This study evaluated the effect of limiting maximal infusion-pump flow rate on suppression of remifentanil-induced cough during target-controlled infusion. Two hundred and ten patients were randomly assigned to receive remifentanil at an effect-site concentration of 4.0 ng.ml(-1) with maximal flow rate limited to 100 (group R(100)), 200 (group R(200)), or 1200 ml.h(-1) (group R(1200)). The number of episodes of cough were recorded and graded as mild (1-2), moderate (3-4), or severe (5 or more). The incidence of cough was 2.9% in group R(100), 5.7% in group R(200) and 25.7% in group R(1200). Patients in group R(100) and R(200) had a significantly lower incidence of cough than those in group R(1200) (p < 0.05). Zero, two and five patients coughed a moderate amount in groups R(100), R(200) and group R(1200), respectively (p < 0.05). Limiting maximal infusion rate during remifentanil TCI suppressed remifentanil-induced cough.