We agree that responses associated with the dose of cocaine administered may be very different from those that occur with cocaine toxicity. However, others have shown that cocaine-induced elevations in core temperature are dose dependent (1, 2). Thus, it is reasonable to suggest that if greater doses of cocaine had been administered, cutaneous vasodilation, sweating, and heat perception may have been even more attenuated. Given that sedating agents decrease internal temperature even in resting healthy persons who do not use cocaine (3), it is not surprising that treatment with sedation agents attenuates the elevation in temperature and improves survival in the cited experimental studies of cocaine toxicity.

We are not proposing that reduced cutaneous vasodilation and sweating is the sole mechanism for elevated internal temperatures due to cocaine use; as indicated in our manuscript, we recognize that increased metabolic activity contributes to elevated internal temperatures. However, our data clearly indicate that in addition to increased metabolism, impaired thermoregulatory responses contribute to the elevation in temperature associated with cocaine use.

Human cutaneous tissue is innervated by a sympathetic adrenergic limb, a sympathetic cholinergic limb leading to sweating, and possibly a sympathetic active vasodilator limb responsible for most of the elevation in skin blood flow during a heat stress (4). Thus, it is far too simplistic to indicate that "enhanced sympathetic outflow should cause diaphoresis." For example, skin-surface cooling causes pronounced increases in skin sympathetic activity but clearly does not cause sweating. If the observed responses were due solely to antimuscarinic effects of cocaine, then only sweating responses would be impaired because local administration of atropine abolishes sweating but does not cause similar alterations in cutaneous vasodilation in humans (5). The observation of similar leftward shifts in the cutaneous vasodilatory and sweating thresholds with cocaine administration therefore strongly suggests that thermoregulatory alterations are central in origin.

We recognize the limitation of using lidocaine as a control solution. However, lidocaine was required to blind the participants, given the anesthetic properties of cocaine. We are unaware of data supporting the contention that lidocaine alters thermoregulatory or perceptual responses during a heat stress. Moreover, one participant in our study also participated in several studies in which similar heat stresses were performed without cocaine or lidocaine. The participant's responses to these heat stresses were virtually identical to those seen in the lidocaine trial but not to those in the cocaine trial.