Long-term follow-up of laparoscopic repair of parastomal hernia using a bilayer mesh with a slit
- First Online:
- Cite this article as:
- Wara, P. & Andersen, L.M. Surg Endosc (2011) 25: 526. doi:10.1007/s00464-010-1205-9
- 196 Views
Open surgery for parastomal hernia has been associated with high morbidity and recurrence rates exceeding 50%. Laparoscopic mesh repair is a promising alternative. Published series on laparoscopic mesh repair of parastomal hernia, however, are few with relative short follow-up.
Seventy-two consecutive patients with 48 paracolostomy and 24 paraileostomy hernias were studied prospectively. Using a two-layer mesh with polypropylene on the parietal side and nonadhesive PTFE toward the viscera, a slit with a central keyhole was cut in the mesh modified after Hofstetter. Covering the fascial defect, the slit was closed laterally. Of the 70 patients discharged alive, 66 were followed for 6 months to 11 years (median = 3 years). Four patients were lost for follow-up.
Lysis of adhesions to the anterior abdominal wall, necessary in 68 patients (94%), was frequently a cumbersome and time-consuming task. In nine patients it was the apparent cause of inadvertent full-thickness enterotomy and seromuscular lesion in three and six patients, respectively. Three procedures (4%) were converted to open surgery. Postoperative complications were observed in 16 patients (22%), nine of whom required repeat surgery (13%). Two patients (3%) died. The median hospital stay was 3 days. Late mesh-related complications were observed in five patients (7%) after 5–34 months. Three patients developed abscess that required mesh removal in two patients, including one patient with a small bowel fistula. In two patients the mesh was removed because of small-bowel obstruction and stenosis of the colon at the level of the mesh, respectively. Parastomal hernia recurred in two patients 1 month and 52 months after surgery (recurrence rate = 3%, 95% confidence interval = 1–10).
Laparoscopic repair of parastomal hernia using bilayer mesh with a slit is associated with a risk of substantial postoperative morbidity, including late mesh-related complications, but a recurrence rate of less than 10%.
KeywordsParastomal herniaLaparoscopic repairBilayer mesh with slit
Open surgery for parastomal hernia with fascial suture or stoma relocation is known to be associated with high morbidity and recurrence rates exceeding 50% [1–3]. Open mesh repair, with the mesh in an on-lay position or placed intraperitoneally, appears to confer a lower recurrence rate [2, 3] but has a significant risk of mesh-related morbidity [2, 4]. Following the first anecdotal reports [5–9], there is accumulating evidence that laparoscopic mesh repair is feasible and has promise in the management of parastomal hernia [10–15].
Two laparoscopic techniques have evolved: the use of mesh with a slit and a central keyhole and use of mesh without a slit, the latter often called a modified Sugarbaker. At present, there appears to be a trend favoring mesh without a slit [10, 13, 14]. However, published series are observational and often have a short follow-up. Controlled trials are lacking and most series suffer from small sample size. The limited data, therefore, make it difficult to draw conclusions.
The aim of this prospective study was to monitor the postoperative morbidity and long-term outcome of the laparoscopic repair of parastomal hernia in which the same technique of using a bilayer mesh with a slit and a central keyhole was used throughout the entire study period.
Materials and methods
From December 1997 through September 2008, 72 consecutive patients underwent elective laparoscopic mesh repair of a paracolostomy hernia (48 patients) or a paraileostomy hernia (24 patients). Data from hospital stay and follow-up were prospectively recorded and stored in a specifically designed and user-friendly Ventral Hernia Database (www.itsosimple.net). Twenty-four patients had a history of previous repair. The male:female ratio was 36:36 and the patients had a median age of 62 years (range = 34–84). The ASA score was I for 27 patients, II for 33 patients, and III for 12 patients. The indications for surgery were poor fitting appliance, skin problems, obstructive symptoms, and pain in 35, 13, 33, and 68 patients, respectively.
The patients were operated on while in a standard supine position, but with the arm opposite the stoma hanging over the head for improved instrumental access to the anterior abdominal wall. Prophylactic antibiotics were not given. Because of an existing midline incision, pneumoperitoneum was established after Palmer with a Veress needle in the left subcostal margin in the midclavicular line in all patients. Three trocars were inserted laterally into the abdominal wall, opposite the stoma. A 30° optic was used. When required, adhesiolysis was achieved with sharp dissection as close as possible to the abdominal wall without the use of any thermal device. If detected, injury to the bowel was sutured. If present, hernia contents were retracted into the abdominal cavity. The whole circumference of the parastomal fascial defect was freed, leaving the hernia sac unresected.
Of the 70 patients discharged alive, 66 were followed in the outpatient clinic after 3 months, 1 year, and 3 years. After the third visit, patients were contacted by telephone, and if there were symptoms indicative of recurrence they were invited back for an examination in the outpatient clinic. Diagnosis of recurrence was made by clinical examination. When in doubt, a CT scan was performed. The median length of follow-up was 3 years (range = 6 months to 11 years). Twelve patients died during follow-up after a median of 3 years (range = 2 months to 8 years). Four patients did not respond and were lost to follow-up.
Adhesions to the anterior abdominal wall were encountered in 68 patients (94%). Adhesiolysis was often a cumbersome and time-consuming task. The median time for lysis was 45 min and the median time for surgery was 90 min. In nine patients, adhesiolysis was the apparent cause of inadvertent full-thickness enterotomy and seromuscular lesion in three and six patients, respectively.
The median size of the largest-diameter parastomal fascial defect was 4 cm (range = 2–8 cm). The median size of the mesh was 14 × 14 cm, with a median overlap of 4 cm. Early postoperative complications were observed in 16 patients (22%), 9 of whom required repeat surgery (13%). The complications included bleeding in two patients (one port site bleeding and one fatal intraperitoneal hemorrhage), peritonitis (one patient with a missed small-bowel injury), peristomal abscess (one patient), small-bowel obstruction unrelated to mesh (one patient), stomal complications (three patients), and a full-thickness hole in the colon necessitating relocation of the stoma (one patient). Seven of the nine complications requiring repeat surgery occurred during the first 3 years of the study period. Seroma and cardiopulmonary complications were observed in seven and five patients, respectively. With exception of the patient with a missed small-bowel injury, none of the remaining eight patients with sutured repair of full-thickness injury of the bowel wall (two patients) or seromuscular lesion (six patients) developed infectious complications. Conversion to open surgery was necessary in three patients (4%). The reasons were dense adhesions in two patients and full-thickness hole in the colon close to the stoma necessitating relocation of the stoma in one patient. Two patients died (3%), one due to a missed bowel injury resulting in multiorgan failure and the other due to uncontrollable bleeding because of portal hypertension unknown prior to surgery.
The median hospital stay was 3 days. Late mesh-related complications that occurred after 5, 7, 14, 25, and 34 months were observed in five patients (7%, 95% confidence interval [CI] = 2–15). Abscess occurred in three patients which required mesh removal in two patients, one of whom had a small-bowel fistula; in the third patient the infection was controlled after drainage of the abscess followed by VAC treatment, leaving the mesh untouched. In the fourth patient the mesh had to be removed due to stenosis of the colon at the level of the mesh. The fifth patient had the mesh removed because of small-bowel obstruction. He also had recurrent Crohn’s disease in the small intestine, but the obstruction was caused by dense adhesions to the mesh.
Parastomal hernia recurred in two patients after 1 and 52 months, respectively, yielding a recurrence rate of 3% (95% CI = 0.3-10).
With increasing observation time the prevalence of parastomal hernia may eventually exceed 50% in patients with a colostomy and 30% in patients with ileostomy. In most patients it is well tolerated and repair is needed only in the presence of intolerable symptoms, mostly due to obstruction, poor fitting appliance, skin problems, or unacceptable cosmetic appearance. Even though the prevalence of parastomal hernia is high, the number of patients discussed in published series is surprisingly small, indicating that only a small fraction of patients are offered surgical repair.
A number of techniques for repairing parastomal hernia have been described. Open suture repair of the fascial defect or stoma resiting are both associated with substantial morbidity and high recurrence rates [1–3]. In addition, stoma resiting carries the risk of incisional hernia at the previous stoma site and in the laparotomy incision. The two methods are no longer recommended for routine use. With open mesh repair the mesh is placed in an onlay, sublay, or intraperitoneal position. Mesh repair has lower recurrence rates [2, 3], but it is claimed that the risk of mesh-related morbidity is substantial [2–4].
Two techniques of open repair with an intraperitoneal mesh have evolved. Hofstetter  made a slit with a central hole and added small radial incisions in a PTFE mesh. He reported no recurrences in 13 patients but failed to reveal data on the extent of overlap and length of follow-up. Sugarbaker  used mesh without a slit. The colon was lateralized and the mesh was sutured to the fascial edge in an inlay position. According to the drawing in his article, the mesh hardly overlaps the defect but it is “cut to fill the defect snuggly.” He reported no recurrences in seven patients after 4-7 years of follow-up, but he did not describe what type of mesh he used. Stelzner et al.  modified Sugarbaker’s technique by using PTFE mesh without a slit but with a minimum overlap of 5 cm. They reported three recurrences in 20 patients after a mean follow-up of 3.5 years (range = 3 months to 7 years).
A laparoscopic approach to parastomal hernia repair combines the benefit of minimally invasive access with the use of an overlapping intraperitoneal mesh. Early reports have been promising but they include small numbers of patients, are retrospective studies, or change the technique of mesh repair during the study period. Controversy remains on whether to use mesh with a slit and keyhole or a mesh without a slit. After the first case reports of the modified Sugarbaker technique [5, 7], acceptable low recurrence rates have been reported, but the length of follow-up was relatively short or was not stated [10, 13–15]. In another study that used the Sugarbaker technique with PTFE mesh, there were 8 recurrences in 41 patients . The disappointing results prompted the authors to abandon that method of repair. After changing to a two-mesh sandwich technique, with polypropylene on the parietal side and antiadhesive polyvinylidene fluoride (PVDF) on the visceral side, in the next 25 patients no recurrence was reported, but the median length of follow-up was only 12 months .
The first reports on use of mesh with a slit were also promising [6, 8, 9, 11, 15], but recently published results have been less satisfactory with higher failure rates [14, 19, 20]. In a retrospective study, Muysoms et al.  reported 8 recurrences in 11 patients repaired with the keyhole technique, but only 2 recurrences in the subsequent 13 patients after changing to the Sugarbaker technique. The follow-up of these patients, however, was relatively short, and during the study period they used a polyester-based mesh, PTFE mesh, and a bilayer composite mesh. However, they failed reveal which mesh they selected for each of the two techniques . In the other two studies with less satisfactory results [19, 20], PTFE mesh with a slit was used.
With exception of the present study, mesh containing polypropylene was used in only two anecdotal reports [6, 7]: in the second period of the study by Berger  and in three patients in the study by Muysoms et al. . In most studies PTFE was the preferred mesh material. It was also used in two studies that had disappointing results with the keyhole technique [19, 20]. Because of lack of controlled data, one can only speculate whether the failures are related to the technique of making the keyhole, the size of the keyhole, or the mesh used. Estimation of an appropriate keyhole size is difficult to standardize, especially when taking the variation of individual patients into consideration. According to Muysoms et al. , the predominant site of hernia recurrence is the central hole. It could be that polypropylene is more likely to induce more fibrosis around the central hole than PTFE, thus preventing hernia from recurring. Using the keyhole technique with PTFE mesh in a relatively large series, Hansson et al.  recently reported a recurrence rate of 37% after a median follow-up of 3 years. It was suggested that the failures were due to shrinkage of the PTFE mesh which made the central hole larger.
The present study indicates that using the keyhole technique with bilayer mesh of polypropylene and PTFE is successful in preventing recurrences. Because of the lack of data from controlled trials, it could be argued that PTFE mesh is probably less suitable for the keyhole technique but may be appropriate as a single mesh overlapping the fascial defect as well as the lateralized exiting bowel. Furthermore, it may be technically less demanding to perform. If postoperative morbidity and recurrence rates prove to be consistently low without serious late complications at the site of the outlet of the bowel, it may end up as the method of choice in the management of parastomal hernias, but this conclusion must await controlled trials.
The properties of polypropylene are well established, enabling fibrous ingrowth into the mesh. It is generally believed that unprotected polypropylene, compared to PTFE, is more prone to have mesh-related complications, but there are no hard data to support this belief. In a relatively large series of 58 patients in which local onlay mesh repair with polypropylene was used, wound infection, fistula, and mesh erosion were observed in only two, two, and one patient, respectively . However, unprotected polypropylene mesh placed intraperitoneally in seven patients caused dense adhesions, small bowel obstruction, and abscess in four, three, and one patient, respectively .
Although based on limited data from a small series, we conclude that unprotected polypropylene should not be used in laparoscopic repair. It requires an additional antiadhesive layer on the side covering the viscera. At present, polypropylene mesh with coating is available as a two-component mesh with PVDF or a bilayer mesh with PTFE. In the present study only one complication in the immediate postoperative period could be attributed to the bilayer mesh, but late mesh-related complications requiring repeat surgery did occur in five patients. It can be speculated whether this is the price to pay for the fairly low recurrence rate in the present study.
To our knowledge there are limited data on the use of collagen-based mesh implants [22, 23]. We are unaware of any published data to support the claims of reduced infection risk when collagen-based mesh is used in the repair of parastomal hernia. The ideal biomaterial for repair of ventral hernia, with parastomal hernia being the most complex, has yet to be developed. At present we lack hard data on what key factors, e.g., selection of mesh, technique of mesh repair, size of overlap, technique of mesh fixation and subsequent shrinkage of the mesh, have a potential impact on recurrence.
The most risky step in laparoscopic parastomal hernia repair is most likely lysis of dense adhesions to the anterior abdominal wall. We agree with Hansson et al.  who emphasizes the importance of avoiding full-thickness injury of the bowel wall, but also seromuscular lesions, which should be sutured when present. Missed injuries to the bowel wall pose a life-threatening risk to the patient.
In conclusion, laparoscopic mesh repair of parastomal hernia can be technically challenging. We found that laparoscopic repair using bilayer mesh with a slit was associated with a recurrence rate of less than 10% but has the risk of substantial postoperative morbidity, including late mesh-related complications.
Drs. P. Wara and L. Maagaard Andersen have no conflicts of interest or financial ties to disclose.