Portal-systemic encephalopathy in a randomized controlled trial of endoscopic sclerotherapy versus emergency portacaval shunt treatment of acutely bleeding esophageal varices in cirrhosis

In patients with cirrhosis and bleeding esophageal varices, there is a widespread belief that control of bleeding by portal-systemic shunts is compromised by a high incidence of shunt-related portal-systemic encephalopathy (PSE). This important issue was examined by a randomized controlled trial that compared emergency and long-term endoscopic sclerotherapy (EST) to emergency direct portacaval shunt (EPCS) in patients with cirrhosis and acute variceal hemorrhage. The study was a community-wide undertaking known as the San Diego Bleeding Esophageal Varices Study. A total of 211 unselected, consecutive patients with biopsy-proven cirrhosis and endoscopically proven, acutely bleeding esophageal varices that required at least 2 units of blood transfusion were randomized to EST (n = 106) or EPCS (n = 105). The diagnostic workup was completed in less than 6 hours and EST or EPCS was initiated within 8 hours of initial contact. Long-term EST was performed according to a deliberate schedule over months. Criteria for failure of EST or EPCS were clearly defined and crossover rescue treatment was applied, whenever possible, when failure of primary therapy was declared. PSE was quantitated by a "blinded" senior faculty gastroenterologist. Four variously weighted components of PSE were graded on a scale of 0 to 4: (1) mental state, (2) asterixis, (3) number connection test, and (4) arterial blood ammonia. PSE was classified as recurrent if 2 or more episodes were documented. All patients (100%) had follow-up for more than 9.4 years or until death. Child's risk classes in the EST and EPCS groups, respectively, were 25% and 30% in class A, 43% and 47% in class B, and 26% and 29% in class C. Mean time from onset of bleeding to EST or EPCS was less than 24 hours, and from study entry to EST or EPCS was 3.1 to 4.4 hours, respectively. EST achieved permanent control of bleeding in only 20% of patients, while EPCS permanently controlled bleeding in every patient (P ≤ 0.001). Survival following EPCS was 3.5 to 5 times greater than that of EST at 5, 10, and 15 years (P ≤ 0.001). The incidence of recurrent PSE following EST (35%) was more than twice the incidence following EPCS (15%) (P ≤ 0.001). EST patients had a total of 179 episodes of PSE and 146 PSE-related hospital admissions, compared with EPCS patients who had 94 episodes of PSE and 87 hospital admissions (P ≤ 0.001). Recurrent upper gastrointestinal bleeding, which was rare in the EPCS group, was a major causative fact