Fever Control Using External Cooling in Septic Shock

To the Editor:

We read with much interest the article by Schortgen and colleagues, in which they showed that external cooling of patients with septic shock and fever was associated with a reduction in vasopressor requirements and early (14-d) hospital mortality (1). We believe two points not addressed by the authors may have contributed to the findings of the study. The first is that the volume of intravenous fluids administered for hemodynamic stabilization in the patients in both the “cooling” and “no-cooling” groups (average of 1.5 and 1.0 L, respectively) was quite small. This suggests that patients in the study may have been less than adequately volume resuscitated, resulting in higher vasopressor requirements in both the cooling and no-cooling groups. Second, although both groups received similar amounts of intravenous fluids, the study design did not take into account the fact that fever causes a substantial increase in insensible water loss. Thus, persistent fever in the no-cooling group would have led to greater insensible losses, and thereby lower intravascular volume, than in the cooling group. It is therefore unclear whether the better outcomes in the cooling group were due to lower temperature per se, or, instead, to better preservation of intravascular volume. Insensible loss of water caused by fever has been known to influence intravascular volume for many years (2). Lowering the temperature of febrile patients as a therapeutic strategy in septic shock is an interesting idea, which we believe should be further tested in studies that take into account the potentially confounding effect of fever on insensible water loss.