In reply
We are grateful to Dr Johnston for his interest in our article. Although, as he points out, many believe fever adversely affects the outcome of acute ischemic stroke, data supporting this conclusion are meager. Moreover, those few studies that have examined the relationship between fever and prognosis in stroke patients are fraught with flaws in experimental design.
Those cited by Dr Johnston, for example, include problems with temperature measurement techniques (axillary in reference 2 and aural in references 4 and 5 of Dr Johnston's letter), arbitrary fever definitions (temperatures ≥37.9°C in reference 2, >37.5° in references 4 and 5), failure to control for diurnal temperature variability (references 2-8), retrospective study design (reference 4), use of historical controls (reference 7), failure to control for the etiology of fevers (references 2-8), and failure to distinguish between fever and hyperthermia (references 2-8).