Urethane-anesthetized rats were used to study the mechanism of cocaine-induced death. Continuous recording of the changes in five physiological parameters, including respiratory rate (RR), electroencephalogram (EEG), blood pressure (BP), electrocardiogram (ECG), and body temperature (BT), were conducted after intraperitoneal (IP) administration of a single dose of cocaine HCl (70 mg/kg). In the control group (normothermic with core body temperature 37.7 +/- 0.1 degree C and spontaneously breathing), the death rate was 88% (15/17), and the average time to respiratory arrest was 12.99 +/- 1.40 min (mean +/- SEM). The first set of experiments investigated the contribution of hypothermia to cocaine-induced death. The hypothermic group (core body temperature 33.9 +/- 0.3 degrees C and spontaneously breathing) had a death rate of 81.5% (22/27), and an average time to respiratory arrest of 16.70 +/- 1.24 min, which was significantly (p les than 0.05) prolonged. A substantial decrease in respiratory rate was seen in normothermic group, while all the other measured parameters remained relatively stable until respiratory arrest. Sequential arterial blood gas data in this group showed a decrease in PaO2 from 116.0 +/- 5.7 mmHg to 57.7 +/- 4.6 mmHg, an increase in PaCO2 from 27.7 +/- 2.2 mmHg to 42.7 +/- 3.0 mmHg, and a decrease in pH from 7.467 +/- 0.039 to 7.357 +/- 0.003. To confirm that respiratory depression was an important mechanism of cocaine-induced death in this model, ten normothermic rats underwent mechanical ventilation, and all survived cocaine exposure. This study points to the important role of respiratory depression as a cause of cocaine-induced death.