Sir.—The article "Corticosteroid Treatment of Laryngotracheitis v Spasmodic Croup in Children" by Koren et al1 addressed an issue of potential interest and importance. However, several limitations of the study make interpretation and generalization of the results difficult.
First, the problem statement, while it outlined problems of past studies, was not sufficiently specific. It was unclear if severe croup or all croup was the subject of interest. Judging by the mean respiratory rates presented, many patients with very mild disease (with normal or borderline-normal respiratory rates) were admitted to the study. Mild croup is usually a self-limited condition. We question the need for studying a drug that may provide only limited relief for a mild illness.
Second, the diagnostic criteria, as presented in Koren and colleagues' Table 2 for laryngotracheitis and spasmodic croup, were not mutually exclusive.1 It was unclear exactly at what time during the admission/hospitalization the