The hepatic steatosis observed in the influenza B virus mouse model of Reye syndrome has been attributed to infectious virus or, alternately, to decreased food intake in the virus-treated mice or impurities in the virus preparation. To resolve this issue, 4- to 6-wk-old male Balb C mice were given, by intravenous injection, 12,800 hemagglutination units of influenza B Lee/40 virus in phosphate buffered saline/1% bovine serum albumin using virus prepared by ultra-centrifugation from infected allantoic fluid, by sucrose density-gradient purification of virus prepared by ultracentrifugation from infected allantoic fluid or by irradiation of virus prepared by ultracentrifugation from infected allantoic fluid to inactivate virus. The infectivity titer of virus prepared by ultracentrifugation from infected allantoic fluid was much higher than that of sucrose density-gradient purified virus prepared from infected allantoic fluid: 50% egg infectious dose for virus prepared by ultracentrifugation from infected allantoic fluid was 3.9 x 10(4)/hemagglutination unit vs. 8.7 50% egg infectious dose/hemagglutination unit for sucrose density-gradient purified virus prepared from infected allantoic fluid. Control mice received phosphate-buffered saline/1% bovine serum albumin or uninfected allantoic fluid diluted in phosphate-buffered saline/1% bovine serum albumin. Mice were fasted to eliminate dietary variation, and livers were obtained 36 hr after virus administration. Of the above treatments, only virus prepared by ultracentrifugation from infected allantoic fluid caused clinical illness and increased hepatic triglycerides (p less than 0.02) compared with controls. Hepatic triglycerides in virus prepared by ultracentrifugation from infected allantoic fluid correlated with histopathological vacuolization scores (r = 0.5773; p less than 0.03).(ABSTRACT TRUNCATED AT 250 WORDS)