Noninvasive alternatives to esophageal pressure (Pes) are needed to evaluate respiratory effort during sleep. Pulse transit time (PTT) is the time taken for pulse pressure to travel from the aortic valve to the periphery. PTT has been shown to be inversely correlated with blood pressure, and can reveal acute changes generated by high pleural pressure swings during pulsus paradoxus. A close relationship has been demonstrated between the increase in Pes and a progressive rise in the amplitude of PTT oscillations. The aim of the present study was to assess the accuracy of PTT for the classification of sleep respiratory events as central or obstructive. Respiratory events occurring during sleep were randomly chosen from 13 unselected male patients (mean apnea-hypopnea index [AHI] = 25.1 per hour of sleep; age = 47.3 yr, body mass index [BMI] = 27.1 kg/m2). Two observers experienced in polysomnography classified 177 events on the basis of the "gold standard method": the measurement of Pes. For 167 events about which the observers agreed, the PTT signal was analyzed visually and independently by the two observers blinded to Pes, in order to reclassify the same sleep respiratory events. The two observers were in agreement for 94.6% of the events scored visually on PTT recordings. We evaluated sensitivity (Se) (Observer 1: 94%, Observer 2: 91%), specificity (Sp) (97% and 95%, respectively), negative predictive value (NPV) (95% and 92%, respectively), and positive predictive value (PPV) (96% and 94%, respectively), of PTT with Pes as the reference. Misclassifications of respiratory episodes were usually due to artifacts or baseline variations of the PTT signal (57%), and occurred during rapid eye movement (REM) sleep (42.8%). PTT has shown a high sensitivity and specificity in differentiating obstructive and central respiratory events, and may become the reference noninvasive tool for this purpose.