The purpose of this study was to determine if depth of halothane anesthesia contributed to the adverse cardiovascular effects of citrate-induced ionized hypocalcemia. Six mongrel dogs were monitored with arterial, central venous, and pulmonary artery flow-directed catheters. Measured end-tidal halothane assured a constant depth of anesthesia, while controlled ventilation and arterial blood gas analysis provided constant acid-base status. Each dog received sodium citrate (USP Fenwal) equivalent to fresh frozen plasma, 1.0 ml.kg-1.min-1, during both deep (D) and light (L) halothane anesthesia. Three dogs received the infusion during L halothane anesthesia first; after a 1-h stabilization period (2.5 h after first infusion) they received a second equivalent infusion during D halothane anesthesia. Three other dogs were studied first with D, then with L halothane. Mean expired halothane (+/- SEM) for group D was 1.52 +/- 0.08%, for group L, 0.85 +/- 0.07%. Significantly greater adverse cardiovascular effects were seen during D halothane anesthesia; four of the six dogs that received citrate during D halothane anesthesia required cessation of the infusion or suffered cardiac arrest. All six infusions during L halothane anesthesia were tolerated. In both groups, significant reductions in ionized calcium [Ca++] (P less than 0.0001) and mean arterial pressure (MAP) (P less than 0.005) were observed; greater reductions in both parameters occurred in group D (P less than 0.0036-0.0005). In group D, but not in group L, cardiac output was depressed compared to baseline (P less than 0.005).(ABSTRACT TRUNCATED AT 250 WORDS)