Ascites and malnutrition are predictive factors for incomplete cytoreductive surgery for peritoneal carcinomatosis from gastric cancer

Transcription

1 The American Journal of Surgery (2013) 205, Clinical Surgery Ascites and malnutrition are predictive factors for incomplete cytoreductive surgery for peritoneal carcinomatosis from gastric cancer Emmanuel I. Benizri, M.D., Ph.D.*, Jean-Marc Bereder, M.D., Amine Rahili, M.D., Jean-Louis Bernard, M.D., Daniel Benchimol, M.D. Department of General Surgery and Digestive Cancerology, Centre Hospitalier Universitaire de Nice, Hôpital de l Archet 2, 151 Route de Saint Antoine de Ginestière, BP 3079, Nice cedex 3, France KEYWORDS: Gastric cancer; Peritoneal carcinomatosis; Cytoreductive surgery; Intraperitoneal chemotherapy Abstract BACKGROUND: Prognosis in peritoneal carcinomatosis from gastric cancer has improved with cytoreductive surgery (CS) and hyperthermic intraperitoneal chemotherapy. The aim of this study was to identify predictive factors for incomplete CS. METHODS: Forty-five patients undergoing laparotomy for gastric cancer with peritoneal carcinomatosis were prospectively included from January 2000 to December In case of optimal CS, patients (n 5 14) received hyperthermic intraperitoneal chemotherapy. Otherwise, the laparotomy was closed or a palliative procedure was performed if necessary. All preoperative data were compared between the 2 groups. RESULTS: Ascites (hazard ratio,.09; 95% confidence interval, ; P ) and nutritional status evaluated by the prognostic nutrition index (hazard ratio,.11; 95% confidence interval, ; P 5.027) were independent predictive factors for incomplete CS. CONCLUSIONS: The selection of patients for CS plus hyperthermic intraperitoneal chemotherapy should include the assessment of nutritional status and the detection of an ascites. Ó 2013 Elsevier Inc. All rights reserved. Peritoneal carcinomatosis (PC) is detected in 10% to 20% of patients with gastric cancer during the initial diagnosis. 1 Moreover, peritoneal recurrence may occur in up to 60% of patients even after curative resection. 2 In the European multicenter Evolution of Peritoneal Carcinomatosis 1 study, prognosis in PC from gastric cancer was very poor, with a median survival time of 3.1 months. 3 Thus, PC is mostly considered a terminal condition in The authors declare no conflicts of interest. * Corresponding author. Tel.: ; fax: address: benizri.e@chu-nice.fr Manuscript received January 11, 2012; revised manuscript May 3, 2012 patients with gastric cancer, and the current standard treatment is palliative systemic chemotherapy. However, a new multimodal therapy linking cytoreductive surgery (CS) and hyperthermic intraperitoneal chemotherapy (HIPEC) has been developed during the past 20 years. This procedure has demonstrated promising results in numerous types of PC, including gastric cancer. Yonemura et al 4 reported a median survival time after complete cytoreduction of 15.5 months. In a multicenter study, similar results were reported, with an overall median survival time of 15 months. 5 In addition, a randomized trial comparing CS alone with CS plus HIPEC recently demonstrated the superiority of the combined treatment. 6 Thus, this procedure could become a standard treatment for selected patients /$ - see front matter Ó 2013 Elsevier Inc. All rights reserved.

2 E.I. Benizri et al. Gastric peritoneal carcinomatosis 669 In all studies, the quality of the surgical cytoreduction, measured by the completeness of cytoreduction (CC) score, is the most decisive prognostic factor. 4 9 This is not surprising, because the depth of penetration of HIPEC does not exceed 2 mm. 10 Thus, if CS is not optimal, HIPEC has no rationale and should not be performed. The completeness of PC resectability is directly correlated with the extent of PC. 11 Computed tomography (CT) is usually considered the best examination for staging PC. Nevertheless, the extent of peritoneal disease is often underestimated, and surgery is required to achieve a precise exploration. 12 Therefore, the challenge is to identify before surgery those patients in whom CS may be incomplete, to avoid unnecessary laparotomy. The main purpose of our study was to find predictive factors for incomplete CS in patients with PC arising from a gastric origin. Secondary aims were to evaluate the efficacy and safety of this procedure. Methods From January 1999 to December 2010, 45 patients with PC from gastric cancer that was apparently resectable were included in this study. All patients underwent laparotomy, with the intention of performing HIPEC. Fourteen patients (31%) had optimal CS according to CC score (0 or 1). HIPEC was performed for this group of patients (the CS- HIPEC group). For the other 31 patients (69%), PC was considered unresectable; that is, CS would have left deposits exceeding 2.5 mm (CC score of 2). For these patients, forming the no-hipec group, the laparotomy was closed or palliative surgery was performed if necessary. All patients were included in a prospective database and provided written informed consent. The study protocol was approved by our institutional ethics committee. Inclusion criteria All primary gastric cancers were confirmed by biopsy. Staging systematically included thoracoabdominal CT with oral and intravenous contrast agents and endoscopic ultrasonography. From 2004, positron emission tomography (PET) was additionally performed. Anesthetic evaluation, echocardiography, and spirometry were performed on all patients. Patients were then selected preoperatively according to the criteria defined by the Peritoneal Surface Malignancy Group adapted for gastric cancer 13,14 : (1) PC from gastric cancer; (2) aged,65 years and good general status (World Health Organization performance status,2); (3) no extraabdominal disease; (4) no multiple, diffuse, and huge tumoral peritoneal deposits on CT; (5) no evidence of intestinal obstruction or involvement; (6) small-volume disease in the gastrohepatic ligament; (7) no evidence of biliary or ureteral obstruction; and (8) no massive and total abdominal involvement on clinical examination. For borderline cases, laparoscopy was performed preoperatively. Surgical procedures Surgery was performed through a large midline incision from xiphoid to pubis. After an extensive and complete exploration of the abdominal cavity, the extent of the PC was calculated for each patient using the peritoneal cancer index (PCI), as described by Sugarbaker et al. 15 If the PC was resectable (CS-HIPEC group), CS was performed by resection or destruction by electrovaporation of all peritoneal lesions. This consists of a series of peritonectomy procedures, visceral resections, or destruction by electrovaporation of all peritoneal implants. Their use depends on the volume, distribution, and depth of invasion of the nodules within the peritoneal surfaces. In case of synchronous PC, a gastrectomy was performed with a modified radical lymphadenectomy (D1.5). Splenectomy with or without distal pancreatectomy was associated in case of locoregional involvement or PC in this area. At the end of this surgical procedure, patients were classified according to CC score as follows: CC score 0 5 no residual macroscopic nodules; CC score 1 5 residual tumor,2.5 mm. 15 This subjective measure was based on a macroscopic visual assessment of any residual lesions. Only these patients were considered to have optimal CS and were eligible for the HIPEC procedure. Mitomycin C was administered into the peritoneal cavity, using the coliseum technique with a Thompson self-retaining retractor. 16 The dose was 12.5 mg/m 2 for men and 10 mg/m 2 for women, in 2 L/m 2 of 1.5% dextrose peritoneal dialysis solution. The heated perfusion solution was infused at a rate of 1 L/min. A heat exchanger kept the intraperitoneal temperature at 42 Cto43 C for 60 to 90 minutes. Anastomoses were usually performed after HIPEC, and 2 suction drains were placed in the peritoneal cavity. Despite a strict selection of patients, multiple causes of incomplete resectability were discovered during surgical exploration. For these patients (the no-hipec group), there was no interest in completing the full combined procedure. Finally, the laparotomy was closed, possibly after a palliative procedure. End points and evaluation criteria The primary aim of this study was to find preoperative factors allowing us to identify patients for whom CS may be incomplete. The parameters studied to achieve this objective were (1) history: completion of preoperative systemic chemotherapy and synchronous or metachronous PC; (2) clinical: sex, body mass index,18 kg/m 2, and presence of ascites before surgery (regardless of quantity); (3) biologic: an increase in the carcinoembryonic antigen level (.5 ng/ml), malnutrition as determined by an assessment of the nutritional status by the prognostic nutrition

3 670 The American Journal of Surgery, Vol 205, No 6, June 2013 index (PNI), calculated as 10! serum albumin (g/dl) 1.005! total lymphocyte count (per mm 3 ) 17 (a PNI value,45 was regarded as indicating moderate to severe malnutrition); (4) imaging: presence of direct or indirect signs related to PC on PET and CT ( 18 F-fluorodeoxyglucose uptake, ascites, peritoneal nodules or bulky mass lesions, peritoneal thickening, mesenteric effacement, and luminal narrowing); (5) histologic: presence of signet ring cells; (6) progression of PC under systemic chemotherapy on the basis of clinical examination, carcinoembryonic antigen level, and imaging; and (7) laparoscopic exploration. Another objective was to analyze the morbidity and mortality observed for all patients. Mortality and morbidity were defined, respectively, as death or medical and surgical postoperative complications occurring,30 days after surgery or until hospital discharge. Finally, we studied the oncologic results by measuring the overall survival in the 2 groups. All patients were followed every 6 months with clinical examinations, thoracoabdominal CT, and carcinoembryonic antigen measurement. Survival was defined from the time of the surgical procedure to death or to the last follow-up date. Statistical analysis Patients were prospectively registered, but the analysis of the assessment criteria was done retrospectively. Mann- Whitney tests were performed to determine intergroup differences between mean values. Bivariate analysis was performed to find correlations between quantitative variables. We used Fisher s exact test to examine the associations between each categorical variable. Variables with P values,.20 on univariate analysis were included in multivariate analysis to identify independent predictive factors for optimal CS. These variables were entered into a logistic regression model. Survival analysis was performed using the Kaplan-Meier method and compared using the log-rank test. No patient was excluded from survival analysis, and no data were missing. P values,.05 were considered significant. Statistical analysis was performed using R version (R Foundation for Statistical Computing, Vienna, Austria). Results Predictive factors for suboptimal cytoreductive surgery A comparison of the preoperative CS-HIPEC and no- HIPEC groups is given in Table 1. On univariate analysis, the presence of ascites before surgery (P ) and nutritional status measured using the PNI (P ) were the only 2 predictive factors for suboptimal CS (Fig. 1). The presence of either of these 2 factors allowed us to predict suboptimal CS in 90% of the patients (28 of 31) in the no-hipec group. On multivariate analysis, these 2 factors were also independent risk factors for incomplete CS (for ascites: hazard ratio,.09; 95% confidence interval, ; P ; for PNI: hazard ratio,.11; 95% confidence interval, ; P 5.027). Analysis of the cytoreductive surgery-hyperthermic intraperitoneal chemotherapy group In the CS-HIPEC group, the mean PCI was (range, 3 22). Eight procedures (57%) were considered CC score 0 cytoreduction, and 6 procedures (43%) were CC score 1 cytoreduction. Procedure-related mortality was null. Five patients (36%) had R1 complication (Table 2). The mean length of hospital stay was days (range, 12 90). The median overall survival duration was 18 months. The survival curve for the CS-HIPEC group is shown in Fig. 2. Analysis of the no-hyperthermic intraperitoneal chemotherapy group In the no-hipec group, the mean PCI was (range, 8 26). Causes of nonresectability in the no-hipec group are reported in Fig. 3. The main cause was the diffuse extent of the PC throughout the abdominal cavity. Palliative surgery was performed in 8 patients (26%). These procedures included digestive bypass (n 5 3), distal gastrectomy (n 5 2), jejunostomy tube (n 5 2), and small bowel resection. Postoperative mortality was 6%. Eight patients (26%) had R1 complication (Table 2). The mean length of hospital stay was days (range, 4 34). The median overall survival duration was significantly lower than that of the CS-HIPEC group: 6 vs 18 months (P ). The survival curve for the no-hipec group is shown in Fig. 4. Comments There is no standard treatment for PC from gastric cancer. Several studies have shown that CS plus HIPEC may significantly improve prognosis in selected patients for whom optimal CS (CC scores of 0 or 1) can be obtained. 4 9 In our peritoneal cancer center, we follow this indication, and we do not perform HIPEC for CC score 2 surgery. Despite compliance with the recommendations of the Peritoneal Surface Malignancy Group adapted for gastric cancer, accurate preoperative staging remains difficult, and the indication to perform HIPEC is frequently based on intraoperative exploration. 13,14 Thus, only 31% of patients had optimal CS and were treated with HIPEC in our study. These results are slightly lower than those reported by Glehen et al 8 or Yonemura et al 4 with, respectively, 51% and 43% of CC score 0 and CC score 1 CS. But in

4 E.I. Benizri et al. Gastric peritoneal carcinomatosis 671 Table 1 Identification by univariate analysis of preoperative predictive factors of suboptimal CS for gastric PC Variable CS-HIPEC group (n 5 14) No-HIPEC group (n 5 31) P Sex 9/14 (65%) 17/31 (55%).74* Male 5/14 (35%) 14/31 (45%) Female 9/14 (65%) 17/31 (55%) Age (y) BMI, 18 kg/m 2 3/14 (21%) 10/31 (32%).72* PNI Synchronous PC 12/14 (86%) 24/31 (78%).70* Preoperative chemotherapy 13/14 (93%) 25/31 (80%).40* Progression with chemotherapy 2/13 (15%) 6/25 (24%).69* Abnormal results on CT 8/14 (57%) 18/31 (58%).96* Abnormal results on PET 4/8 (50%) 13/21 (62%).69* Carcinoembryonic antigen level.5 6/14 (43%) 16/31 (52%).75* Ascites 2/14 (14%) 20/31 (65%).0031* Signet ring cells 7/14 (50%) 21/31 (68%).32* Preoperative laparoscopy 4/14 (29%) 16/31 (52%).19 Data are expressed as number (percentage) or as mean 6 SD. Significant P values are in boldface type. BMI 5 body mass index; CS 5 cytoreductive surgery; CT 5 computed tomography; HIPEC 5 hyperthermic intraperitoneal chemotherapy; PC 5 peritoneal carcinomatosis; PET 5 positron emission tomography; PNI 5 prognostic nutrition index. *Fisher s exact test. Mann-Whitney test. their series, all patients underwent HIPEC, and some patients who had undergone only exploratory laparotomy were probably not included. To avoid unnecessary laparotomy, we looked for other factors that might be useful to select the right candidates for the combined treatment. Thus, we showed that 2 preoperative factors were independent risk factors of nonresectability: the persistence of ascites and nutritional status assessed using the PNI. The detection of ascites just before surgery is a predictive factor of suboptimal CS that has already been reported for colorectal PC but never for gastric cancer. 18 In an expert consensus document, Bozzetti et al 14 reported that only 33% of participants recommended the full procedure in case of ascites. Surprisingly, no study has ever shown that the presence of ascites was correlated with the extent of PC. Cancer-associated malnutrition is multifactorial: reduced nutritional intake, alterations in nutrient metabolism, and production of agents by the tumor (such as hormones or proinflammatory cytokines) are implicated in the pathophysiology of malnutrition. Nutritional status has already been correlated with surgical resectability rates, as in esophageal cancer. 19 Severe malnutrition is often an indicator of tumor extension. In our series, the mean PNI was 40 in the no-hipec group. This very low rate demonstrates severe malnutrition and reflects indirectly on the tumor extension. Figure 1 Prediction of performance of HIPEC by regression analysis of the PNI. The threshold value of the PNI is measured at 45. Gray area represents the confidence interval. Figure 2 Overall survival curve of the CS-HIPEC group (14 patients with gastric PC treated with optimal CS plus HIPEC). The median overall survival was 18 months.

5 672 The American Journal of Surgery, Vol 205, No 6, June 2013 Table 2 Variable Postoperative results for patients with gastric PC CS-HIPEC group No-HIPEC group Number of patients Postoperative mortality 0 2 Complications 5 (36%) 8 (26%) Digestive fistulas 2 1 Occlusion 2 2 Pulmonary embolism 1 0 Septic Shock 1 0 Catheter infection 0 1 Pneumopathy 0 1 Jaundice 0 1 Urinary infection 0 2 Number of reoperations 2 (14%) 0 Hospital stay (days) Data are expressed as number (percentage) or as mean 6 SD. CS 5 cytoreductive surgery; HIPEC 5 hyperthermic intraperitoneal chemotherapy; PC 5 peritoneal carcinomatosis. Yonemura et al 11 clearly established that CS was correlated with the extent of PC. For gastric cancer, experts have recommended full-procedure CS plus HIPEC with PCI % In this situation, the probability of complete macroscopic CS is highest. However, we confirm that conventional imaging techniques underestimate the extent of intraperitoneal spread. Indeed, we showed that the presence of an abnormality on CT was not a predictive factor for optimal CS. Similarly, Esquivel et al 20 showed that the extent of PC on the basis of CT is inaccurate, and its impact on the management of patients is modest. PET appears to be a promising alternative imaging modality with potentially higher accuracy in identifying peritoneal lesions and the overall extent of PC. Pfannenberg et al 21 reported that the combination of PET and CT was accurate in estimating the extent of carcinomatosis. In practice, PET usually fails to detect subcentimeter lesions, whereas the presence of such lesions is conclusive in determining the resectability of patients. In our study, 18 F-fluorodeoxyglucose uptake did not emerge as a predictive factor for optimal CS. Figure 3 Causes of nonresectability in the no-hipec group (31 patients with gastric PC for whom CS could not be performed). Figure 4 Overall survival curve of the no-hipec group (31 patients with gastric PC for whom CS could not be performed). The median overall survival was 6 months. Some teams propose laparoscopy for the initial staging of gastric PC, with the hope of avoiding unnecessary laparotomy (detection of superficial liver metastases or diffuse extension of the small bowel). 22 Valle and Garofalo 23 demonstrated that laparoscopic PCI assessment was possible in 98% of patients. Moreover, laparoscopy is associated with a shorter recovery and less pain than laparotomy. However, in our practice, laparoscopy is sometimes difficult to perform (iterative surgery, adhesions) and is not very accurate (assessment of the retroperitoneal space and the posterior segments of the liver). In this study, when laparoscopy was performed, it did not accurately predict the extent of PC. There is also the potential risk for metastasis on port trocars, which must be taken into account. For all these reasons, experts believe that laparoscopy is probably useful, but not fundamental. 24 In the no-hipec group, surgical exploration was still useful for 26% of patients who underwent palliative surgery. Postoperative mortality was 6%, but this did not affect patients who had palliation. This involved 2 patients with massive PC, who died of occlusion. The 2 groups were not comparable, because their treatment and their tumor extent were different. Under these conditions, overall survival was better in the CS- HIPEC than in the no-hipec group. With a median overall survival duration of 18 months, the oncologic results of the combined procedure are similar to those reported in the literature after optimal CS. 4 9 These results suggest that optimal CS plus HIPEC slightly improves the survival rate in selected patients. Gastric PC should not be considered a terminal stage, as evidenced by 2 patients in our series who were alive 5 years after surgery in the CS-HIPEC group compared with no patient in the no-hipec group. Moreover, although we did not assess the quality of life in this study, other investigators have reported improved quality of life after CS and HIPEC in long-term survivors. 25 In conclusion, the selection of patients with gastric PC eligible for the complete procedure linking CS with HIPEC

6 E.I. Benizri et al. Gastric peritoneal carcinomatosis 673 remains difficult. This study suggests that we must be very careful with malnourished patients with or without the presence of ascites. These 2 factors are an indication of advanced disease and are predictive of nonresectability. Moreover, the present series suggests that optimal CS plus HIPEC may improve survival with acceptable morbidity. References 1. Bando E, Yonemura Y, Takeshita Y, et al. Intraoperative lavage for cytological examination in 1,297 patients with gastric carcinoma. Am J Surg 1999;178: Sugarbaker PH, Yonemura Y. Clinical pathway for the management of resectable gastric cancer with peritoneal seeding: best palliation with a ray of hope for cure. Oncology 2000;58: Sadeghi B, Arvieux C, Glehen O, et al. Peritoneal carcinomatosis from non-gynecologic malignancies: results of the EVOCAPE 1 multicentric prospective study. Cancer 2000;88: Yonemura Y, Kawamura T, Bandou E, et al. Treatment of peritoneal dissemination from gastric cancer by peritonectomy and chemohyperthermic peritoneal perfusion. Br J Surg 2005;92: Glehen O, Gilly FN, Arvieux C, et al. Peritoneal carcinomatosis from gastric cancer: a multi-institutional study of 159 patients treated by cytoreductive surgery combined with perioperative intraperitoneal chemotherapy. Ann Surg Oncol 2010;17: Yang XJ, Huang CQ, Suo T, et al. Cytoreductive surgery and hyperthermic intraperitoneal chemotherapy improves survival of patients with peritoneal carcinomatosis from gastric cancer: final results of a phase III randomized clinical trial. Ann Surg Oncol 2011;18: Scaringi S, Kianmanesh R, Sabate JM, et al. Advanced gastric cancer with or without peritoneal carcinomatosis treated with hyperthermic intraperitoneal chemotherapy: a single western center experience. Eur J Surg Oncol 2008;34: Glehen O, Schreiber V, Cotte E, et al. Cytoreductive surgery and intraperitoneal chemohyperthermia for peritoneal carcinomatosis arising from gastric cancer. Arch Surg 2004;139: Glehen O, Gilly FN, Boutitie F, et al. Toward curative treatment of peritoneal carcinomatosis from nonovarian origin by cytoreductive surgery combined with perioperative intraperitoneal chemotherapy: a multi-institutional study of 1,290 patients. Cancer 2010;116: Los G, Mutsaers PH, van der Vijgh WJ, et al. Direct diffusion of cisplatinum in intraperitoneal rat tumors after intraperitoneal chemotherapy: a comparison with systemic chemotherapy. Cancer Res 1989;48: Yonemura Y, Bandou E, Kawamura T, et al. Quantitative prognostic indicators of peritoneal dissemination of gastric cancer. Eur J Surg Oncol 2006;32: Cotte E, Passot G, Gilly FN, et al. Selection of patients and staging of peritoneal surface malignancies. World J Gastrointest Oncol 2010;2: Esquivel J, Elias D, Baratti D, et al. Consensus statement on the loco regional treatment of colorectal cancer with peritoneal dissemination. J Surg Oncol 2008;98: Bozzetti F, Yu W, Baratti D, et al. Locoregional treatment of peritoneal carcinomatosis from gastric cancer. J Surg Oncol 2008;98: Sugarbaker PH, Averbach AM, Jacquet P, et al. Hyperthermic intraoperative intreperitoneal chemotherapy (HIIC) with mitomycin C. Surg Technol Int 1996;5: Beaujard AC, Glehen O, Caillot JL, et al. Intraperitoneal chemohyperthermia with mitomycin C for digestive tract cancer patients with peritoneal carcinomatosis. Cancer 2000;88: Onodera T, Goseki N, Kosaki G. Prognostic nutritional index in gastrointestinal surgery of malnourished cancer patients. Nippon Geka Gakkai Zasshi 1984;85: Elias D, Benizri E, Vernerey D, et al. Preoperative criteria of incomplete resectability of peritoneal carcinomatosis from non-appendiceal colorectal carcinoma. Gastroenterol Clin Biol 2005;29: Belghiti J, Langonnet F, Bourstyn E, Fekete F. Surgical implications of malnutrition and immunodeficiency in patients with carcinoma of the oesophagus. Br J Surg 1983;70: Esquivel J, Chua TC, Stojadinovic A, et al. Accuracy and clinical relevance of computed tomography scan interpretation of peritoneal cancer index in colorectal cancer peritoneal carcinomatosis: a multiinstitutional study. J Surg Oncol 2010;102: Pfannenberg C, Königsrainer I, Aschoff P, et al. 18 F-FDG-PET/CT to select patients with peritoneal carcinomatosis for cytoreductive surgery and hyperthermic intraperitoneal chemotherapy. Ann Surg Oncol 2009;16: Leake PA, Cardoso R, Seevaratnam R, et al. A systematic review of the accuracy and indications for diagnostic laparoscopy prior to curative-intent resection of gastric cancer. Gastric Cancer 2011;15- (suppl):s Valle M, Garofalo A. Laparoscopic staging of peritoneal surface malignancies. Eur J Surg Oncol 2006;32: Yan TD, Morris DL, Shigeki K, et al. Preoperative investigations in the management of peritoneal surface malignancy with cytoreductive surgery and perioperative intraperitoneal chemotherapy: experts consensus statement. J Surg Oncol 2008;15: Glockzin G, Schlitt HJ, Piso P. Peritoneal carcinomatosis: patients selection, perioperative complications and quality of life related to cytoreductive surgery and hyperthermic intraperitoneal chemotherapy. World J Surg Oncol 2009;7:5.