Impact of Operative Blood Loss on Survival in Invasive Ductal Adenocarcinoma of the Pancreas Shunji Nagai, MD, PhD,* Tsutomu Fujii, MD, PhD,* Yasuhiro Kodera, MD, PhD,* Mitsuro Kanda, MD, PhD,* Tevfik T. Sahin, MD,* Akiyuki Kanzaki, MD,* Suguru Yamada, MD, PhD,* Hiroyuki Sugimoto, MD, PhD,* Shuji Nomoto, MD, PhD,*Shin Takeda, MD, PhD,* Satoshi Morita, PhD,Þ and Akimasa Nakao, MD, PhD* Objectives: The aim of this study was to determine the prognostic factors and assess the impact of excessive operative blood loss (OBL) on survival after pancreatectomy for invasive ductal adenocarcinoma. Methods: From the retrospective analysis, 271 patients were eligible for evaluation. Overall survival was assessed to clarify the prognostic determinants, including patient characteristics, perioperative factors, and tumor characteristics. Results: The overall survival was significantly affected by the amount of OBL. The median survival times were 26.0, 15.3, and 8.7 months for OBL less than 1000, 1000 to 2000, and greater than 2000 mL, respec- tively (G1000 vs 1000Y2000 mL, P = 0.019; 1000Y2000 vs 92000 mL, P G 0.0001). Operative blood loss greater than 2000 mL remained an independent prognostic factor in multivariate analysis (P = 0.003; hazards ratio, 2.55). Operative blood loss of 2010 mL was found to be an appropriate cutoff level to predict early mortality within 6 months after resection (sensitivity, 0.660; specificity, 0.739). Male sex, year of resection, and plexus invasion were independently associated with OBL greater than 2000 mL. Conclusions: Excessive OBL was found to be a prognostic determi- nant of survival after surgery for pancreatic cancer. Operative blood loss can be used to stratify the risk for pancreatic cancer mortality. Success- ful curative resection with limited blood loss can contribute to improved survival. Key Words: pancreatic cancer, operative blood loss, postoperative complication, blood transfusion, prognostic factor Abbreviations: OBL - operative blood loss, DGE - delayed gastric emptying, ROC - receiver operating characteristic, MST - median survival time, HR - hazards ratio (Pancreas 2011;40: 3Y9) P ancreatic cancer is one of the most difficult malignancies to cure. Curative resection is considered to be the most im- portant factor for determining the outcome in patients with pancreatic adenocarcinoma. 1,2 Notably, surgical resection is su- perior to chemoradiation for locally invasive pancreatic cancer without distant metastases or major arterial invasion and im- proves survival. 3 The following factors have been reported to be associated with overall survival in pancreatic cancer: patient demographics such as age and sex; histopathologic factors such as tumor size, differentiation, blood vessel or neural invasion, lymph node status, and resection margins; and perioperative factors such as type of resection, operative blood loss (OBL), red blood cell transfusion, operation time, era of resection, and chemother- apy. 4Y15 Operative blood loss and red blood cell transfusion are also significant prognostic determinants for other cancer types, such as hepatocellular carcinoma and gastric cancer. 16,17 Pan- createctomy is one of the most difficult and challenging opera- tions and sometimes leads to massive blood loss and prolonged operation time. Thus, it seems to be very important to better understand the effect of surgical stress on survival in pancreatic cancer. Although OBL has been proposed as a prognostic factor in several studies of pancreatic cancer, 5,7,11 no study has demon- strated a stratified risk for cancer mortality for OBL. Further- more, to the best of our knowledge, no studies have determined whether the amount of OBL may be associated with early cancer mortality. This was a retrospective study to identify prognostic factors after curative resection of pancreatic cancer. The experience of the large, single-center is favorable for this type of analysis, be- cause experienced surgeons regularly perform pancreatectomies and the postoperative course is under a well-organized perio- perative management protocol, which means that there should be little bias regarding treatments. The aim of this study was to detect prognostic factors through comprehensive evaluation, focusing on perioperative factors, particularly OBL. In addition, we wished to statistically clarify the negative impact of OBL on early cancer mortality and determine whether there is a threshold value for increasing the risk of early cancer mortality. MATERIALS AND METHODS Patient Selection Between July 1981 and June 2009, there were 614 operative cases of invasive ductal adenocarcinoma of the pancreas at the Department of Surgery II, Nagoya University, and 416 patients underwent curative resection. The medical records of these patients were reviewed retrospectively. After resection, the patients were categorized based on the International Union Against Can- cer classification system, sixth edition. 18 Patients at stages III and IV were considered inappropriate for this analysis because the influence of invasion to the celiac or superior mesenteric artery would likely mask other factors. Therefore, 104 patients at stage III or IV were excluded from this study, and the remaining 312 patients (204 males and 108 females) were evaluated. After analyzing the operational and anesthetic records, 41 patients were excluded because of insufficient data. Thus, 271 patients were finally included in this study. ORIGINAL ARTICLE Pancreas & Volume 40, Number 1, January 2011 www.pancreasjournal.com 3 From the *Department of Surgery II, Nagoya University Graduate School of Medicine, Nagoya, Aichi; and †Department of Biostatistics and Epide- miology, Yokohama City University Medical Center, Yokohama, Kanagawa, Japan. Received for publication May 7, 2010; accepted June 30, 2010. Reprints: Tsutomu Fujii, MD, PhD, Department of Surgery II, Nagoya University Graduate School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya, Aichi 466-8550, Japan (e-mail: fjt@med.nagoya-u.ac.jp). No financial support was received for this study. Preliminary data from this study were presented in part at the 40th anniversary meeting of the American Pancreatic Association and the Japan Pancreas Society, November 2009, Honolulu, HI. Copyright * 2011 by Lippincott Williams & Wilkins Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.