OBJECTIVE Severe hypertension responds to treatment with nifedipine given orally or sublingually. Nicardipine hydrochloride, a water soluble dihydropyridine analogue similar to nifedipine, has less of a negative ionotropic effect and produces less reflex tachycardia than nifedipine. Our purpose was to assess the antihypertensive efficacy and safety of intravenous nicardipine in a group of patients with severe hypertension (defined as a supine diastolic blood pressure of more than 120 mm Hg). METHODS Eighteen patients with severe hypertension received treatment with intravenous nicardipine. Nicardipine titration was performed using doses of 4 to 15 mg/hour to achieve therapeutic goal (diastolic blood pressure 95 mm Hg or less or decrease in diastolic blood pressure of more than 25 mm Hg). After this therapeutic end-point was reached, patients received maintainance therapy with nicardipine for varying lengths of time: one hour (Group I), six hours (Group II), or 24 hours. When blood pressure control was lost, patients in Groups I and II entered a second maintenance period lasting a maximum of 24 hours. Onset and offset of action of nicardipine at various infusion rates and times of infusion were measured. RESULTS Onset time to achieve therapeutic response was rapid at 15 mg/hour (0.31 +/- 0.13 hours) when compared with lower doses (1.11 +/- 0.36 hours at 4 mg/hour; 0.54 +/- 0.09 hours at 5 mg/hour; 0.52 +/- 0.09 hours at 7 to 7.5 mg/hour). Those who showed a therapeutic response received maintenance infusions with nicardipine for one (n = 7), six (n = 6), or 24 (n = 5) hours. Sustained blood pressure control at a constant rate of nicardipine infusion was seen in all patients during the maintenance period. After discontinuation of nicardipine, the time for offset of action (increase in diastolic blood pressure of 10 mm Hg or more) was independent of duration of infusion. Decreases in both systolic and diastolic pressures correlated well with plasma nicardipine levels. Heart rate increased by about 10 beats/minute, but this increase did not correlate with plasma nicardipine levels. Side effects were minimal, consisting of headache and flushing. In seven patients, local phlebitis developed at the site of infusion. This occurred after at least 14 hours of infusion at a single site, and the incidence can probably be reduced by shortening the infusion time at a single site. CONCLUSIONS Nicardipine appears to be a safe and effective drug for intravenous use in the treatment of severe hypertension.