| Literature DB >> 3977449 |
J G Chandler, C H Van Meter, D L Kaiser, S E Mills.
Abstract
The course of 121 shunted cirrhotic patients, managed according to a prospective protocol over a period of 10 years, was analyzed to determine predictors of 30-day and long-term survival. Forty-five per cent of the patients underwent emergent decompression within 12 hours of active bleeding, and 34% of the shunts were selective. Logistic regression linked early mortality to bilirubin and blood-urea nitrogen (BUN) (p = 0.001), and long-term survival to the presence of preoperative ascites and higher levels of alkaline phosphatase (p = 0.027), but neither variable set was a more accurate predictor than Child Class. Emergency shunt patients had greater risk of early death, 44% vs. 17% for patients shunted electively (p = 0.001), but beyond 30 days, their Kaplan-Meier survival curves were identical. Independently, angiographic prograde portal flow was favorably associated with short-term (p = 0.003) but not prolonged survival. The presence of Mallory bodies, fatty metamorphosis, and acute periportal inflammation, alone or in combination, had no prognostic value. Continued post-operative alcohol ingestion jeopardized long-term survival (p = 0.017). Survival of nonalcoholics was enhanced by selective as opposed to total splanchnic decompression (p = 0.009).Entities:
Mesh:
Year: 1985 PMID: 3977449 PMCID: PMC1250737 DOI: 10.1097/00000658-198504000-00013
Source DB: PubMed Journal: Ann Surg ISSN: 0003-4932 Impact factor: 12.969