The preferred method of vascular access for routine hemodialysis is via an arteriovenous fistula into which two needles are inserted. Single-needle access is an alternative to this method. Thorough evaluations of the efficiency of the single-needle method are lacking. The present study evaluates the weekly time-averaged urea concentrations (TACUrea) and protein intake (PI) in 76 patients routinely treated with single-needle dialysis on a chronic basis. Furthermore, other clinical parameters of dialysis adequacy, such as hematocrit, nerve conduction velocity, hospitalization rate, and cumulative survival also were evaluated, as well as fistula survival. TACUrea appeared to be 53.2 +/- 6.0 mg/100 mL, a value corresponding to adequate two-needle dialysis with low morbidity, as outlined by the American National Cooperative Study Group. PI was evaluated from urea generation rates (G) and from outpatient dietary records. G was 6.07 +/- 2.42 mg/min, and the corresponding PI was calculated to be 1.07 +/- 0.28 g/kg body weight/24 h. PI estimated from dietary records was 1.14 +/- 0.43 g/kg body weight/24 h. The relation of total clearance over distribution volume (KT/V) averaged 0.98 +/- 0.23. Mean hematocrit and nerve conduction velocity during 2 years of follow-up ranged between 24% and 26% and 38 and 40 m/s, respectively. Hospitalization rate during 1 year was 18%. Five years' cumulative survival was 64% for the period 1975 to 1985, and 79% for the period 1980 to 1985. Five years' fistula survival was 74%, a value higher than in four studies on two-needle dialysis of comparable extent. It is concluded that urea kinetic data and other parameters of dialysis adequacy indicate that the efficiency of the single-needle technique is at least as good as that obtained in the more currently used two-needle technique. Subsequently, the current reluctance towards single-needle dialysis as a routine procedure in chronic renal failure, appears to be unjustified.