To validate the accuracy and consistency of respiratory inductive plethysmography (RIP) in measuring tidal volume after an overnight sleep, tidal volumes of 18 patients with suspected sleep-disordered breathing and 8 normal volunteers were measured simultaneously with RIP (VTRIP) and with an ultrasonic airflow meter (VTUFM) before and after an unstrained overnight sleep on supine and lateral decubitus. The bias of the VTRIP was expressed as (VTRIP-VTUFM)/ VTUFM.100%, limits of agreement between VTRIP and VTUFM was measured by averaged bias +/- 2 s. Results showed that in normal subjects, the bias of RIP before and after overnight sleep was precise and consistent in both supine (0.7% and -1.6%) and lateral decubitus (3.7% and -0.56%). In these patients, the bias of RIP before and after sleep in supine also remained small (1.9% and 1.7%), but it became larger in lateral decubitus (24.5% and 20.4%) and 11.5% exceeded the limits of agreement observed in the evening. The patients' body mass indices (BMI) were higher than those of normal subjects (median 34.2 vs. 27.8 kg/m2). Pooled data showed that the bias of VTRIP in the morning on lateral decubitus but not on supine was correlated to BMI (Spearman R = 0.32, n = 52, P = 0.02). Thus, we were led to conclude that the accuracy of VTRIP overnight was precise and consistent in normal subjects, but the deviation of VTRIP measured on lateral decubitus in patients especially in those with excessive obesity was greater, thus, the method should not be used for quantitative determination.