Abstract

OBJECTIVE: To describe trends in the clinical pattern of Reye's syndrome in the British Isles between 1982 and 1990; and to determine the relation between any changes and the June 1986 warnings against the use of aspirin in children. DESIGN: Development, and application to reported cases, of a scoring system designed such that patients showing the typical clinical and pathological features of 'classical' Reye's syndrome scored highly. The relations between 'Reye scores' and a number of explanatory variables were explored using multivariable analysis. SETTING: British Isles. SUBJECTS: 445 cases fulfilling the Reye's syndrome case definition reported to the surveillance scheme between January 1982 and December 1990. MAIN OUTCOME MEASURE: Individual 'Reye score'. RESULTS: Cases with high scores were more likely to have occurred in the 4 1/2 year period before June 1986 compared with the subsequent period (p < 0.006). Numbers of cases in the low and intermediate score categories declined by about 50% after June 1986, whereas those in the high category fell by 79%. High scorers were more likely to have received aspirin (p < 0.0001) and were older than intermediate and low scorers (p < 0.008). No relation was identified between score and season of onset. CONCLUSIONS: The decline in Reye's syndrome after the aspirin warnings cannot be explained entirely, as has been proposed, by improved diagnosis of 'Reye-like' inherited metabolic and other disorders: this would not account for the greater decline of the high scoring subgroup which also contained those cases most likely to resemble 'classical' Reye's syndrome and to have received aspirin. This study provides further evidence for the role of aspirin in a subset of cases meeting the standard diagnostic criteria for Reye's syndrome and supports the need to consider this disorder as a heterogeneous group of conditions including Reye-like inherited metabolic disorders.