Transitory Stoma at the Time of Complete Cytoreductive Surgery Affects Survival for Patients with Advanced-stage Ovarian Cancer

Discussion

A TPS in advanced ovarian cancer is associated with a high risk of recurrence. Moreover, our series showed an impact of TPS on OS. One explanation for these results is that patients with a TPS did not comply with the adjuvant chemotherapy schedule and therefore received fewer cycles of adjuvant chemotherapy.

These results raise three questions: At the time of CRS, what are the situations in which a TPS is required? How can malnutrition and dehydration be avoided? When should stomal closure occur?

In theory, in many countries, a patient who undergoes CRS with a bowel resection for advanced ovarian cancer should have an adequate nutritional status (serum albumin >30 g/l and adequate weight) and preoperative immunonutrition with Oral Impact® that contains a combination of arginine, n-3 fatty acids, and nucleotides (3 bags/day) over 7 days (16). Comorbidities such as the smoking status, coronary artery disease and diabetes were not significantly associated with anastomotic leakage in the series reported by Tseng et al., comprising 331 patients with FIGO stage II-IV ovarian cancer (44 with a TPS) (9). Ultimately, comorbidities should be taken into consideration for indicating CRS but should not systematically include a TPS. At the time of surgery, an anastomotic defect at intraoperative testing remains an incontestable indication for a TPS. Four additional elements can help the surgeon decide to create a TPS: the total length of the bowel resected, the level of the colorectal anastomosis, the number of anastomoses and perioperative anesthesiology criteria. In the gynecological literature, data are lacking for the three surgical elements. Only one study demonstrated that the length of the colonic resection was a predictive factor for anastomotic leakage. This should be confirmed, as mentioned by the authors (8). Anesthesiological management during surgery should monitor hemodynamic parameters in order to reduce the risk of complications such as anastomotic leakage. However, in the end, objective parameters indicating whether to perform a TPS or not are lacking. Ultimately, the rate of TPS should remain as low as possible due to consequences such as dehydration, acute renal failure and weight loss. The rate of TPS was 9% in our series, and 13% in the series reported by Tseng et al. (9).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Prognostic impact of a transitory protective stoma (TPS) at the time of complete cytoreductive surgery for advanced-stage ovarian cancer on overall survival.

A Gynaecologic Oncology Group Study published by Hess et al. showed that change in body weight during primary chemotherapy was a strong prognostic factor. Indeed, in that series of 792 patients with a stage III FIGO ovarian cancer, the relative risk of death increased by 7% with each 5% decrease in body weight [hazard ratio (HR)=0.93, 95% confidence interval (CI)=0.88-0.99, p=0.013] (17). Moreover, Yim et al. analyzed the nutritional status in 213 patients with an advanced ovarian cancer who underwent surgery followed by chemotherapy. The authors found that residual disease >1 cm (HR=2.251, 95% CI=1.218-4.159, p=0.010) and malnutrition before treatment (HR=2.360, 95% CI=1.228-4.535, p=0.010) and at the beginning of the last course of chemotherapy (HR=5.896, 95% CI=2.723-12.764, p<0.001) were independent predictors of OS (18). These two studies argue in favor of preventing weight loss after CRS. Even if patients are able to eat and drink the day after surgery with fast-track protocols [according to Enhanced Recovery After Surgery Guidelines (19)], parenteral nutrition remains useful especially in the case of a TPS in order to maintain body weight. In our series 77% of patients with a TPS had received parenteral nutrition at hospital which continued at home. Glasgow et al. reported a 30-day re-admission rate of 30% in a series of 53 patients who had undergone an ileostomy (in the context of ovarian cancer in 86%). The reasons for re-admission were dehydration (38%) and acute renal failure (33%) (20). Villafranca et al. developed a protocol for patients with a stoma to detect and manage dehydration and renal dysfunction. This protocol emphasizes that surgeons and nutritionists should be included in the multidisciplinary team (21). However, even with this protocol, the authors reported a high rate of re-admission (28%) and found no reason to explain this high rate in 57% of cases. These studies argued in favor of stomal closure as soon as possible (21).

The timing of stomal closure continues to fuel debate. In the literature on colorectal cancer, morbidities related to stomal closure increased from 5% to 30% from 10 days to 3 months after reverse surgery (22). Lasithiotakis et al. proposed early stomal closure in their series with a median delay of 8 days. The 26 patients were monitored on day 6 to check the integrity of the anastomosis and were randomized into two groups: early closure group (day 8) and late closure group (median 57 days) (23). Earlier closure was not found to be significantly superior in terms of ease of reversal and costs of stomal care. No difference was found in terms of morbidities (23). The timing of TPS closure is crucial due to the potential impact on delaying the start of adjuvant chemotherapy. Indeed, in an ancillary study of the Gynaecologic Oncology Group protocol 218, Tewari et al. showed in a multivariate analysis that a delay was predictive of poorer OS (p<0.001) and a delay of more than 25 days increased the risk of death in patients with stage IV ovarian cancer (24). Anastomotic leakage after CRS with bowel resection increases this delay, as shown by Kalogera at al. (12). TPS may reduce the risk of anastomotic leakage but exerts an impact on adherence to the chemotherapy schedule, as shown in our study (22% of the TPS group complied with the chemotherapy schedule versus 58% in the no-TPS group, p<0.05). Moreover, anastomotic leakage and a TPS created due to a defect at intraoperative testing should be a contraindication to adjuvant therapy with bevacizumab. This drug is administered as maintenance adjuvant therapy because it vas found to significantly increase progression-free survival in these patients (25, 26). Our study found that a TPS significantly increased the risk of recurrence (p=0.002). Tseng et al. found no impact of a TPS on progression-free survival or OS. However, our series of 99 patients only concerned CRS (CC0 according to Sugarbaker's classification) whereas Tseng et al. reported on more patients (n=331) but only 167 had undergone a CC0 resection and residual disease was the main prognostic factor (9). On the other hand, our study was a retrospective study with a small number of patients in the TPS group and 59% of the 99 patients had received neo-adjuvant chemotherapy.

In conclusion, the nutritional status of patients with advanced ovarian cancer should be adequate before CRS. A TPS should be created in selected cases and should be managed at hospital and at home using parenteral nutrition and adequate hydration to reduce the risk of re-admission. The TPS should be closed as soon as possible (before 10 days) so that the beginning of chemotherapy is not delayed. If a TPS is created due to an anastomotic defect at intraoperative testing, TPS closure should be organized later without delaying the beginning of chemotherapy and without compromising the chemotherapy schedule. Management of the TPS is crucial in order to administer adjuvant chemotherapy as planned and to adhere to the timing of the treatment schedule.