Analysis of 6,747 pancreatic neuroendocrine tumors for a proposed staging system.

OBJECTIVES: Currently, no reasonable staging system exists for pancreatic neuroendocrine tumors (PNET) to guide treating physicians. The aim of this study was to devise a staging system of relevant prognostic factors to better predict overall survival in PNET. METHODS: A prospective 300 patient cohort and a review of the Surveillance Epidemiology and End Results database identified 6,447 patients with PNET from 1973 to 2008. Significant prognostic factors were created for an initial. Tumor: T (T1: ≤3 cm and localized to pancreas, T2: >3 cm and localized to the pancreas; T3: extension to adjacent organs and vessels), grade: G (G1: well/moderate and G2: poor/undifferentiated), and metastasis: M (M0: no distant mets, M1: distant mets) staging system. RESULTS: Significant predictors of survival on multivariate analysis included age, size, grade, and metastasis. Based on the TGM staging system: stage 1 (T1-2, G1, M0), stage 2 (T1-2, G2, M0), stage 3 (T3G2M0, Tany, G1, M1), stage 4: (Tany, G2, M1) was created with survival being statistically different between stages (p < 0.0001). Median survival rates were stage 1, 55 months; stage 2, 50 months; stage 3, 46 months; and stage 4, 25 months. CONCLUSIONS: Incorporation of this newly developed staging system into clinical practice will improve the ability to predict prognosis and aid in stratification of patients for clinical trials.

Laparoscopic left pancreatectomy: complication risk score correlates with morbidity and risk for pancreatic fistula.

BACKGROUND: Surgeons are performing laparoscopic left pancreatectomy (LLP) with increasing frequency; however, determinants of perioperative outcome after LLP are not well defined. We evaluated factors contributing to morbidity after LLP. METHODS: Records from patients undergoing LLP from 2000 to 2008 from nine academic medical centers were evaluated to assess risk factors for perioperative complications. Extent of pancreatic resection was determined by the length of the gross pancreatic specimen. Complications and pancreatic fistula rates were assessed, and a model was developed to identify those at risk of postoperative adverse events. RESULTS: Among the 219 LLP cases, indications were cystic neoplasms in 122 (56%), solid neoplasms in 83 (38%), and chronic pancreatitis in 14 (6%). Thirty-day morbidity and mortality were 39% and 0, respectively. Major complications occurred in 11%. Pancreatic fistulae were detected in 23%, with clinically important fistulae (International Study Group on Pancreatic Fistula Definition grade B/C) seen in 10%. On multivariate analysis, only greater estimated blood loss (EBL), higher body mass index (BMI), and longer length of resected pancreas were associated with major complications. A complication risk score consisting of 1 point each for BMI >27, pancreatic specimen length >8 cm, or EBL > or =150 mL predicted an increased risk of complications and pancreatic fistulae. CONCLUSIONS: The risk of major complications after LLP is 11%, with clinically important pancreatic fistulae occurring in 10%. A complication risk score incorporating BMI, extent of pancreatic resection, and EBL correlates with all end points evaluated. The complication risk score should be used when quality outcome measures are evaluated.

Left-sided pancreatectomy: a multicenter comparison of laparoscopic and open approaches.

OBJECTIVES: To compare perioperative outcomes of laparoscopic left-sided pancreatectomy (LLP) with traditional open left-sided pancreatectomy (OLP) in a multicenter experience. SUMMARY AND BACKGROUND DATA: LLP is being performed more commonly with limited data comparing results with outcomes from OLP. METHODS: Data from 8 centers were combined for all cases performed between 2002-2006. OLP and LLP cohorts were matched by age, American Society of Anesthesiologists, resected pancreas length, tumor size, and diagnosis. Multivariate analysis was performed using binary logistic regression. RESULTS: Six hundred sixty-seven LPs were performed, with 159 (24%) attempted laparoscopically. Indications were solid lesion in 307 (46%), cystic in 295 (44%), and pancreatitis in 65 (10%) cases. Positive margins occurred in 51 (8%) cases, 335 (50%) had complications, and significant leaks occurred in 108 (16%). Conversion to OLP occurred in 20 (13%) of the LLPs. In the matched comparison, 200 OLPs were compared with 142 LLPs. There were no differences in positive margin rates (8% vs. 7%, P = 0.8), operative times (216 vs. 230 minutes, P = 0.3), or leak rates (18% vs. 11%, P = 0.1). LLP patients had lower average blood loss (357 vs. 588 mL, P < 0.01), fewer complications (40% vs. 57%, P < 0.01), and shorter hospital stays (5.9 vs. 9.0 days, P < 0.01). By MVA, LLP was an independent factor for shorter hospital stay (P < 0.01, odds ratio 0.33, 95% confidence interval 0.19-0.56). CONCLUSIONS: In selected patients, LLP is associated with less morbidity and shorter LOS than OLP. Pancreatic fistula rates are similar for OLP and LLP. LLP is appropriate for selected patients with left-sided pancreatic pathology.

Promises and pitfalls of anti-angiogenic therapy in clinical trials.

A significant body of research has implicated the process of angiogenesis in the growth and spread of tumors. Elucidation of the mechanisms of tumor angiogenesis has led to the development of multiple anti-angiogenic agents. However, the perceived differences between the results of preclinical studies and those of early phases of clinical trials have led to questions being asked regarding the efficacy of these agents. There are many reasons for this discrepancy, including difficulties in the appropriate interpretation of preclinical data and clinical trial design. Further insights into the complex process of angiogenesis are essential for the development of effective anti-angiogenic regimens.