Anaerobic threshold measurements determined either invasively by analysis of arterial lactate concentration (lactate threshold) or noninvasively by respiratory gas exchange analysis (ventilatory threshold) were compared in patients with chronic congestive heart failure. Sixteen patients performed symptom-limited maximal exercise on a bicycle ergometer using a continuous ramp protocol with measurement of arterial lactate concentration at 1 minute intervals, and continuous breath-by-breath analysis of respiratory gas exchange. A specific lactate threshold point was detected in only 7 patients. These 7 patients had significantly greater peak oxygen uptake than did the 9 in whom no specific lactate threshold point was detected (15.9 +/- 1.0 vs 10.5 +/- 0.5 ml/kg/min; p less than 0.05). Ventilatory threshold significantly correlated with lactate threshold in these 7 patients. In the remaining 9 patients, neither lactate nor ventilatory threshold could be reliably determined with methods used in the present study.