Abstract OBJECTIVES In healthy individuals, increasing pulmonary blood flow during exercise also increases the % of the diffusing capacity of the lungs for carbon monoxide (DLCO%), but its evolution after lung resection is unknown. In this study, our goal was to measure changes in exercise DLCO% during the first 3 days after anatomical lung resection. METHODS We performed a prospective observational study on consecutive patients with non-small-cell lung cancer scheduled for anatomical resection, except pneumonectomy, during a 6-month period. Patients underwent measurement of the DLCO% by a single-breath technique adjusted by the concentration of haemoglobin—before and after standardized exercise the day before and 3 consecutive days after surgery. The delta (Δ) variation (basal versus exercise) was calculated. The number of functioning resected segments was calculated by bronchoscopy. Postoperative pain and pleural air leak were estimated using a visual analogue scale and graduated conventional pleural drainage systems, respectively, and their influence on ΔDLCO each postoperative day was evaluated by linear regression analysis. RESULTS Fifty-seven patients were included. The visual analogue scale of pain and pleural air leaks were not correlated to Δ values (model R2: 0.0048). The evolution of Δ values during 3 postoperative days showed a progressive recovery of values, but on the third day, DLCO% capacity during exercise was still impaired (P < 0.01), especially in patients who underwent a resection of more than 3 functioning segments. CONCLUSIONS Physiological increase in DLCO% during exercise is still impaired on the third postoperative day in patients undergoing resection of more than 3 functioning pulmonary segments. This fact should be considered before discharging those patients after anatomical lung resection.