Pulmonary morbidity of diaphragmatic surgery for stage III/IV ovarian cancer

Transcription

1 DOI: /j x Gynaecological oncology Pulmonary morbidity of diaphragmatic for stage III/IV ovarian cancer E Chéreau, a M Ballester, a F Selle, b A Cortez, c C Pomel, d,e E Darai, a R Rouzier a a Department of Gynecology-Obstetrics b Department of Medical Oncology and c Department of Pathology, Hôpital Tenon, Assistance Publique des Hôpitaux de Paris, CancerEst, Université Pierre et Marie Curie Paris, Paris, France d Department of Surgery, Centre Jean Perrin, Clermont Ferrand, France e Saint Bartholomew Hospital, London, UK Correspondence: Dr E Chéreau, Service de Gynécologie-Obstétrique, Hôpital Tenon, 4 rue de la Chine, Paris, France. elisabeth.chereau@gmail.com Accepted 31 March Published Online 14 May Objective To determine the morbidity of diaphragmatic peritonectomy. Design Prospective cohort study. Setting A Gynecology Department of a University Hospital. Population From 2005 to 2007, thirty-seven consecutive patients underwent for stage IIIC or IV ovarian cancer. Methods Patients were separated into a diaphragmatic group (n = 18) and a control group (n = 19). Diaphragmatic may consist of coagulation, stripping or muscle resection. Main outcome measures Postoperative course and outcome were analysed. Results Patients in group 1 (diaphragmatic ) underwent more intestinal resection (89% versus 37%, P = 0.01) and pelvic (94% versus 63%, P = 0.02) or para-aortic lymphadenectomy (94% versus 53%, P = 0.04). Neither the mean estimated blood loss (960 ml versus 909 ml) nor the rates of intra-operative blood transfusion (11 versus 9) were significantly different between the two groups. The mean operative time was higher in group 1 (480 minutes versus 316 minutes, P < 0.05). There were thirteen postoperative complications in group 1 and eight in group 2 (P = 0.065). In group 1, the main complication was pleural effusion (seven cases): four patients required secondary pleural drainage, two required only pleural puncture and one had both procedures. There were more complete cytoreduction in group 1 than in group 2 (89% versus 63%, P = 0.068). Conclusions Diaphragm peritonectomies and resections are an effective way to cytoreduce diaphragm carcinomatosis and increase the rate of optimal debulking. Such procedures frequently result in pleural effusion, but with no long-term morbidity. Keywords Diaphragm, morbidity, ovarian cancer, pleural effusion. Please cite this paper as: Chéreau E, Ballester M, Selle F, Cortez A, Pomel C, Darai E, Rouzier R. Pulmonary morbidity of diaphragmatic for stage III/IV ovarian cancer. BJOG 2009;116: Introduction Both retrospective and prospective reports have demonstrated that optimal cytoreduction for advanced epithelial ovarian cancer is the cornerstone of effective treatment. 1 In patients with stage IIIC/IV disease, diaphragm involvement is very common. Bulky diaphragm disease has been reported as one of the most commonly reported sites of disease that lead to suboptimal cytoreduction and therefore to a lower rate of survival. Surgical procedures already exist to treat diaphragm disease; they typically increase the rate of complete and optimal debulking and yield better survival compared to patients with residual disease found only on the diaphragm. 2 Given this, incorporation of extensive upper abdominal procedure has become necessary to achieve optimal cytoreduction. As most reports are made by highly specialised centres, reporting the morbidity and caveats of diaphragmatic in early experience is necessary to speed the implementation of such extensive upper abdominal procedures. 3,4 Moreover, the lack of a control group in most previous reports may have limited the interpretation of results. In our institution, diaphragmatic was introduced relatively recently. The aim of this study was to determine the morbidity of diaphragmatic (resection/stripping) in comparison with a control group without such ª 2009 The Authors Journal compilation ª RCOG 2009 BJOG An International Journal of Obstetrics and Gynaecology

2 Pulmonary morbidity of diaphragmatic Methods During the period from 2005 to 2007, thirty-seven consecutive patients underwent for stage IIIC or IV ovarian cancer in Tenon hospital. This study population was divided into two groups: eighteen had diaphragmatic peritoneal during the complete cytoreduction (group 1) and nineteen did not have diaphragmatic (group 2). We considered both adjuvant and neo-adjuvant chemotherapy because the purpose of this study was to determine the morbidity and feasibility of the surgical act in isolation. The decision of whether or not to perform diaphragmatic was dependent on the extent of disease using the Peritoneal Cancer Index (PCI) score, 5 and the surgeon s choice (age of the patient, extent of the carcinosis, number of procedures required ). In our practice, extensive radical procedures were only performed and combined (multiple bowel resection, splenectomy) if they were likely to result in a complete cytoreduction. Individual records of all patients were reviewed and analysed. Patient and tumour characteristics, intra-operative findings, postoperative course and outcome were analysed. Diaphragmatic may consist of stripping, muscle resection or electro-evaporation. Residual disease was reported as complete (no residual disease) or not. The size of the diaphragmatic resection was also studied. Intraoperative units of blood transfusion, operative time and length of hospitalisation (including length of time in the intensive care unit) were all recorded. We graded complications according to the Memorial Sloan-Kettering Cancer Center surgical secondary events grading system, that is: 0: No events observed within 30 days postoperatively; 1: Use of oral medications, bedside interventions to treat an event; 2: Use of intravenous medications, total parenteral nutrition, enteral nutrition or blood transfusion to treat an event; 3: Interventional radiology, therapeutic endoscopy, intubation or operation required to treat an event; 4: Residual and lasting disability requiring major rehabilitation or organ resection; 5: Event resulting in death of patient. The kind and duration of drainage, secondary pleural puncture requirement and postoperative complications were also studied. Surgical Procedures 6 The abdomen was opened from xiphoid to pubis. The first step in the surgical course was sus-mesocolic because we considered this as a limiting step for complete cytoreduction. The complete mobilisation of the liver was performed by resection of hepatic round and falciform ligaments, as well as the sectioning of triangular ligaments and layers of coronary ligaments if carcinomatosis was not adhesive to the liver. These allowed us to explore the entire right and left diaphragm (including omental bursa). In the case of adhesions between the tumour and liver, the liver was not initially completely mobilised. We started the peritoneal stripping of the right hemi diaphragm: the peritoneum that constitutes the edge of the abdominal incision was stripped off the posterior rectus sheath. To secure this peritoneal layer, clamps were positioned approximately every 8 cm. This allowed traction to be achieved on the tumour specimen throughout the right upper quadrant. The dissection beneath the diaphragm was performed with electro eventually using the ball-tip to avoid diaphragm perforation. In cases where adhesions between the tumour and the liver occurred, the stripping of the tumour from the muscular surface of the diaphragm continued until the bare area of the liver was encountered. At this point, the tumour on the anterior surface of the liver was electro-evaporated until the liver surface was encountered. With both blunt and ball-tip electrosurgical dissection, the tumour was lifted off the dome of the liver by moving through or beneath Glisson s capsule. Isolated patches of tumour on the liver surface were electro-evaporated. Tumour from beneath the right hemidiaphragm, from the right subhepatic space, and from the surface of the liver forms a type of envelope as it was removed en bloc. The dissection was simplified greatly if the tumour specimen could be maintained intact. We sometimes used digestive clamps to help peritoneal stripping during diaphragm ampliation. In cases of major invasion of the diaphragm near hepatic veins, mobilisation of the liver and opening of the diaphragm, may allow isolation and clamping of the suprahepatic and infrahepatic vena cava, hepatic vein and the hilum to avoid major bleeding. The peritoneal stripping of the left hemidiaphragm was performed similarly to the right side: the peritoneum was stripped away from the left posterior rectus sheath to begin the peritonectomy in the left upper quadrant of the abdomen. In most cases, upper quadrant peritonectomy involved the stripping of all tissue from beneath the hemidiaphragm to expose diaphragmatic muscle (Figure 1A), the adrenal gland, the superior aspect of the pancreas on the left side and the cephalad half of Gerotta s fascia. When the tumour was densely adhesive to the tendinous mid-portion of the left or right hemidiaphragm, the peritoneum was freed laterally to the tendinous diaphragm. A digestive clamp was used to isolate the diaphragm and to perform the resection without opening the pleural space. When a pleural effraction occurred, the fibrous tissue infiltrated by the tumour was resected, usually requiring an elliptical excision of a portion of the hemidiaphragm. The defect in the diaphragm was closed with interrupted sutures. A chest tube was eventually placed according to the extent of the resection in cases of pleural effraction. Alternatively, the pneumothorax was aspirated with a Foley ª 2009 The Authors Journal compilation ª RCOG 2009 BJOG An International Journal of Obstetrics and Gynaecology 1063

3 Chéreau et al. (A) (B) Figure 1. (A) Operative aspect after surgical resection of right diaphragmatic peritoneum. (B) Liver mobilisation after resection of falciform ligaments and sectioning of triangular ligaments (arrow: section of left triangular ligament) and layers of coronary ligaments: the complete exploration (star: stomach) reveals carcinomatosis located on the omental bursa (double arrow). catheter placed through the diaphragm defect, which was closed without the pleural drain in place. Statistical analysis Data were analysed using the Fischer exact test, the chi-square test, the Student s t-test and the non-parametric Mann Whitney test if necessary. Differences were considered significant at a level of P < Results The median age of patients in group 1 (diaphragmatic ) was 51 years, whereas it was 64 years in group 2 (control group) (P < 0.05). Nearly, all the patients had stage IIIC disease (94 and 89% for groups 1 and 2 respectively), and most of them had a serous histological type (67 and 74%). There were no significant differences between the two groups in body mass index, stage, tumour grade, histological type or preoperative CA 125 level (Table 1). More patients in group 2 than in group 1 received neo-adjuvant chemotherapy (58% versus 22%, P = 0.045). Patients in group 1 had a more extensive surgical procedure. Whereas there were no significant differences in hysterectomy, salpingo-oophorectomy, omentectomy, appendectomy, splenectomy or cholecystectomy, patients in group 1 more often underwent intestinal resection (89% versus 37%, P = 0.01), pelvic (94% versus 63%, P = 0.02) or para-aortic lymphadenectomy (94% versus 53%, P = 0.04) and lesser omentum resection (39% versus 11%, P = 0.04). Table 1. Patients and tumor characteristics Group 1: diaphragmatic % Group 2: no diaphragmatic % P Median age <0.05 BMI Stage ns III C 17/ /19 89 ns IV 1/18 6 2/19 11 Tumor grade 1 2/ /19 11 ns 2 10/ / / /19 32 Histologic type Serous 12/ /19 74 ns Tubulo-papillar 2/ /19 11 Endometrioid 3/ /19 5 Others 1/18 6 2/19 11 Preoperative CA 125 level 4/15 27 ns / / / / / /18 11 >1500 6/18 33 Neo adjuvant chemotherapy 4/ / All patients had right diaphragmatic. Thirteen patients underwent a stripping of the right diaphragmatic peritoneum, while five patients had electro-evaporation. A 1064 ª 2009 The Authors Journal compilation ª RCOG 2009 BJOG An International Journal of Obstetrics and Gynaecology

4 Pulmonary morbidity of diaphragmatic resection of the tendinous part of the diaphragm was performed in seven cases. Eight patients (44%) also had left diaphragmatic : three had only coagulation, whereas five had stripping and three had muscle resection. There were seven entrances in the pleural cavity during stripping or muscle resection (six on the right side and one on the left side). At the end of the surgical course, there were more patients with complete cytoreduction in group 1 than there were in group 2 (89% n = 2 versus 63% n = 7), but the difference fell short of reaching statistical significance (P = 0.068; Table 2). In group 1, two patients did not have complete cytoreduction because of an extensive non resectable disease on small bowel mesentery. In group 2, in five cases, there were also an extensive non resectable disease on small bowel mesentery, one other patient had diaphragmatic carcinomatosis associate with caecum and sigmoid disease and her general status could not permitted multiple digestive resections. The last patient had incomplete cytoreduction explained by residual diaphragmatic disease because of an inextirpable carcinomatosis. The mean estimated blood loss (960 ml versus 909 ml) and the number of units of blood transfused (2 units versus 1.6 units) were Table 2. Surgical procedures performed Group 1 % Group 2 % P Standard Hysterectomy 16/ / ns Unilateral or bilateral 18/ / ns salpingo-oophorectomy Omentectomy 18/ /19 95 ns Intestinal resection 16/ / Appendectomy 8/ /19 37 ns Pelvic lymphadenectomy 17/ / Para-aortic lymphadenectomy 17/ / Extensive Splenectomy 2/ /19 0 ns Cholecystectomy 1/18 6 0/19 0 ns Lesser omentum resection 7/ / Diaphragmatic Right 18/ <0.01 Coagulation only 5 Stripping 13 + muscle resection 8 Left 8/ <0.01 Coagulation only 3 Stripping 5 + muscle resection 3 Pleural effraction 7 Size of the resection (mm) 101 Complete resection 16/ / not significantly different between the two groups and neither was the percentage of patients who received intraoperative blood transfusion (11 versus 9, ns). The mean operative time was higher in group 1 (480 minutes versus 316 minutes, P < 0.05) (Table 3). There were thirteen postoperative complications in group 1 and eight in group 2 (P = 0.065). In grading by the MSKCC score, however, there were no differences noted between the two groups (Table 4). There were more pulmonary complications in group 1 because of the diaphragmatic. The main complication was pleural effusion. In the seven patients with pleural effusion, four required secondary pleural drainage, two only pleural puncture and one had both procedures. Of note, out of the seven patients with diaphragm resection, three needed a secondary drainage or puncture. Conversely, four patients who needed thoracocentesis or pleural puncture had no entrance made into the pleural cavity during diaphragmatic (not significant [ns]). A pneumothorax occurred after a pleural puncture in one patient. There were two pulmonary embolisms in group 1 and none in group 2. None of these patients had compression stockings for the prevention of venous thromboembolism. There were no differences between the two groups in the number of rehospitalisations or surgical re-interventions. There was no long-term morbidity after discharge of hospitalisation and none of the patients presented remaining thoracic pain or respiratory deficiency. The mean duration time between the surgical course and the beginning of chemotherapy was 36 days in group 1 versus 27 in group 2 (ns). Table 3. Postoperative complications Group 1 Group 2 P Patients with one or more postoperative complication Details of complications Fever/infection 5 4 Digestive fistula 2 0 Urinary fistula 0 1 Lymphocyst 3 0 Patients with one or more pulmonary complications Pulmonary embolism 2 0 Pleural effusion 7 1 Pneumothorax 1 0 Secondary drainage 5 0 Pleural puncture 3 0 Re-hospitalisation 5 2 ns Iterative 3 3 ns ª 2009 The Authors Journal compilation ª RCOG 2009 BJOG An International Journal of Obstetrics and Gynaecology 1065

5 Chéreau et al. Table 4. Surgical outcome Discussion Group 1 Group 2 P Estimated blood loss Mean (ml) ns Units of blood transfused No transfusion 6 8 Transfusion 11 9 Mean (units) ns Operative time <480 minutes >480 minutes 5 1 Mean (minutes) <0.05 Complications Grade Grade Length of hospitalisation < > Mean (days) <0.05 Length of hospitalisation in intensive care unit None Mean (days) ns Day of perfusion discharge ns Mean morphine used (mg) ns Mean time between and chemotherapy (days) ns For advanced ovarian cancer cases, the residual disease after is the most relevant determining factor for survival. 7 Given this, the aim of ovarian cancer is to obtain time complete cytoreduction. In our series, diaphragmatic peritoneum involvement was present in 25 cases. We note that this is in accordance with previous reports. 1 Surgical treatment of diaphragm disease is an important part of complete cytoreduction. 3,4 We demonstrate in this report that the main morbidity is pulmonary. However, we did not observe complications rating greater than grade 3 according to the Memorial Sloan-Kettering Cancer Center surgical secondary events grading system. In this study, diaphragmatic increased operative time, length of hospitalisation and pulmonary complications. However, there was neither increase in general postoperative complications nor in the other parameters that we studied. The parameters studied here were blood loss, units of blood transfused, length of hospitalisation in the intensive care unit, rehospitalisation or reoperation. In our study, the main complication was pleural effusion. It is important to note that surgical procedures (coagulation, stripping or muscle resection) were chosen according to the case-specific invasion of the diaphragm by the carcinomatosis. Opening of the pleural cavity was typically required because the carcinomatosis was infiltrative and full-thickness resection was required to obtain complete resection of diaphragmatic disease. In our experience, entrance in the pleural cavity during the surgical course increased the number of postoperative pleural punctures or drainages (two drainages and two punctures in the seven patients with pleural opening). At the end of the study period, we decided to systematically place a chest tube in anticipation of pleural opening. This approach remains to be evaluated. In his study, Eisenhauer 8 has reported that the only predictive factor for pleural effusion after diaphragmatic was liver mobilisation. In our study, pleural effusion occurred in 38% of the patients who underwent diaphragmatic. In Table 5, we summarise studies focusing on diaphragmatic in stage IIIC/IV ovarian cancers. In these series, the rates of postoperative pleural effusion ranged from 10 to 59%. Data from this table suggests that the rate of postoperative pleural effusion was inversely correlated with the number of chest tubes placed during the. This question should be prospectively addressed in future work. In his study, Eisenhauer 8 concluded that systematic pleural drainage was not justified during the time when the pleural space was opened. Indeed, even if a high rate of patients needed secondary pleural puncture or drainage, there were no major complications in the postoperative course. Studying Eisenhauer s work, however, only 40% of patients had extensive including pelvic and para-aortic lymphadenectomy. Of note, the percentage of diaphragmatic was greater in our study than in most previously published studies (from 14 to 50%). The high rate of pleural effusion in the absence of drainage may be explained by technical differences such as complete liver mobilisation and extent with almost systematic para-aortic and pelvic lymphadenectomy. We believe that complete liver mobilisation is necessary to ensure complete exploration and complete cytoreduction; as shown in Figure 1B, carcinomatosis of the omental bursa cannot be identified without complete liver mobilisation. To complete the education to ovarian cancer, a rotation in a department of hepatic may help. Of note, we report a pleural effusion in group 2. This complication raises the issue of other explanations for the pleural effusions than simply the diaphragmatic. One explanation could be volemic surcharge during or after or lymphatic absorption of lymphorrhoea. Two patients in group 1 had pulmonary embolism. Of note, both patients did not have compression stockings for the prevention of postoperative venous thromboembolism. We strongly recommend this prevention for ultra-radical of ovarian cancer. 9 Even if diaphragmatic increases short-term 1066 ª 2009 The Authors Journal compilation ª RCOG 2009 BJOG An International Journal of Obstetrics and Gynaecology

6 Pulmonary morbidity of diaphragmatic Table 5. Review of studies on diaphragmatic for ovarian cancer Number of patients N or N1/N2* Time of Percentage of radical Pleural opening Chest tube Pleural effusions Pneumothorax Pleural punctures or chest tube Deppe Gynecol Oncol Primary 100% < 2 cm 0% 0% 0% Montz Gynecol Oncol Primary 93% 28% 28% 0% 7% 0% Kapnick Gynecol Oncol Primary (16) 100% < 1 cm 52% 52% 0% 33% 0% Recurrent (5) Cliby Gynecol Oncol Recurrent (85%) 80% 24% 24% 9.8% 5% 15% Eisenhauer Gynecol Oncol /215 27% Primary 27% 12% 58% 2% 15% Dowdy Gynecol Oncol /382 14% Primary (37) 43% (52% 30% 0% 12.5% Secondary (19) residual < 1 cm) Devolver Int J Gynecol 69/137 50% Primary (38), 85% 7% 0% 59% 6% 17% Cancer interval (29) Secondary (2) Current study 18/37 (48%) Primary 89% 38% 0% 38% 5% 44% N1, diaphragmatic ; N2, total patients during the study period. morbidity, most of the complications are grade 1 or 2 in MSKCC score and there are no major consequences in postoperative course and long-term morbidity. Chemotherapy was started 36 days after in group 1 and 27 days after in group 2 (ns). Even if diaphragmatic delays the onset of adjuvant chemotherapy, some studies have shown that there is no impact on the efficacy or the rate of survival This study compared morbidity of diaphragmatic with that of a control group. This is of particular importance because diaphragmatic is often part of a radical procedure with multiple bowel resection. We acknowledge that, to some extent, the groups have dissimilar characteristics. First, patients in group 2 were older. Nevertheless, these groups are representative of the surgeon s choices when facing a stage IIIC/IV cancer with the necessity to balance the benefits and risks of extensive. Preoperative imaging included CT-scan, MRI or both and could not be included in the operative decision. Actually, CA125 and imaging were completed by a laparoscopy to decide for a cytoreduction or a neoadjuvant chemotherapy. We currently use Fagotti modified 14 score to select patients. Second, group 2 was heterogeneous: seven patients of group 2 had diaphragmatic carcinomatosis but 12 did not. Third, patients in group 2 were more likely to have received neo adjuvant chemotherapy. However, as reported by Morice et al., 15 the is less morbid after neo adjuvant chemotherapy and would theoretically bias our study in favour of the group 2. Therefore, it is unlikely that these weaknesses have a major impact on our results as we mainly focus on pulmonary morbidity. The aim of the present study was not to determine the survival benefit of diaphragmatic, but to compare morbidity data with a control group. One study have reported follow-up results from a cohort of patients with diaphragmatic, in which overall survival was improved compared to patients without diaphragmatic, even compared to patients optimally debulked for reasons other than diaphragmatic disease. 3 We see this as a further argument for the benefits of optimal cytoreduction, including diaphragmatic resection of peritoneal carcinosis. Disclosure of interest No conflict of interest. Contribution to authorship R.R. and E.C. carried out the study concept and design. E.C., M.B. and F.S. were responsible for acquisition of the data. E.C., R.R., C.P., M.B., F.S. and A.C. carried out the analysis and interpretation of data. E.C. and R.R. drafted the manuscript. R.R. contributed to the statistical expertise. R.R. and E.D. supervised the study. Details of ethics approval Local Research Ethics Committee. Funding None. Acknowledgements None. j ª 2009 The Authors Journal compilation ª RCOG 2009 BJOG An International Journal of Obstetrics and Gynaecology 1067

7 Chéreau et al. References 1 Bristow RE, Tomacruz RS, Armstrong DK, Trimble EL, Montz FJ. Survival effect of maximal cytoreductive for advanced ovarian carcinoma during the platinum era: a meta-analysis. J Clin Oncol 2002;20: Chi DS, Franklin CC, Levine DA, Akselrod F, Sabbatini P, Jarnagin WR, et al. Improved optimal cytoreduction rates for stages IIIC and IV epithelial ovarian, fallopian tube, and primary peritoneal cancer: a change in surgical approach. Gynecol Oncol 2004;94: Aletti GD, Dowdy SC, Podratz KC, Cliby WA. Surgical treatment of diaphragm disease correlates with improved survival in optimally debulked advanced stage ovarian cancer. Gynecol Oncol 2006;100: Cliby W, Dowdy S, Feitoza SS, Gostout BS, Podratz KC. Diaphragm resection for ovarian cancer: technique and short-term complications. Gynecol Oncol 2004;94: Sugarbaker PH, Jablonski KA. Prognostic features of 51 colorectal and 130 appendiceal cancer patients with peritoneal carcinomatosis treated by cytoreductive and intraperitoneal chemotherapy. Ann Surg 1995;221: Sugarbaker PH. Peritonectomy procedures. Ann Surg 1995;221: Hacker NF, Berek JS, Lagasse LD, Nieberg RK, Elashoff RM. Primary cytoreductive for epithelial ovarian cancer. Obstet Gynecol 1983;61: Eisenhauer EL, D Angelica MI, Abu-Rustum NR, Sonoda Y, Jarnagin WR, Barakat RR, et al. Incidence and management of pleural effusions after diaphragm peritonectomy or resection for advanced mullerian cancer. Gynecol Oncol 2006;103: Committee on Practice Bulletins Gynecology, American College of Obstetricians and Gynecologists. ACOG Practice Bulletin No. 84: Prevention of deep vein thrombosis and pulmonary embolism. Obstet Gynecol 2007;110: Aletti GD, Long HJ, Podratz KC, Cliby WA. Is time to chemotherapy a determinant of prognosis in advanced-stage ovarian cancer? Gynecol Oncol 2007;104: Gadducci A, Sartori E, Landoni F, Zola P, Maggino T, Maggioni A, et al. Relationship between time interval from primary to the start of taxane-plus platinum-based chemotherapy and clinical outcome of patients with advanced epithelial ovarian cancer: results of a multicenter retrospective Italian study. J Clin Oncol 2005;23: Rosa DD, Clamp A, Mullamitha S, Ton NC, Lau S, Byrd L, et al. The interval from to chemotherapy in the treatment of advanced epithelial ovarian carcinoma. Eur J Surg Oncol 2006;32: Paulsen T, Kaern J, Kjaerheim K, Haldorsen T, Tropé C. Influence of interval between primary and chemotherapy on short-term survival of patients with advanced ovarian, tubal or peritoneal cancer. Gynecol Oncol 2006;102: Brun JL, Rouzier R, Uzan S, Daraï E. External validation of a laparoscopic-based score to evaluate resectability of advanced ovarian cancers: clues for a simplified score. Gynecol Oncol 2008;110: Morice P, Dubernard G, Rey A, Atallah D, Pautier P, Pomel C, et al. Results of interval debulking compared with primary debulking in advanced stage ovarian cancer. J Am Coll Surg 2003;197: Deppe G, Malviya VK, Boike G, Hampton A. Surgical approach to diaphragmatic metastases from ovarian cancer. Gynecol Oncol 1986;24: Montz FJ, Schlaerth JB, Berek JS. Resection of diaphragmatic peritoneum and muscle: role in cytoreductive for ovarian cancer. Gynecol Oncol 1989;35: Kapnick SJ, Griffiths CT, Finkler NJ. Occult pleural involvement in stage III ovarian carcinoma: role of diaphragm resection. Gynecol Oncol 1990;39: Dowdy SC, Loewen RT, Aletti G, Feitoza SS, Cliby W. Assessment of outcomes and morbidity following diaphragmatic peritonectomy for women with ovarian carcinoma. Gynecol Oncol 2008;109: Devolder K, Amant F, Neven P, van Gorp T, Leunen K, Vergote I. Role of diaphragmatic in 69 patients with ovarian carcinoma. Int J Gynecol Cancer 2008;18: ª 2009 The Authors Journal compilation ª RCOG 2009 BJOG An International Journal of Obstetrics and Gynaecology