Background: The utility of the device software in assessing residual apnea hypopnea index (AHI) on adaptative servoventilation (ASV) treatment has not been evaluated. Objectives: To assess the precision of the smart card device in assessing residual AHI in patients treated with ASV. Methods: Seven patients with central apneas (CA) and 9 patients with complex sleep apnea (CompSA) were offered ASV. A 3 months polysomnography (PSG) with the patient9s device was performed and the AHI recorded on the ASV smart card (ASV-AHI) was compared to the physician recorded PSG-AHI. Results: median baseline AHI was 50 (p25th-75th 48-81) with a central apnea index (CAI) of 37 (p25th-75th 22-57) in CA patients, and 46 (p25th-75th 41.5-60.5) with a CAI of 44 (p25th-75th 19-54.5) in CompSA patients. Two patients refused treatment, 3 used BiPAP autoSV Advanced and 11 AutoSet CS. The mean difference between PSG-AHI and ASV-AHI was 11.9 ± 9.6 (95% limits of agreement -6.90, 30.71), mostly representative of underscoring hypopneas, to a greater extent at a higher residual AHI. According to ASV device software, ASV correctly treated (AHI Conclusions: ASV device software underestimated residual AHI compared to manual scoring of PSG to a greater extent at a higher residual AHI.