Another interesting point arising from the results of Chouri-Pontarollo et al3 concerns the impact of bilevel NIV in OHS. Indeed, 7 of 15 of their patients had low CO2 response at baseline, displaying a typical situation of chronic alveolar hypoventilation. No clinical characteristics could differentiate the patients having normal or impaired CO2 sensitivity at baseline since they had similar body mass index, similar daytime hypoxemia and daytime hypercapnia levels, similar indexes of nocturnal respiratory disturbances, and similar severity of desaturation during sleep. According to our current understanding of how NIV works, we would have anticipated that domiciliary NIV would have reset the respiratory centers and therefore improved CO2 sensitivity, but this was not the case for five of seven patients (71%). Clearly, NIV improved Paco2, restored sleep architecture (increased stage 3–4 and rapid eye movement sleep), corrected nocturnal desaturations, and decreased respiratory disturbance index and microarousals. The mechanisms of such sleep and daytime Paco2 improvements, without any significant change in CO2 sensitivity and without concomitant weight loss, have to be clarified. Because assessments were performed after only 5 nights under NIV, further studies should appraise the time course evolution of ventilatory responses in OHS patients receiving domiciliary NIV.