We measured erythrocyte Na + Li + and Na + H + countertransport (CT) activity (millimoles per liter per cell per hour) in 10 healthy control subjects (age, 38 +/- 4 years; body mass index, 25 +/- 1 kg/m 2 ) and in 25 hypertensive patients with non-insulin-dependent diabetes mellitus ([NIDDM] age, 49 +/- 3 years; body mass index, 29 +/- 1 kg/m 2 ; fasting plasma glucose, 157 +/- 12 mg/dL) 4 weeks after discontinuation of previous antihypertensive treatment. Na + Li + CT was significantly increased in hypertensive NIDDM patients compared with controls (0.56 +/- 0.04 v 0.30 +/- 0.30, P < .01), whereas Na + H + CT was similar to control levels (21 +/- 1 v 20 +/- 2). A positive correlation was found between Na + Li + CT and the severity of insulin resistance (r = .69, P < .01), mean arterial pressure ([MAP] r = .64, P < .01), plasma triglyceride concentration (r = .46, P < .05), and plasma total cholesterol (r = .41, P < .05). An inverse correlation was found between Na + Li + CT activity and plasma insulin concentration (r = -.47, P < .05). No relationship was observed between Na + Li + CT activity and either creatinine clearance or proteinuria. Stepwise multiple regression analysis for all metabolic variables and blood pressure showed that only the severity of insulin resistance was positively correlated with increased Na + Li + CT activity. Na + H + and Na + Li + CT activity were not aftered by 3 hours of euglycemic physiologic hyperinsulinemia (84 +/- 3 μU/mL). Hypertensive NIDDM subjects were treated for 3 months with captopril, nifedipine, or doxazosin. After captopril, a reduction of Na + H + CT was observed (22 +/- 4 v 13 +/- 2, P < .05); Na + Li + CT decreased after doxazosin (0.56 +/- 0.06 v 0.45 +/- 0.05, P < .05) and nifedipine (0.52 +/- 0.06 v 0.42 +/- 0.05, P < .05). In conclusion, in hypertensive NIDDM subjects, (1) Na + Li + CT is increased and is correlated with the level of insulin CT activity is reduced by captopril, and Na + Li + CT is decreased by doxazosin and nifedipine.