Surgery News - May 2008 - (Page 15)

M AY 2 0 0 8 • SURGERY NEWS ONCOLOGY Prognostic Score Stratifies Pancreatic Cancer Risk BY MITCHEL L. ZOLER Else vier Global Medical Ne ws atient age, tumor grade, and presence of distant metastases were the most powerful prognostic factors for predicting survival following resection of pancreatic, neuroendocrine tumors in a review of almost 4,000 patients. This analysis led to development of a prognostic score formula that was effective at stratifying patients into high-, medium-, and low-risk subgroups, investigators reported in the Annals of Surgery. The incidence of pancreatic neuroendocrine tumors is increasing, and until now physicians have not had a widely accepted system for predicting patient outcomes following tumor resection. “The prognostic score will provide information regarding expected survival, assist in adjuvant treatment decisions, and allow patient stratification for clinical trials,” wrote Dr. Karl Y. Bilimoria, a surgeon at Northwestern University, Chicago, and his associates (Ann. Surg. 2008;247:490-500). Using data collected in the National Cancer Data Base, the investigators based their analysis on 3,851 adult U.S. patients who underwent resection of a pancreatic neuroendocrine tumor during 1985-2004. The database is a registry maintained by the American College of Surgeons’ Commission on Cancer and the American Cancer Society. The patients’ median age was 56 years, and P those younger than 18 were excluded. Neuroendocrine carcinomas made up 84% of the total, and 78% had wellor moderately well-differentiated tumors. About 62% of patients had tumors 4.0 cm or smaller in diameter, 36% had nodal metastases, and 20% had distant metastases. Multivariable Cox proportional hazard models were used to identify factors significantly associated with survival following resection of these cancers. Factors significantly linked with worse survival were older age, nonfunctional status, presence of metastases to the liver and other distant sites, high tumor grade, and resection by pancreaticoduodenectomy. Tumor history indicating gastrinoma was linked with significantly improved survival. The most powerful predictive factors from this group were age, tumor grade, and presence of distant metastases. Dr. Bilimoria and his associates used these findings to devise a postresection prognostic score, assigning various point values to different clinical presentations within these three parameters (see chart). After calculating scores for all of the patients in the database, they found that the patients fell into three groups with significantly different 5-year survival rates. Those with a score of zero had a 5-year survival rate of 77%. A score of one or two corresponded to a 5-year survival rate of 51%, while a score of three or more showed a 36% survival rate. Adding other variables into the score did not appreciably improve its prognostic ability. Survival Prognostic Score After a Pancreatic Neuroendocrine Tumor Resection Clinical features Age 75 (yr) Tumor grade Well or moderately differentiated Poorly differentiated Presence of distant metastases None Liver Other Total score 0 1-2 3 Points 0 1 2 0 1 0 1 3 ELSEVIER GLOBAL MEDICAL NEWS 5-year survival rate 77% 51% 36% Note: Scoring system based on an analysis of 3,851 adults. Source: Annals of Surgery Pancreatic Surgery Outcomes Vary Widely Among High-Volume Units BY ROBERT FINN Preop Chemo Stretches Esophageal Ca Survival B Y F R A N L O W RY Else vier Global Medical Ne ws H U N T I N G T O N B E A C H , C A L I F. — Volume alone is not a reliable single predictor of quality or outcomes following pancreatic surgery, according to a study presented by Dr. Taylor S. Riall at the Academic Surgical Congress. In a study of 12 high-volume pancreatic surgery units in Texas, Dr. Riall and her colleagues from the University at Galveston found high degrees of variability in several outcome measures including the proportion of patients discharged to a skilled nursing facility, their mean length of stay, and the amount of time between admission to the hospital and operation. On the other hand, after adjustment for a number of confounding variables, the investigators found no significant differences in in-hospital mortality among the high-volume hospitals. “As pay-for-performance becomes increasingly important, hospitals and surgeons will be under increased pressure to provide evidence of the volume of care they deliver,” said Dr. Riall, an ACS Fellow. “[But] some high-volume centers do not have ideal outcomes, and it is likely that some low-volume centers do.” The investigators looked at all hospitals in Texas where pancreatic resections of any type were performed between 1999 and 2005. They used criteria developed by the Leapfrog Group for Patient Safety, a coalition of more than 150 public and private health care purchasers, to define high-volume centers. High-volume centers were those that performed an average of more than 10 pancreatic resections per year, and at least 10 resections in 4 of the 7 years included in the study. Among the 12 high-volume centers in Texas, the number of pancreatic resections performed during the 7-year period ranged from 78 to 608, a significant difference. In-hospital mortality ranged from 0.7% to 7.7%. While this unadjusted difference was statistically significant, statistical significance disappeared after the investigators controlled for age, gen- der, race, risk of mortality, admission status, diagnosis, procedure type, and insurance status. Dr. Riall explained this by noting that some of the hospitals with higher mortality rates tended to operate on patients at higher risk. Those patients who did not die in the hospital usually were discharged home, but some required discharge to a skilled nursing facility. The unadjusted percentage of patients discharged to a skilled nursing facility varied significantly among the high-volume hospitals, from less than 1% to 41%. The differences remained significant even after adjustment for confounders in a multivariate analysis. Compared with the highest-volume hospital, the other hospitals discharged patients to skilled nursing facilities twofold to ninefold more frequently. Another measure of quality is how often patients are sent to the operating room within 24 hours of admission. Compared with the highest-volume hospital, six of the hospitals operated within 24 hours significantly more frequently, and one got patients to the OR significantly less frequently within that time frame. Length of stay and total hospital charges also varied significantly from hospital to hospital. Mean hospital charges, for example, ranged from just over $40,000 to just over $100,000 per patient. The differences remained significant after multivariate analysis. Dr. James Neifeld, an ACS Fellow who was asked to comment on the study, called the results “provocative,” but said that the differences between hospitals in how often patients are sent to the OR within 24 hours of admission may just represent the patients they see. “For example, if one institution has most of their patients coming from the medical service . . . it will be a different group of patients than those referred to an institution after the complete evaluation is done at the other site or as an outpatient,” said Dr. Neifeld, chair of the surgery department at Virginia Commonwealth University, Richmond. In addition, he noted, it’s unclear whether the hospitals were similar in terms of other prognostic factors, tumor size, or portal vein resection. Dr. Riall declared she had nothing to disclose. Else vier Global Medical Ne ws O R L A N D O — Updated data from the Medical Research Council Oesophageal Cancer Working Group’s OEO2 trial confirm that giving chemotherapy with cisplatin and 5-fluorouracil prior to surgery prolongs disease-free and overall survival in patients with resectable esophageal cancer. At a median follow-up of 6 years, the 5-year overall survival rate was 23% for patients randomized to neoadjuvant chemotherapy, compared with 17% for those who had surgery alone, Dr. William H. Allum reported at a meeting on gastrointestinal cancers sponsored by the American Society of Clinical Oncology. The initial analysis reported 2year survival based on a median follow-up of three years (Lancet 2002;359:1727-33). Although this result favored neoadjuvant chemotherapy, there were concerns that this was too short a period for conclusive results, said Dr. Allum, the Royal Marsden Hospital, London. The OEO2 trial randomized 802 patients with operable squamous cell carcinoma or adenocarcinoma of the esophagus either to surgery with two preoperative courses of chemotherapy, or to surgery alone. The chemotherapy sessions with cisplatin and 5-fluorouracil were spaced 3 weeks apart, after which patients were sent to surgery within 24 weeks. The patients’ median age was 63 years; 75% were men, about a third had squamous histology, and two-thirds had adenocarcinoma. The 2-year overall survival for patients given preoperative chemotherapy was 43% versus 34% for those in the surgery-only arm. Serious adverse events were not greater with chemotherapy. At 6 years of follow-up, the results favoring chemotherapy were maintained irrespective of age, gender, and tumor site. Dr. Allum reported hazard ratios of 0.82 (P = .003) for disease-free survival and 0.84 (P = .03) for overall survival. There have been 320 deaths in the chemotherapy arm and 335 in the surgery-only arm; 78% of deaths were cancer related. “We conclude that preoperative chemotherapy improves survival in operable esophageal cancer and should be considered as a standard of care,” he said, adding that it should not be the only standard of care. The OEO5 trial, begun in September 2004, also studies patients with resectable esophageal cancer; it compares the chemotherapy of OEO2 with a combination of epirubicin, cisplatin, and capecitabine hypothesized to be more effective in the neoadjuvant setting. Resul

Table of Contents Feed for the Digital Edition of Surgery News - May 2008

Surgery News - May 2008 Contents New Lung Approach Speeds Extubation Innovative GI Procedures May Improve Diabetes Quality Programs Differ on Risk Data Crystal Ball Medical Modeling Ventricular Valve Taking Stock

Surgery News - May 2008

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.