Introduction

When presented with a primary ventral hernia, the surgeon and patient have a myriad of different approaches available to them. These range from a choice between suture and mesh repair, various mesh types, and a minimally invasive or open approach. Despite numerous publications on the topic, there is limited high-level evidence, and surgeons do not have a consensus on the best approaches for these patients [1].

Moreover, patients with a primary ventral hernia tend to have heterogeneity in their characteristics and presentation, adding a level of complexity to the preoperative evaluation [2]. The presence of a concomitant rectus diastasis and a female patient within reproductive age can alter the management plan for a primary ventral hernia repair. Furthermore, a young or thin patient may benefit from a different repair than an older or obese patient.

While previous studies have identified mesh repair as superior in larger ventral hernias, mesh repair in smaller hernias remained contentious, supported mainly by retrospective studies, registry-based data, and meta-analysis [3,4,5,6,7]. However, a recent randomized control trial evaluating mesh repair in primary umbilical hernias between 1 and 4 cm found similar improved recurrence rates with the use of mesh [8]. Furthermore, a meta-analysis of all randomized control trials conducted on this issue found similar outcomes with strong recommendation towards mesh use even in smaller hernias [9].

In this invited review, we discuss recent additions to the literature and recommendations for primary ventral hernias while concurrently evaluating current practices by surgeons participating in the Americas Hernia Society Quality Collaborative (AHSQC) to address any divergences and potential for improvement.

Methods

After receiving institutional review board (IRB) approval, the Americas Hernia Society Quality Collaborative (AHSQC) was used to acquire data for all primary ventral hernias. The AHSQC data registry is a prospectively maintained, surgeon-entered, point of care continuous quality improvement registry. Details regarding the design, implementation, and data quality assurance of the registry are discussed elsewhere [10].

A total of 224 surgeons in 215 sites contributed to the data, with 47% of patients from academic settings, 35% from private practice setting, and 18% from private practice with academic affiliation. The study population included patients with uncomplicated primary epigastric or umbilical hernias. An uncomplicated primary ventral hernia was defined as elective, non-recurrent, and with hernia width ≤ 10 cm. Patients that had concomitant procedures performed and non-clean wounds were also considered to have complicated hernias and were excluded. Patients that did not have mesh used and fascial closure was not achieved were excluded, as this was not considered a durable hernia repair. Figure 1 describes the inclusion and exclusion criteria for identifying the study population.

Fig. 1
figure 1

Inclusion and exclusion criteria

Full size image

A retrospective review of the prospectively collected data was conducted. Relevant variables included patient demographics, comorbidities, operative information, and post-operative outcomes.

The primary outcomes of interest were the surgical approach, including the use of mesh, the type of mesh used, the position of mesh, and the use of minimally invasive approaches. The type of mesh used was defined as follows: (1) flat mesh: a flat sheet of mesh excluding preformed meshes; (2) mesh patch: a flat mesh with mesh arms or deployment system to aid in mesh fixation; and (3) others: preformed meshes, mesh plugs, and bilayer meshes with a connector.

Secondary outcomes of interest include post-operative wound and medical morbidity at 30 day follow-up. Post-operative wound events collected in the AHSQC include Surgical Site Infection (SSI), Surgical Site Occurrence (SSO), and SSO requiring procedural intervention (SSOPI) [11]. SSI was defined according to the Center for Disease Control and Prevention (CDC) classifications as superficial, deep, or organ space [12]. Surgical Site Occurrence (SSO) included any SSI, in addition to wound cellulitis, non-healing incisional wound, fascial disruption, skin or soft-tissue ischemia, skin or soft-tissue necrosis, serous or purulent wound drainage, stitch abscess, seroma, hematoma, infected or exposed mesh, or development of an enterocutaneous fistula. Procedural interventions to be considered SSOPI included wound opening, wound debridement, suture excision, percutaneous drainage, partial mesh removal, and/or complete mesh removal.

Descriptive statistics were used to analyze the data, using exact frequencies, percentages, means and standard deviations, as well as medians and interquartile ranges wherever appropriate. The results are summaries of observed events in the study sample and not adjusted for potential confounders. To emphasize the descriptive nature of the analysis, no population inference was performed and no p values were reported. Two separate analyses were conducted, one looking at the use of mesh vs suture repair, and one looking at open vs minimally invasive repairs.

Results

Mesh vs suture

A total of 7030 patients met inclusion criteria. Of those, 69% underwent mesh repair, and 31% underwent suture repair. The cohort included 80% umbilical hernias and 20% epigastric hernias. The mean age was similar between mesh and suture repair (52 ± 13 vs 50 ± 15, respectively). The sample mean BMI was higher in the mesh group (32 ± 7 vs 28 ± 5). Consequently, the prevalence of obesity (≥ 30 kg/m2) in the mesh group was twice as much as its prevalence in the suture group (61% vs 35%, respectively) in the study sample. Patients undergoing mesh repair were more likely to have comorbidities (55% vs 43%). This was driven mostly by diabetes (12 vs 6%) and hypertension (37% vs 26%). The sample median hernia width in the mesh group was higher than the suture group (2 [2, 3] vs 1 [1, 2], respectively). Pain and an enlarging hernia that is interfering with activity were the two most common indications for repair (pain: 81% in mesh vs 80% in suture; enlarging hernia: 61% in mesh vs 39% in suture). Table 1 highlights the demographics, patient, and hernia characteristics for this cohort.

Table 1 Demographics, patient, and hernia characteristics
Full size table

The majority of the operations were performed through an open approach (60% for mesh repair and 97% for the suture repair). The most common mesh material used was permanent synthetic (98%), placed in the intraperitoneal space (46%) or the preperitoneal space (42%). The most common type of mesh used was flat (53%) and patch (46%). Table 2 provides the operative details for this cohort.

Table 2 Operative details
Full size table

Table 3 details all 30-day wound morbidity. At 30 day follow-up, there were more SSOs and SSOPIs reported for mesh repairs (SSO: 6% vs 3% and SSOPI: 2% vs < 1%). In the mesh group, the most common procedural management was percutaneous drainage (59% vs 0% in the suture group). In the suture group, wound opening was the most common procedure (71% vs 31% in the mesh group). SSI rate was similar in both groups. Table 4 illustrates 30-day medical morbidity, readmission, and reoperation. Note that the comparisons of 30-day outcomes are not adjusted for potential confounding factors or patient characteristics; rather, the results are simple summaries of observed events in the study sample.

Table 3 30 day medical morbidity, readmission, and reoperation
Full size table
Table 4 30 day wound morbidity
Full size table

Minimally invasive (MIS) vs open

Of the entire cohort, 28% underwent minimally invasive repair, while 72% underwent open repair. The mean age was similar in both groups (52 ± 13 in the MIS group vs 51 ± 14 in the open group). The sample mean BMI for the MIS group was higher than the open group (33 ± 7 vs 30 ± 6). Consequently, the MIS group had higher rates of obesity (66% vs 47%) in the study sample. Patient undergoing MIS repair were more likely to have comorbidities (57% vs 49%). This was mostly driven by diabetes (14% vs 8%) and hypertension (39% vs 32%). Table 1 highlights the demographics, patient and hernia characteristics for this cohort.

The majority of the MIS repairs were performed with the robot (55%) followed by laparoscopy (38%). Mesh was used in 96% of the MIS repairs and in 58% of the open repairs. The most common mesh was permanent synthetic (98% in MIS repairs and 99% in open repairs). The mesh was more likely to be place in the preperitoneal position when performing an open repair (50% in open repairs vs 29% in MIS repairs), while the intraperitoneal position was most common when performing MIS repairs (58% in MIS repairs vs 37% in open repair). A flat sheet was used for the majority of MIS repairs (88% in MIS repair vs 29% in open repairs), while a mesh patch was the most common in open repairs (70% in open repairs vs 8% in MIS repairs).

Table 3 details all 30-day wound morbidity. Similar rate of SSOs, SSIs, and SSOPIs was observed between the two groups. Table 4 illustrates 30-day medical morbidity, readmission, and reoperation. Again, these results are descriptive summaries of observed events in the study sample.

Tables 5 and 6 provide descriptive details of operative and demographic details for each group of surgeon affiliation (academic, private, private with academic affiliation).

Table 5 Surgeon affiliation operative details
Full size table
Table 6 Surgeon affiliation demographic details
Full size table

Discussion

In this analysis, we report on the current practices of surgeons participating in the AHSQC for primary uncomplicated umbilical and epigastric ventral hernia repair. Only one-third of this cohort received a suture repair, while the rest received a mesh repair. The majority of the mesh repairs were performed through an open approach with a permanent synthetic mesh patch placed preperitoneally, while less than one-third received MIS repair with a flat sheet of permanent synthetic mesh placed intraperitoneally. The use of the robot was more common than laparoscopy. In general, patients receiving a hernia repair with mesh or with an MIS approach tended to be more complex (higher BMI, larger hernias, and more comorbidities). To better evaluate these practices, it is important to contrast our findings to the current literature available for primary ventral hernia repair.

The initial consideration for patients presenting with a primary umbilical hernia is whether to use mesh or a tissue repair. Several studies with different designs report a tendency towards improved recurrence rates with the use of mesh. In 2001, a randomized control trial comparing suture to mesh repair found significant improvement in recurrence rates with mesh repair (1% with mesh vs 11% with suture at mean 64 months of follow-up) [13]. Although these were promising findings, the study had no hernia size cutoff, and almost one-third of the population had hernias larger than 3 cm repaired with a different mesh than smaller hernias. Therefore, the question of whether to provide a mesh repair in smaller umbilical hernias remained relevant. This predicament was the study question of a randomized control trial published in 2018 by Kaufmann et al. which included only patients with hernia diameters between 1 and 4 cm. In their study, they found a recurrence rate of 4% in the mesh group compared to 12% in the suture group at a median follow-up of 25.1 months [8]. These results were similar to reports in prior registry-based data and meta-analysis [5, 7]. Subsequently, a meta-analysis of all randomized control trials conducted to assess suture vs mesh repair for umbilical hernias found that mesh repair reduced the risk of recurrence with a relative risk of 0.28. Conclusively, the use of mesh for the repair was recommended for all hernias between 1 and 4 cm [9].

As the recommendation for mesh use is backed by multiple studies in the literature with different designs and similar conclusions, it is important to investigate the proper indications for its use, especially in an era of mesh apprehension and mesh litigation. In both previously mentioned randomized control trials, the mean age of the entire cohort was in the mid-50s. Whether a young healthy patient would have the same benefit when using mesh is unknown. In the trial by Kaufmann, the patient population was composed mostly of male patients (almost 84%), with a mean BMI of 28 kg/m2. Finally, most of the cohort in that study (71%) had hernias between 1 and 2 cm, and when analyzed separately, the difference in the recurrence rate slightly diminishes, but remains statistically significant (2% in the mesh group and 8% in the suture group) [8]. In the setting of an 88–92% success rate in patients with small hernias repaired primarily, a suture repair in a non-obese patient remains reasonable, especially if they do not desire mesh. However, if an obese patient presents, the benefit of mesh repair may be more pronounced, as an increasing BMI has been found to be associated with recurrence [14]. Therefore, in an obese population, mesh repair may be better indicated. While our analysis has shown that surgeons are more likely to offer a mesh repair, this was more selective towards more complex patients and hernias (patients that are obese have larger hernia widths and have more comorbidities). While the selection towards providing suture repair for thinner healthier patients with smaller hernias is understandable, the recent randomized control trial suggests that even in those patients, a significant benefit is seen with mesh repair. Therefore, it is important to share those results with these patients and involve them early in decision making to individualize each treatment plan.

Female patients within reproductive age with a primary ventral hernia present a predicament within this discussion. If the hernia is left unrepaired, there is a possibility for incarceration and obstruction with pregnancy due to the increased intra-abdominal pressure. However, if repaired, future pregnancies may disrupt the durability of the repair. Retrospective reviews found that pregnancy was independently associated with recurrence [15, 16], with mesh repair reported to result in less recurrences after pregnancy compared to suture repair [17]. Female patients within reproductive age should be counselled on the potential implications of their hernia management. As no current studies report on the risk of incarceration during pregnancy, asymptomatic primary ventral hernia may be managed by watchful waiting. When symptomatic, a mesh repair provides better durability. However, if a recurrence occurs, a reoperation after suture repair may be more feasible, and therefore, mesh repair can be postponed until reproduction is complete. Further research is required to assess these hypotheses before any formal recommendation can be made.

If mesh repair is deemed appropriate by the surgeon and the patient, another consideration is the type of mesh to use. While mesh has been shown to improve recurrence, the type of the mesh used differs amongst the aforementioned studies. In the trial by Kaufmann and colleagues, the mesh used was a flat sheet of mesh placed in the preperitoneal plane [8]. The trial by Arryyo et al. used two meshes, a flat sheet for larger hernias, and mesh plug for smaller ones [13]. Other meshes have been noted in the literature, most prominently mesh patches, mesh plugs, and bilayer mesh with a connector. In our study, the majority of the surgeons used either a flat mesh (52%) or a mesh patch (45%). Several studies have reported low recurrence rates with the use of the patch (0–9%) and no concerning complications [18,19,20,21]. However, when comparing mesh patch to a flat mesh, the MOROHEUS Trial, a randomized control trial, found no benefit to the patch, increased reoperations due to complications, and slight increase in recurrence rates (7.8% in patch repair vs 3.3% in flat mesh repair) [22]. The trial, however, was not powered enough to detect this difference. This reiterates the importance of studying these meshes in a research setting, ensuring adequate power and correct comparisons to detect differences in outcomes. The use of mesh plugs and bilayer mesh is similarly limited by the retrospective nature of the literature [23, 24] and a randomized control trial that is not powered to detect specific differences [25]. Therefore, evidence for the use of mesh in primary ventral hernias is limited to a flat mesh. While other options are viable, their use should be cautioned until high-level evidence is available. In our analysis, the majority of surgeons performing hernia repair through an open approach used a mesh patch (70%). A thorough study evaluating long-term recurrence rates and complication is required before any recommendation is provided. Preferably, this would be through a randomized control trial powered to detect these potential differences in outcomes.

A minimally invasive approach to primary hernia repair is also available as an option for these patients. A laparoscopic repair with mesh placed in the intraperitoneal position is the more traditional approach; however, our cohort had higher rates of robotic repair compared to laparoscopy. A propensity score matched analysis comparing laparoscopy to open repair found similar recurrence rates, fewer SSIs, with increased risk of port-site hernias and developing a bulge (not related to recurrence or seroma) [26]. A systemic review found similar outcomes with lower recurrence rate in the laparoscopic group. However, the pooled recurrence rate and effect size remained unknown [27]. An intraperitoneal mesh repair performed with the robot has also been described, with studies finding similar outcomes to laparoscopic repair [28, 29]. More recently, the enhanced-view totally extraperitoneal (eTEP) repair and the endoscopic mini/less open sublay technique (EMILOS) have been described for incisional and ventral hernias, with no safety concerns reported [30, 31]. The advantage of these techniques is utilizing the retromuscular space for mesh placement through an MIS approach, isolating the mesh from the abdominal viscera, and potentially avoiding complications associated with intra-abdominal access and mesh (bowel injury, port-site hernias, and adhesions between prosthetic and bowel). However, these new techniques were not studied exclusively in primary ventral hernias, and their use within this specific population is not yet described.

While minimally invasive approaches to primary ventral hernias are considered a valuable option, there is no clear evidence on the indications for these approaches. At large, studies on MIS repair for primary ventral hernias tend to have more obese populations [26, 28, 30, 32, 33]. This was similarly observed in our analysis (66% obese patient in MIS group vs 47% in the open group). While several studies have shown benefit to using laparoscopic repair in this population, they often do not report hernia width and include both incisional and primary ventral hernias with wide range of hernia sizes [34, 35]. Primary ventral hernias with smaller diameter may behave differently, and the benefit to MIS repair may not be as evident. In general, even for the non-obese population, heterogeneity of hernia sizes exists within MIS studies, and there is no clear evidence for a hernia width cutoff to benefits from such repairs.

The SAGES guidelines for laparoscopic ventral hernia repair report the advantages of laparoscopic ventral hernia repair over open repair given the decreased wound complication rate and similar recurrence rates and post-operative pain [36]. However, the use of MIS in our cohort seems to reflect an aversion to these approaches, as less than one-third of our patients received an MIS repair. Since the majority of these repairs involved placing the mesh intraperitoneally (58%), this could be driven out of fear of mesh complications related to placing the mesh in contact with abdominal viscera rather than the MIS repair itself. Of the 1% readmitted after an MIS repair, 55% were due to gastrointestinal complications (vs 29% in open repairs). Similarly, 62% of reoperations were due to bowel obstruction in MIS repairs (vs 12% in open repairs). However, due to the unadjusted nature of our analysis, associations with these complications cannot be drawn. Finally, the preperitoneal space was used for mesh positioning in 29% of MIS mesh repairs. This could be the eTEP approach as previously mentioned. However, no subgroup analysis was made to detect any differences in outcomes between the two different mesh positions.

The presence of a concomitant rectus diastasis with a primary ventral hernia requires special consideration. In a retrospective study, the presence of rectus diastasis has been linked to increased rate of recurrences with concomitant suture repair of primary hernia [37]. While other retrospective studies and a randomized control trial studying repair of rectus diastasis report acceptable recurrence rates and improved outcomes compared to baseline, they lack a comparison arm where the rectus diastasis is not repaired [38,39,40]. Due to paucity in the literature on this topic, a concomitant rectus diastasis repair is not currently recommended unless within a research setting to establish the potential benefit. The AHSQC does not collect the presence of concomitant rectus diastasis, and therefore, we are unable to comment on our cohort’s current practices.

Our study has several limitations that deserve mention. As inherent in any retrospective analysis, our results reflect a selective group of patients with bias in each offered approach, and since no adjustment analysis was conducted, no inference to the larger population of hernia repairs can be drawn. We are also unable to comment specifically on outcomes of different approaches in our patient population, as our analysis was descriptive in nature and no adjustment or subgroup analysis was conducted. In addition, since this is a large database study, a degree of error in data collection and measurement is expected.

Another limitation is the exclusion criteria implemented to define an “uncomplicated primary ventral hernia”. Since the AHSQC does not differentiate between planned concomitant procedures and those occurring as a complication, we elected to exclude all patients with any concomitant procedure performed. A similar limitation is present in excluding all non-clean wounds. Finally, our analysis is limited to experience and practice of AHSQC surgeons in the United States of America and their patient population. Importantly, the AHSQC represents both academic and non-academic surgeons across the United States which improves the generalizability of the results within this region.

Conclusion

Recent high-level literature recommends the use of mesh repair (flat mesh) in all patients with a primary hernia ≥ 1 cm in width. Despite this, AHSQC surgeons currently offer suture repair in one-third of cases. While the evidence for mesh use is limited to the use of flat mesh through an open approach, AHSQC surgeons most commonly performed the repair using a mesh patch and were more selective towards larger hernias and obese patients. Further research is required to evaluate the safety of mesh patches, and a mesh repair should be offered to a young non-obese healthy patient, as they benefit similarly from the use of mesh.