Incisional hernia is a common complication after laparotomy with a reported incidence of 20 % [1]. Suture repair of ventral hernias alone yields unsatisfactory results, with a recurrence rate of 68 %, and thus, mesh repair is the preferred method [2]. LeBlanc and Booth [3] first described the laparoscopic ventral hernia repair technique, applying the concept first espoused by Rives et al. [4] of broad overlapping prosthetic coverage, allowing even distribution of intraabdominal forces along the surface of the mesh in a position deep to the fascia. Subsequent studies show the technique of laparoscopic ventral hernia repair (LVHR) to have lower postoperative morbidity compared to open hernia repairs [5]. The LVHR with intraperitoneal mesh is well accepted. However, the long-term sequelae of intraperitoneal mesh, particularly in the event of a subsequent abdominal operation, has not been well described.

In one study of elective abdominal surgery, adhesiolysis, alone, was associated with an increased incidence of sepsis, intraabdominal complications, wound infection, a longer hospital stay, and higher hospital costs [6]. Mesh can complicate future abdominal surgery. In one large study of reoperation after ventral hernia repair, prior mesh was associated with a 20 % risk of enterotomy or unplanned bowel resection [7]. More specifically, mesh position may play a more significant role. Compared to extraperitoneal mesh, intraperitoneal polypropylene mesh is associated with a need for small bowel resection in 20 % of patients resulting in enterocutaneous fistulas 5 % of the time [8]. Having a hernia specialty, referral practice, we operate on many patients who have had prior intraperitoneal mesh hernia repairs elsewhere. We have noted difficulty in reoperating these patients due to the adhesions between the mesh and the viscera. This has resulted in enterotomies and unanticipated bowel resections. We have also seen delayed intraperitoneal mesh infections which developed shortly after a reoperation where the prior intraperitoneal mesh was transgressed in some form. As a result of this experience, we have long advocated against placing mesh within the peritoneal cavity, due to a perceived risk of complications at a future surgery. We understand that we are only seeing the complications of intraperitoneal mesh repairs (numerator), without any knowledge of the larger number of patients who live life, uneventfully, after intraperitoneal mesh repair (denominator). In this report, we identify the characteristics of the patients in whom we performed LVHR, who went on to have reoperations to investigate whether prior intraperitoneal mesh complicated future surgery.

Methods

Using the Greenville Health System Hernia Center database, a retrospective review was performed to identify patients who underwent LVHR by us between August 2005 and May 2014 and then subsequently underwent an additional operation. Patient demographics, details of operative technique, indication for reoperation, reoperation performed, and outcomes of those reoperations were recorded. Primary outcome measures of the reoperations were surgical site infection (SSI) and unplanned bowel events such as enterotomy or bowel resection (EBR). An SSI was based on criteria established by the Centers for Disease Control (CDC) and was divided into superficial, deep, and organ space [9]. This study was performed with the approval from the Greenville Health System Institutional Review Board.

Continuous bivariate analysis was done using Students t test. Discrete variables were compared using Chi-squared test, or Fisher’s exact test for small sample sizes (n < 5). P-values <0.05 were considered indicative of statistical significance. All data analyses were completed using R statistical software (R version 3.0.2).

Results

A total of 757 patients underwent LVHR at the Hernia Center of the Greenville Health System. Conversion from laparoscopy to open occurred in 24 patients, these patients were excluded from analysis. Subsequent abdominal operations were performed in 17 % (125 patients). The demographics and perioperative information of the entire LVHR population as well as the non-reoperated and reoperated groups are shown in Table 1 and Table 2. Overall, most patients were female (62.8 %) and obese with an average BMI of 33.9 kg/m2. Hypertension was the most common comorbidity (61.3 %) among our patient population, and the majority (97.9 %) were American Society of Anesthesiologists class 2 or 3.

Table 1 Patient demographics
Full size table
Table 2 Index hernia repair perioperative data and follow-up
Full size table

The majority of the initial repairs were a clean wound class (77.4 %), and 37.7 % of the patients had prior hernia repairs. Defect sizes were large with an average hernia size of 115 cm2. Mesh used was ePTFE (57 %), barrier-coated polyester (35 %), and barrier-coated polypropylene (8 %). Spontaneous mesh infection rate was 2 % (n = 15; ePTFE 12, polyester 2, polypropylene 1). All patients with mesh infection ultimately underwent mesh removal. After a mean follow-up of 19.4 months, the overall hernia recurrence rate was 8.4 %. Operative time for the index LVHR was higher for the group which ultimately required a reoperation. There was no statistical difference in any of the other base demographics or perioperative data of either subset.

The incidence of subsequent abdominal operation after an index LVHR at our center was 17 % (125 patients); 26 patients underwent more than one reoperation. The mean time from index hernia repair to reoperation was 2.2 years. As shown in Table 3, the most common indication for reoperation was recurrent hernia (33 patients, 26.4 %), followed by bowel obstruction (18 patients, 14.4 %), hepatopancreaticobiliary surgery (17 patients, 13.6 %), infected mesh removal (15 patients, 12 %), gynecologic surgery (10 patients, 8 %), colorectal surgery (8 patients, 6.4 %), bariatric surgery (4 patients, 3.2 %), trauma (1 patient, 0.8 %), and miscellaneous (19 patients, 15.2 %).

Table 3 Indications for reoperation
Full size table

Of the 18 patients operated on for bowel obstruction, 55 % had barrier-coated polyester, 33 % ePTFE, and 11 % barrier-coated polypropylene. As a function of the indications for reoperation, Table 4 demonstrates the complications associated with these reoperations. The rate of SSI was 5.6 %, occurring in 7 patients. The incidence of EBR at reoperation was 4 %, occurring in 5 patients, all of whom were reoperated for complete bowel obstruction. No other indication for reoperation resulted in EBR. The mesh in these cases was barrier-coated polyester (3) and ePTFE (2). In each case, the EBR was the result of direct mesh–prosthetic adhesions.

Table 4 Complications of reoperation
Full size table

A subsequent mesh infection occurred after reoperation in 3 patients. The first patient was reoperated for bowel obstruction, required an ileocecectomy, and developed an SSI and subsequently a mesh infection. The second patient underwent placement of a peritoneal dialysis catheter, which was later removed, and then eventually developed the mesh infection. The third patient underwent a diagnostic laparoscopy to manage a large, symptomatic seroma, dorsal to the mesh which later went on to become infected.

Discussion

Patients who underwent LVHR at our center over a nine-year period had a 17 % incidence of requiring a subsequent abdominal operation. This is similar to other investigators who have reported rates between 17 and 25 % [7, 10, 11]. One could foresee that throughout a patient’s life, however, the need for reoperation would continue indefinitely, particularly as they age and develop other surgical maladies.

Prior to this study, we had felt that intraperitoneal mesh was associated with an increased rate of EBR at reoperation due to tenacious mesh–visceral adhesions, as well as potential secondary mesh infection. This belief was largely based upon a vast experience with complex reoperative hernia repairs, in patients who have had prior intraperitoneal mesh placed at outside institutions. The rationale was biased, however, as we were only treating patients with complications requiring reoperative hernia repair or management of a mesh infection, without having a complete understanding of all the other patients who have had intraperitoneal mesh placed, without sequelae. Analysis of our own series of LVHR patients demonstrated that the actual risk of an adverse event at the time of a subsequent abdominal operation in patients with intraperitoneal mesh was quite low, with an EBR rate of 4 % and secondary mesh infection rate of 2.4 %.

To understand the relationship between adhesions and EBR, we look to large series addressing this point. In a study of 755 patients undergoing elective abdominal surgery, adhesions were encountered in 62.9 % of patients. Adhesiolysis was associated with an increased incidence of postoperative sepsis, intraabdominal complications, wound infection, a longer hospital stay, and higher hospital costs [6]. This study demonstrates how the mere presence of adhesions increases the complexity of a reoperation. Adhesions can also be exacerbated by a foreign body within the peritoneal cavity; thus, mesh exposed to the viscera could certainly be associated with significant adhesion formation. A large study of the National Surgical Quality Improvement Project database analyzed 1124 patients who had reoperations after prior ventral hernia repair. The presence of a prior mesh was associated with a 20 % risk of EBR, a fourfold increase over patients who had undergone a prior sutured hernia repair [7]. Our series, however, demonstrated a significantly lower incidence of EBR (4 %). Another study looked at 1444 patients with prior hernia repairs from the Veteran’s Administration Hospital System. At a median time of 80 months, 25 % of these patients had undergone a subsequent abdominal operation, most of which were for recurrent hernia. On multivariate analysis, polypropylene underlay and inlay and absorbable/biologic mesh significantly increased the operative time for subsequent surgery. Repair type, mesh type, or mesh position had no significant effect on risk of inadvertent enterotomy. Conversely, other investigators found the type and position of the mesh to be important. Intraperitoneal polypropylene mesh from prior hernia repair was associated with a need for small bowel resection in 20 % of patients and enterocutaneous fistulas in 5 % at the time of subsequent surgery, compared to none if the mesh was originally extraperitoneal [8]. This study demonstrated the effect of unprotected polypropylene mesh exposed to the viscera, which is no longer a common surgical practice. When the viscera are to be exposed to mesh, coated meshes with a tissue-separating layer are used. These permanent or temporary barriers serve to separate the bowel from the mesh until a mesothelial layer of tissue develops across the mesh, preventing the bowel from becoming adherent to the mesh. Our use of permanent and temporary barrier meshes likely explains our lower rate of EBR at the time of reoperation. Every instance of EBR in our series occurred exclusively in patients who were being reoperated for bowel obstruction. The EBR occurred at the point of dense mesh–visceral adhesions which interestingly was never the point of actual obstruction. We speculate that patients who develop dense adhesions to mesh may also have a predilection for developing dense entero-enteric adhesions resulting in bowel obstruction. There may have been an association with the development of a bowel obstruction and the mesh type used for repair; however, the numbers are too small to draw any real conclusions. Of the 18 patients operated on for bowel obstruction, 55 % had barrier-coated polyester, 33 % ePTFE, and 11 % barrier-coated polypropylene. Despite this, EBR occurred in 5 patients; 3 barrier-coated polyester; and 2 ePTFE patients.

Our overall rate of SSI after reoperation was 5.6 %. This rate is not surprising, considering all the SSI events occurred in patients who had undergone a reoperation in the face of some level of contamination. The major concern, however, would be the potential for secondary infection of the original intraperitoneal mesh. Our rate of secondary mesh infection was 2.4 %. This compares favorably with the 2 % rate of mesh infection we experienced after index LVHR. Thus, in our experience, reoperating a patient with prior intraperitoneal mesh is safe, without any increased risk of having the mesh to become secondarily infected, despite performing a contaminated field surgical procedure.

Conclusion

In a large consecutive series of LVHR, 17 % of patients required a subsequent abdominal at a mean of 2.2 years. The most common indication for reoperation was hernia recurrence, followed by bowel obstruction, and hepatobiliary surgery. The overall incidence of enterotomy or unplanned bowel resection (EBR) at reoperation was 4 %, occurring exclusively in patients who were reoperated upon for complete bowel obstruction. The incidence of secondary mesh infection after subsequent operation was 2.4 %. Although reoperation after LVHR is generally safe, surgeons should discuss with their patients the potential long-term implications of having an intraperitoneal mesh and how it may impact future abdominal surgery.