Ovarian cancer is the third most common cancer of the female reproductive tract, yet has the highest case fatality ratio of all gynecologic malignancies. Every year, approximately 23,000 American women are diagnosed with the disease and about 14,000 die, despite aggressive management.[1,2] One of the main reasons for the high fatality ratio among ovarian cancer patients relates to the fact that the majority are diagnosed with advanced disease. Table 1 summarizes the International Federation of Gynecology and Obstetrics (FIGO) staging system for ovarian cancer.[3] The distribution by FIGO stage of over 4,000 ovarian cancer patients is presented in Table 2.[3] Surgery followed by chemotherapy is the standard approach to the management of ovarian cancer. Despite many chemotherapeutic advances over the past 2 decades, surgery remains the cornerstone of effective disease management. For patients who present with early-stage disease, surgery involves a comprehensive staging procedure with the goal of thoroughly evaluating the peritoneal cavity and the retroperitoneum to identify clinically inapparent sites of metastasis. The rationale and procedures performed during staging are described elsewhere and are not the focus of this review.[4] For patients with advanced-stage ovarian cancer, the goal of surgery is cytoreduction or tumor debulking. "Primary" cytoreduction is performed at the time of diagnosis, prior to the initiation of cytotoxic chemotherapy. In cases of suboptimal primary cytoreduction, "interval" cytoreduction has been described as a second attempt at surgical debulking after the initiation of primary chemotherapy, but prior to its completion. "Secondary" cytoreduction generally refers to surgical debulking performed at the time of disease recurrence. The role of surgical cytoreduction in the management of advanced and recurrent ovarian cancer is the subject of this review. Primary Cytoreduction As stated previously, the majority of ovarian cancer patients present with advanced disease, for which the goal of surgery is to document the stage and extent of the disease, and more importantly, to resect as much grossly visible tumor as possible with the knowledge that microscopic and perhaps macroscopic tumor will be left behind. For most solid tumors, such as lung, pancreatic, and gastric cancer, surgical "cytoreduction" or "debulking" has no role in management because no effective chemotherapy protocol exists for these malignancies. However, there are effective chemotherapy protocols for ovarian cancer. Consequently, both theoretical and clinical benefits have been demonstrated after surgical cytoreduction for the majority of newly diagnosed cases of ovarian cancer, as well as in select cases of recurrent disease. Theoretical Benefits
The most important theoretical effect of primary cytoreductive surgery is its impact on the chemosensitivity of the residual tumor nodules. Chemotherapeutic agents exert maximum cytotoxicity on tumors that are adequately perfused, well differentiated, and have a high mitotic index. Smallervolume residual tumors after cytoreduction have an adequate blood supply creating an oxygen- and nutrientrich environment that promotes cell division. Bulky noncytoreduced tumors, however, outgrow their blood supply, creating a nutrient-deficient and hypoxic environment that halts cell division. Cytoreduction, therefore, maximizes the chemotherapeutic response because the cell-cycle-specific and -nonspecific chemotherapeutic agents act when the cells have a high mitotic rate and are, therefore, most susceptible to cytotoxic drugs. Clinical Benefits
Griffiths[5] was the first investigator to quantify the diameter of the largest residual tumor in relation to chemotherapy response rates and survival. He demonstrated an inverse relationship between maximal residual tumor and patient survival. Since his landmark study, many authors have confirmed their observations.[6-9] Today, the majority of studies use the diameter of the largest remaining tumor nodule as the measurement of residual disease. Patients are divided into optimal and suboptimal groups, using a variety of cutoff points-from any visible disease to 3 cm-to define optimal cytoreduction. Overall, a clear clinical benefit is observed when patients are optimally cytoreduced. These patients have higher response rates to chemotherapy and also have higher median survivals.[10] In two large Gynecologic Oncology Group (GOG) studies, Hoskins et al further clarified the role of primary cytoreductive surgery in patients with advanced ovarian cancer.[6] In the first study, the survival of patients who were found to have abdominal disease ≥ 1 cm during surgery was compared to that of patients with abdominal disease of ≥ 1 cm that had then been surgically cytoreduced to ≤ 1 cm. If surgery was the only important prognostic factor, then survival should have been the same in the two groups. However, patients with small-volume disease survived longer than patients who were cytoreduced to small-volume disease. Further analysis showed that the age of the patient, the grade of the tumor, and the number of residual tumor nodules were independent prognostic factors. Although this study did not show primary cytoreduction to be ineffective, it did show that other factors, including tumor biology, were also important. In the second study, cytoreduction to ≤ 2-cm residual disease resulted in a significant survival benefit, but all residual diameters > 2 cm had equivalent survival. Therefore, unless cytoreduction resulted in the maximum residual tumor diameter being ≤ 2 cm, surgical cytoreduction did not improve survival.[11] Three distinct groups of patients emerged from these two studies: (1) those with no grossly visible residual disease; (2) those with optimal residual disease (≤ 2 cm); and (3) those with suboptimal disease (> 2 cm in diameter). The 4-year survival rates for these patients were 60%, 35%, and 20%, respectively. Table 3 summarizes the largest studies in the literature from both cooperative groups[12,13] and single institutions[ 14-16] that have analyzed the survival of patients with stage III ovarian cancer by residual disease after primary cytoreductive surgery and postoperative platinum-based chemotherapy.

• Optimal Cytoreduction Parameters- Although the benefits of optimal cytoreduction for advanced ovarian cancer appear to be well established, as Table 3 demonstrates, the most appropriate cutoff point for optimal cytoreduction remains controversial. In his seminal paper, Griffith[ 5] used 1.5 cm as the cutoff point in defining optimal cytoreduction. Other authors have used values between any visible disease and 3 cm.
In 2001, we reported a 28% survival for 73 patients with advanced ovarian cancer cytoreduced to between 1- and 2-cm residual disease, which was not significantly different from the 21% 5-year survival of patients with > 2-cm residual disease. These survival rates were both significantly lower than the 50% 5-year survival for patients cytoreduced to ≤ 1 cm. These data support the current GOG definition of optimal cytoreduction as that with residual disease ≤ 1 cm.[16] At our institution, the rates of optimal cytoreduction for advanced ovarian cancer historically have been less than 50%.[16-18] In January 2001, with strong data supporting the survival benefits of optimal cytoreduction,[ 12-16] our service began incorporating the use of extensive upper abdominal cytoreductive procedures including diaphragm-stripping/resection, splenectomy, distal pancreatectomy, liver resection, resection of tumor from the porta hepatis, and cholecystectomy, if needed, to achieve optimal (≤ 1-cm residual) cytoreduction. We studied two separate groups of patients. Group 1 consisted of patients who had their primary surgery at our institution between November 1998 and May 2000-a time when extensive upper-abdominal procedures were not utilized during primary cytoreductive procedures. Group 2 patients underwent their primary surgery between January 2001 and May 2002-a time when a more aggressive approach to surgical cytoreduction was used. In the first group of patients, optimal cytoreduction was attained in 50% of patients, compared to 76% in the second group (P < .01). Although operative time and estimated blood loss increased, complications did not increase significantly.[19]
• Technologic Advances Allowing Maximal Cytoreduction-Other advances in surgical technology have allowed surgeons to maximize cytoreductive efforts, enabling them to remove virtually all visible disease. The main, new technologic advances consist of the cavitational ultrasonic surgical aspirator and the argon-beam coagulator. Bristow and Montz[20] have shown the feasibility and success of surgical cytoreduction using the argon-beam coagulator. They reported that the use of this device resulted in a 94% successful cytoreduction rate, compared to 64% without it. In a similar study, patients cytoreduced with the cavitational ultrasonic surgical aspirator had lower morbidity, decreased length of hospital stay, and a lower serum level of CA-125 at 1 and 2 months postsurgery, compared to patients not cytoreduced with this technology.[21]

Meta-analyses
Despite numerous studies demonstrating improved survival of advanced ovarian cancer patients who were optimally cytoreduced, critics argue that it is not the surgical cytoreduction that leads to prolonged survival, but rather, the inherent biology of the tumor (which also enables optimal cytoreduction). The meta-analysis by Hunter and colleagues[22] is often cited to support this argument. They analyzed 58 articles encompassing 6,962 advanced ovarian cancer patients and found that a 10% increase in maximal cytoreductive surgery resulted in a 4.1% increase in median survival. However, the use of platinum- based chemotherapy produced an estimated 53% increase in median survival. Hunter et al concluded that surgery had a minor effect on the survival of women with advanced ovarian cancer. This meta-analysis has been criticized because of the low rate of optimal debulking and the lack of treatment with optimal platinumcontaining chemotherapy.[23] More recently, Bristow et al[24] performed a meta-analysis on the survival effect of maximal cytoreduction for ovarian cancer during the era of platinum-based chemotherapy. This analysis evaluated 81 studies involving 6,885 patients with stage III/IV ovarian cancer and found that during the platinum era, maximum cytoreduction was a powerful determinant of cohort survival. These investigators showed further that each 10% increase in maximal cytoreduction resulted in a 5.5% increase in median survival. They concluded that expert centers with optimal cytoreduction rates of 75% or greater offered a 50% increase in median survival over less experienced centers with optimal cytoreduction rates of 25% or less. Stage IV Disease
The role of primary cytoreduction in the management of stage III disease is well defined; however, its benefit in stage IV disease is not as clear. Most studies that address survival in ovarian cancer patients evaluate only stage III disease or a combination of stage III and IV disease. In stage IV disease- which by definition includes extraperitoneal or intrahepatic metastasis- the benefit of surgical cytoreduction has been questioned. An early study by Wharton et al[25] reported a 4-year survival of 9% in patients who were cytoreduced to ≤ 2 cm. This study was performed before the platinum-based chemotherapy era, which may explain the poor outcomes.[25] Additionally, Goodman et al reported no significant survival advantage for patients who underwent optimal cytoreductive surgery for stage IV epithelial ovarian cancer compared to patients who were suboptimally cytoreduced.[26] Contrary to these reports, most other studies have shown a survival advantage for primary cytoreduction in patients with stage IV ovarian cancer.[ 17,27] Curtin and colleagues[17] reported on 97 patients with stage IV disease, of which 51% were optimally cytoreduced (residual disease ≤ 2 cm). Median survival for the optimally vs suboptimally cytoreduced group was 40 vs 18 months (P = .0136). Bristow and investigators[27] have also shown the value of optimal debulking in stage IV patients with liver metastases. In this study, patients with optimal hepatic and extrahepatic residual disease had a median survival of 50.1 months, compared to 27 months for those with suboptimal hepatic but optimal extrahepatic residual disease, and 7.6 months for patients with both suboptimal hepatic and extrahepatic residual disease. Table 4 summarizes recent studies of surgical cytoreduction for stage IV ovarian cancer.[17,26-30] Neoadjuvant Chemotherapy Neoadjuvant chemotherapy, which is given to patients with advanced ovarian cancer prior to cytoreductive surgery, has several theoretical advantages over primary cytoreduction. These include (1) improvement of performance status, especially in elderly patients and those with pleural effusions and ascites, (2) decrease in the extent and morbidity of surgery by reducing tumor volume preoperatively, and (3) increase in the percentage of patients undergoing optimal cytoreduction. Several retrospective studies have evaluated this approach, but to date, no prospective randomized data exist. Vergote et al[31] reported on 285 patients with advanced ovarian cancer treated between 1980 and 1997. In the period from 1980 to 1988, all patients underwent primary debulking surgery. From 1989 to 1997, patients were surgically evaluated to determine whether they should receive primary chemotherapy (43%) or primary cytoreductive surgery (57%). The crude 3-year survival in the later part of the study was 42%, compared to 26% during the earlier period, when all patients underwent primary debulking surgery. Ansquer et al[32] reported the French multicenter experience with neoadjuvant chemotherapy for ovarian cancer deemed unresectable by either laparoscopy (61%) or laparotomy (39%). Patients received a median of four cycles of chemotherapy preoperatively, with 80% responding to chemotherapy and subsequently undergoing debulking, which was optimal in 91%. The authors concluded that neoadjuvant chemotherapy in unresectable ovarian cancer led to the selection of chemosensitive patients who, in the majority of cases, could subsequently undergo optimal cytoreduction. In addition, aggressive cytoreduction was avoided in patients with initial chemoresistance. Table 5 summarizes the studies that have compared the neoadjuvant approach to standard primary cytoreduction.[ 33-36] Many of the patients managed with neoadjuvant chemotherapy had significant medical comorbidities, and except for the series reported by Kuhn and colleagues,[ 35] none of the median survivals using the neoadjuvant chemotherapy approach those attained with optimal primary cytoreduction. Therefore, many authorities are hesitant to uniformly use this strategy, because it may result in a significant number of patients being deprived of the opportunity for optimal cytoreduction and 5-year survival rates of up to 50%. However, if preoperative evaluation could predict which patients had disease so extensive that optimal cytoreduction could not be performed, these patients would be ideal candidates for neoadjuvant chemotherapy. Retrospective studies have evaluated the accuracy of computed tomography (CT) scanning and serum CA-125 level in selecting candidates for optimal cytoreduction. These studies report sensitivities, specificities, positive predictive values, and negative predictive values ranging from 50% to 100%, 63% to 100%, 61% to 100%, and 75% to 100%, respectively.[ 37-40] We are currently evaluating the ability of preoperative CT scan and serum CA-125 to predict optimal primary cytoreduction for patients with advanced ovarian cancer in a multicenter prospective trial. The role and indications for neoadjuvant chemotherapy may be further elucidated after the completion of a randomized prospective trial being performed jointly by the European Organization for the Research and Treatment of Cancer (EORTC) and the National Cancer Institute of Canada (NCI-C). In this trial, patients with biopsy-proven stage IIIC/IV ovarian cancer will be randomized to primary cytoreduction vs neoadjuvant chemotherapy. Both arms will receive six cycles of paclitaxel(Drug information on paclitaxel) and platinum chemotherapy. The trial is expected to close to accrual in 2005.[23]