Does Admission Body Temperature Predict Mortality After Acute Stroke?
To the Editor:
We read with great interest the recent article by Kammersgaard et al1 on the admission body temperature and prognosis after acute stroke. The conclusion is that low body temperature on admission is considered to be an independent predictor of good short-term outcome. The authors also concluded that the study suggests that admission body temperature seems to be a major determinant even for long-term mortality after stroke. The results are beautiful, and these conclusions are easy to be accepted in the light of many experimental studies showing the protective effect of hypothermia on the ischemic brain damage in the focal cerebral ischemia model (see a review2). However, we consider these conclusions possibly misleading.
An interesting article about body temperature in acute stroke was published in this journal3 from the same country just 1 year ago. Two of the authors (J.R., H.N.) of the present article1 belong to the same institute from which the previous article3 came. Although there might be some misunderstanding4 in some parts of the explanations,3 this study is epoch-making. Because Boysen and Christensen’s study has documented that body temperature in acute stroke can change very rapidly even within 6 to 8 hours after onset, a new paradigm is necessitated to study the relationship between body temperature and outcome in acute stroke. Six hours is too long to treat as one group. The current article1 did not cite this important work3 and paid no attention to this point.
In Boysen and Christensen’s study,3 725 patients were admitted within 6 hours of onset and were investigated. Fifty percent of the patients were admitted within 2 hours of onset. Mean temperature on admission of the patients who died within 7 days (36.5°C) or 3 months (36.5°C) tended to be lower than in survivors (36.6°C). In 93 patients, body temperature on admission was <36°C. These patients had more severe strokes than patients with higher temperatures did. Patients with body temperature >37.5°C on admission had better stroke scale than those with admission temperature ≤37.5°C did. These results are quite contrary to those from the article of Kammersgaard et al.1 We think this difference may come from the fact that Kammersgaard et al treated the patients admitted within 6 hours of onset as 1 group.
Boysen and Christensen’s study3 and our own study5 have demonstrated that 4 hours after onset of cerebrovascular diseases (cerebral ischemia, intracerebral hemorrhage, and subarachnoid hemorrhage) may be a turning point for the body temperature in the acute phase of these diseases. Therefore, we would like to know how many patients were admitted within 4 hours of onset in their study.1 We suspect most patients investigated in this study1 admitted 4 to 6 hours after onset.
Kammersgaard LP, Jorgensen HS, Rungby JA, Reith J, Nakayama H, Weber UJ, Houth J, Olsen TS. Admission body temperature predicts long-term mortality after acute stroke: the Copenhagen Stroke Study. Stroke. 2002; 33: 1759–1762.
Ginsberg MD, Busto R. Combating hyperthermia in acute stroke: a significant clinical concern. Stroke. 1998; 29: 529–534.
Boysen G, Christensen H. Stroke severity determines body temperature in acute stroke. Stroke. 2001; 32: 413–417.
Takagi K. Body temperature in the acute phase of stroke. Stroke. 2002; 33: 2154–2155.
We thank Drs Takagi and Fujimaki for their interest in our study of the effect of admission body temperature on long-term mortality after stroke.1 First, Drs Takagi and Fujimaki suggest that analyzing mortality for all patients admitted within 6 hours may be wrong. They furthermore suggest that patients should be stratified into smaller groups because body temperature changes during the early hours after stroke onset. In our study,1 body temperature was measured directly on admission to hospital. Of the total number of 390 patients considered, 278 (71%) were admitted within 4 hours and 118 (30%) within 2 hours after stroke onset.
Drs Takagi and Fujimaki draw attention to the fact that body temperature may change after onset of stroke and vary between patients. This is an interesting issue, but our study was not designed to measure body temperature at different time points after stroke onset and that was not the aim of the study. We think that the only way to firmly conclude on the role for hypothermia in stroke treatment is to conduct randomized controlled clinical trials where body temperature is kept low for some time after stroke onset. Nevertheless, our study suggests that body temperature remains a strong independent predictor of long-term mortality, even after adjustment for other basic characteristics (eg, onset stroke severity, atrial fibrillation, and hypertension) that, unlike body temperature, a priori may be expected to continue to be a risk factor for mortality even after the acute phase of stroke.