrence at 24 months’ follow-up.³³ Additionally, a systematic review by Mellia et al found a 9.1% average recurrence rate.¹⁸ Our findings support that the use of P4HB mesh in patients with and without a history of prior VHR results in acceptable, low recurrence rates following loss of mesh integrity.

Complications related to mesh are a common challenge associated with both the use of synthetic and biologic meshes. Despite biologic mesh successfully mitigating risks associated with synthetic mesh, biologic mesh has been found to increase risk of SSOs and, specifically, seroma. SSO risk with biologic mesh exceeds 50% per Roth et al,²⁷ and seroma risk exceeds 20% per Alicuben et al³⁴  and Diaz-Siso et al.³⁵ Our experience with P4HB biosynthetic mesh demonstrates favorable rates of SSO (21.9%), with notably low rates of seroma (1.8%), surgical site infection (9.5%), and mesh infection (1.8%). These findings are consistent with those of Roth et al²² and Mellia et al.¹⁸ We also found P4HB mesh to share the infection-resistant benefits of biologic mesh. Although patients with a history of prior mesh infection had higher rates of postoperative mesh infection, a history of mesh infection did not increase risk of SSO, reoperation, or recurrence on multivariate analysis. 

Mesh placement in the retrorectus plane is known to produce lower recurrence rates than onlay mesh in VHR with synthetic and biologic mesh.³⁶ Early data on VHR with P4HB mesh has been consistent with this and supports the use of retrorectus mesh compared with onlay.¹⁸ In our experience, there were no significant differences in outcomes among patients who underwent onlay or retrorectus mesh placement. Additionally, multivariate analysis found no impact of mesh plane on rates of SSO, readmission, reoperation, or recurrence. Although this is inconsistent with the current data, it is possible the similar outcomes are a result of optimal patient selection via intraoperative factors not identified in this analysis. While it is the authors preference to use retromuscular sublay as the plane of mesh placement, onlay is performed if the posterior rectus sheaths are surgically absent or unavailable. The absence of the posterior sheaths or inability to be reapproximated is not uncommon in patients with multiple prior abdominal surgeries or repairs.  

Identifying predictors of postoperative outcomes is important to physicians’ ability to assess the risks of their operations and to allow patients to make informed decisions about their care. Predictors of hernia recurrence have been evaluated in the setting of VHR with synthetic and biologic mesh and summarized in a meta-analysis by Parker et al³⁷; however, biosynthetic mesh has limited data on predictors of postoperative outcomes. In this study, we identified predictors of SSO, readmission, reoperation, and recurrence. Risk of SSO was increased by obesity and COPD, which may be due to associated microvascular disease and altered immune modulators impairing wound healing.^38,39 SSOs were found to significantly increase the risk of readmission and reoperation but not recurrence, which is consistent with findings of other studies.¹⁸ A history of previous VHR was associated with increased risk of readmission and recurrence, likely due to the aforementioned compounded loss of tissue strength.^30,31 Additionally, hypertension was a risk factor for reoperation and recurrence, attributable to microvascular disease impairing proper wound healing.⁴⁰ Finally, immunosuppression significantly increased risk of hernia recurrence, likely due to impaired healing.⁴¹ Protective factors from negative postoperative outcomes included diabetes, component separation, previous mesh infection, and concurrent abdominal procedures. These positive predictors of recurrence were generally consistent with findings in synthetic and biosynthetic mesh; however, diabetes and prior mesh infection being protective factors is inconsistent and lack a physiologic explanation. These inconsistencies are likely the result of low overall recurrence rates impairing the conclusions. Overall, we found that the risk of postoperative complications is significantly increased by comorbidities associated with impaired wound healing. The identification of these predictors is an important first step to provide physicians with tools to accurately assess surgical risks; however, further studies with larger patient populations are needed to refine and expand these models.

Limitations of this study include its retrospective nature, the limited sample size, the lack of a comparison group, and the follow-up time. Additionally, these data were obtained from a single surgeon in a single health system. Finally, the low recurrence rates, in conjunction with a limited sample size, limit the power of the multivariate analysis of recurrence.

Ventral hernia repair with P4HB biosynthetic mesh produces acceptable rates of recurrence in both onlay and retrorectus planes with a favorable complication profile to biologic and synthetic meshes. Predictors of hernia recurrence include a history of prior hernia repair, hypertension, and immunosuppression.

Affiliations: ¹Division of Plastic Surgery, Department of Surgery, University of Rochester, Rochester, New York; ²Division of Plastic and Reconstructive Surgery, Department of Surgery, University of Missouri, Columbia, Missouri; ³Division of Plastic Surgery, Department of Surgery, University of Pennsylvania, Philadelphia, Pennsylvania; ⁴Division of Plastic and Reconstructive Surgery, Department of Surgery, Northwell Health, New Hyde Park, New York; ⁵Division of Traumatology, Surgical Critical Care, and Emergency Surgery, Department of Surgery, University of Pennsylvania, Philadelphia, Pennsylvania

Correspondence: Stephen J Kovach III, MD; stephen.kovach@pennmedicine.upenn.edu

Results of this study were first presented at the 2022 American Hernia Society Annual Meeting in Charlotte, North Carolina. 

Funding: This study received no financial support.

Ethics: This study was approved by the University of Pennsylvania Institutional Review Board.

Disclosures: The senior author, SJK, is a consultant for Bard (Murray Hill, NJ), BD (Franklin Lakes, NJ), and Integra LifeSciences (North Billerica, MA). These affiliations did not play any role in study design, data collection or analysis, and manuscript writing. No other authors have disclosures.