Successful cardioversion is dependent on the delivery of sufficient current. Current is determined by energy and transthoracic impedance (TTI). Our purpose was to assess factors affecting TTI in humans. Twenty-eight patients undergoing elective cardioversion were monitored up to 48 hours after shock delivery. We also studied 10 normal subjects, who did not receive shocks, for comparison. TTI declined 8% in the first hour in patients who received shocks, but also 6% in the normal subjects who wore the same pads for 1 hour but received no shocks. Twenty-four hours later, TTI returned to baseline in the nonshocked subjects but remained reduced (93% of control, p less than 0.05) in the shocked patients. There was a correlation between the total energy delivered and the decline in TTI (r = 0.69). TTI was 9% lower at end-expiration than at end-inspiration (55 +/- 3 vs 50 +/- 3 ohms, mean +/- standard error of the mean, p less than 0.01, paired t test). In the normal subjects, when a nonsalt-containing coupling agent (ultrasound coupling gel) was compared with a salt-containing gel (Redux paste), TTI was 20% higher (65 +/- 5 vs 54 +/- 4 ohms, p less than 0.01) with the nonsalt-containing gel. When no coupling agent was used, TTI was markedly higher, 160 +/- 18 ohms (p less than 0.01 vs control). After 1 hour, TTI decreased 6% in the normal subjects when salt-containing gel was used, but did not change when a nonsalt gel was used.(ABSTRACT TRUNCATED AT 250 WORDS)