Incisional hernias post cytoreductive surgery/peritonectomy and hyperthermic intraperitoneal chemotherapy: a systematic review and meta-analysis

Cytoreductive surgery (CRS) combined with hyperthermic/heated intraperitoneal chemotherapy (HIPEC) is an effective management strategy for advanced peritoneal malignancies [1‐3]. CRS aims for complete tumour removal, involving extensive peritoneal and visceral resection [4]. Once optimal cytoreduction has been achieved, HIPEC is employed intraoperatively and, in select cases, is followed by early postoperative intraperitoneal chemotherapy (EPIC) [5, 6]. However, it is associated with complications including bowel perforation, anastomotic leak and incisional hernias (IH), alongside a postoperative morbidity and mortality reported in the range 22–41% and 2–5%, respectively [7‐12]. The overall incidence of IH in those undergoing laparotomy has been documented in the literature to exceed 20% [13‐16]. Late morbidity and in particular the occurrence of an IH have not been well studied in those with peritoneal malignancies managed with CRS/HIPEC [17, 18].

Although the true incidence is unclear, several studies have reported an IH incidence between 7 and 17% [19‐21]. CRS/HIPEC represents a complex surgical intervention of considerable duration [22]. Notwithstanding, this procedure poses a potential risk for hernia development, given several inherent factors. Primarily, a significant proportion of CRS/HIPEC patients have a history of previous abdominal surgeries, a factor well-documented to increase hernia susceptibility due to abdominal wall weakening [23, 24]. Moreover, the lengthy duration of the CRS/HIPEC procedure necessitates sizable incisions, thereby subjecting the abdominal wall to heightened stress and augmenting the likelihood of herniation [22, 25, 26]. The intraperitoneal delivery of chemotherapy during CRS/HIPEC can result in immunosuppression, further compromising abdominal wall integrity [21, 27]. Nonetheless, the precise proportion of patients developing an IH following CRS/HIPEC remains largely unexplored, as existing studies predominantly focus on short-term morbidity and long-term oncological outcomes [28].

Understanding the proportion of IH, risk factors, and outcomes related to IH post-CRS/HIPEC is essential for risk assessment, prevention, and optimal management. Further research is needed to refine preventitive strategies, standardize surgical techniques, and assess long-term outcomes to enhance patient care. The aims and learning points of this systematic review and meta-analysis is to assess the proportion of patients, risk factors and outcomes in patients who develop IH post CRS with or without HIPEC and how this information can be utilized to enhance clinical decision making for the betterment of patient outcomes and quality of life.

Our search yielded ninety four articles of which nine studies were selected for data extraction [10, 19‐21, 32‐34, 44]. Studies selected were published between 2014 and 2023, conducted in six countries. A total of 1416 patients were included in our analysis. Study characteristics and patient demographics are found in Table 2 and 3, respectively. All but one study was conducted retrospectively [19]. All but one study took place at a single institution [33]. One study did not specify its location but collected data from a prospectively maintained database [32]. Eight studies included patients who underwent both CRS and HIPEC. Spencer et al. included patients who underwent CRS only [33]. Cascales Campos et al. described two groups, CRS only and a group who underwent both CRS and HIPEC [10]. Patients who underwent HIPEC received variable regimens, but all with either platinum agents such as oxaliplatin/cisplatin, mitomycin or both. Pathologies were wide ranging with the majority described as ovarian cancer, peritoneal mesothelioma, colorectal cancer and appendiceal cancer. Three studies included recurrent disease [10, 20, 32]. Spencer et al. and Wong et al. reported on ovarian cancer and mesothelioma in isolation, respectively [33, 34].

We performed a systematic review and meta-analysis regarding the occurrence of IH post CRS plus or minus HIPEC. From our results, we report a pooled proportion of patients developing an IH of 12% in the HIPEC group, 7% in the non-HIPEC group and 11% overall. There was evidence of statistical heterogeneity in the HIPEC group and between groups. Considering the odds of developing an IH post CRS/HIPEC we reported an OR of 1.9, which was not statistically significant, indicating further research is required to determine clinical significance. These results indicate that IH may be more likely in the HIPEC group. We also observe no statistically significant difference between mean follow-up times in HIPEC or non-HIPEC groups, which can affect the rates of IH observed [45].

Rates of IH post midline laparotomy, not specifically related to CRS, of 10–30% have been described [46]. The pooled proportion of IH post CRS plus or minus HIPEC of 11% is at the lower end of expected rates. This may be due to a number of reasons, including closure technique, BMI, previous surgery, age and gender [47‐50]. As well as this, the actual rates of IH post CRS plus or minus HIPEC may be higher. Considering midline laparotomy in general, only 75% of IH were seen to occur within 2 years of surgery in previous studies [48].

Beadles et al. have shown incidence rates of IH emergency repair in elderly women and men of 23.5 and 32.0 per 100,000 population in the United States, respectively [51]. This serves to highlight the impact IH can have on patient outcomes, and healthcare systems.

In an obese cohort undergoing midline laparotomy, required IH repairs rates of 29% have been reported in the literature [52], with the rate of incarcerated IH repair reported as 3.7% [53]. The expected rates of IH in midline laparotomy in conjunction with peritonectomy may be expected to be higher. Within the included studies, 11 out of 28 patients underwent surgical correction of their IH, with one surgery classed as an emergency due to incarceration [10]. Tuttle et al. reported 10 patients whom underwent surgical repair out of 26 IH [21]. 4 out of 14 IH were repaired in Ravn et al.’s publication, with one case classed as an emergency obstruction [19]. 12 out of 19 IH were repaired electively, in Struller et al.’s study [20]. 7 from 265 patients underwent non-emergency IH repair in an ovarian cancer cohort [33]. What must be considered is the benefit of CRS and HIPEC in contrast with the risks of emergency IH repair and morbidity associated with this procedure, in an immunosuppressed patient population.

Regarding ventral hernias, laparoscopic as opposed to open cases have been described as a more cost effective method of repair when hernias recur, however, all are economically costly [54]. In the case of IH repair post peritonectomy, open surgery may be the most effective option due to the fact it may no longer be possible to place a pre-peritoneal mesh. Additionally a retro-rectus approach may not be feasible if the posterior rectus sheath is resected, leaving the option of an onlay repair, which has its own complications [55, 56]. If open repair is undertaken this will further increase repair economic cost.

Our review also identified risk factors that may suggest patients are more likely to develop an IH as described in Table 3.

The primary malignancy was seen to affect IH rates post CRS/HIPEC, with pseudomyxoma peritonei and mesothelioma patients more likely to develop an IH (p = 0.022) [20]. A colorectal primary has also been described as a risk factor for IH by Cascales Campos et al. (p = 0.01), while Spencer et al. details a suboptimal CRS as a risk factor (p < 0.001), which may be considered a surrogate of primary cancer aggressiveness [10, 33]. Nutritional status was also reported as a risk factor for IH in one study (p < 0.001) [10], which is in agreement with previous literature regarding inguinal hernias [57]. Peritoneal cancer/carcinomatosis index (PCI) has been described as accurate in predicting outcomes, however, others have questioned its benefit [58, 59]. Parikh described a PCI greater than 20 as a high burden of disease, but failed to show statistical significance in relation to wound complications post CRS, however, they did not specifically analyze PCI in relation to IH [32]. Wong et al. also reported the effect of PCI on outcomes. They did not analyze PCI in relation to IH but did find PCI > 20 to correlate with overall survival [34]. Of note, our included studies did not report the effect of stoma formation on IH rates, however, previous research has shown rates of anastomotic leak and prognosis seem to be within the established range when stomas are fashioned in CRS [60, 61]. Further research relating stoma formation to IH outcomes may be clinically useful.

The use of meshes in patients with peritoneal metastases has been questioned [62]. However, the use of mesh reconstruction in patients post CRS/HIPEC/laparotomy has been shown to be safe and effective [46, 63, 64]. Only a small cohort of patients in one study included in this analysis reported mesh use [32], however, ongoing studies (ClinicalTrials.gov identifier: NCT03953365) relating to the outcomes regarding mesh use post CRS/HIPEC may further enhance patient outcomes regarding IH. One study included in our analysis did not show a IH development rate that was statistically significant between mesh and no mesh groups [32].

The major limitation of this meta-analysis is an inherent limitation of each of the included studies. The follow-up time was likely insufficient to detect all IH post surgeries. The HIPEC group had a mean follow-up of 30 months and the non-HIPEC group had follow-up of 18 months, falling short of the recommended minimum follow-up period of 36 months [45]. Another limitation is the lack of standardisation in follow-up times, and while we utilized 12 months as an inclusion minimum there is likely to be a difference in IH picked up with longer follow-up, however, in this patient cohort longer follow-up may be difficult due to patient mortality prior to IH development [19‐21]. Previously described limitations of the statistical methods are also valid [39]. Due to the low volume of papers describing non-HIPEC cohorts this meta-analysis is likely underpowered to detect all outcome differences between HIPEC and non-HIPEC groups, and there is a risk of type II error occurring as a result. Further studies may consider evaluating the clinical significance of HIPEC versus non-HIPEC IH rates.

The proportion of patients developing an IH post CRS plus or minus HIPEC is in the range expected, considering midline laparotomies in general. This analysis suggested that HIPEC may contribute to a greater proportion of patients developing an IH, however, this finding was not statistically significant. Further studies may be clinically useful to further investigate HIPEC’s role in IH development.