Ventilator induced diaphragm dysfunction may occur through a variety of mechanisms, the most established being diaphragm disuse atrophy. Diaphragm biopsy is the most definitive method of evaluating diaphragm atrophy, but its invasive nature limits its use in critically ill patients. Therefore, ultrasonographic measurements of diaphragm thickness are commonly used to evaluate diaphragm atrophy. However, whether all changes in diaphragm thickness observed on ultrasound are specific to structural changes due to atrophy or hypertrophy versus the conditions under which measurements are taken (ie, ventilator mode and presence of sedation) is uncertain. We conducted a prospective observational study comparing diaphragm thickness measurements on spontaneous and controlled modes of mechanical ventilation (MV) in a randomized crossover fashion. Adult patients receiving invasive MV for less than 48 hours were included. Pressure support was adjusted to ensure that spontaneous tidal volume was similar to the tidal volume under assist control. Positive end expiratory pressure and fraction of inspired oxygen were not changed between measurements. In patients receiving sedatives, additional measurements were taken on assist control after an interruption of sedatives and analgesics. Measurements of the diaphragm were taken at end expiration (Tdi) and peak inspiration, allowing calculation of diaphragm thickening fraction (TFdi). The investigator obtaining the images was blinded to the mode of MV. 85 patients were enrolled, 66 with images on controlled and spontaneous modes and 40 with images before and after an interruption of sedatives. There was no difference in tidal volume between comparisons. When comparing controlled and spontaneous modes, no difference was observed in Tdi (mean difference-0.02 mm, 95% CI-0.09 to 0.05 mm, p= 0.62) or TFdi (mean difference 2.2%, 95% CI-2.8% to 7.2%, p= 0.38). Tdi increased after an interruption of sedatives (mean difference 0.09 mm, 95% CI 0.02 to 0.15 mm, p= 0.012), representing a mean increase of 6%(95% CI 1% to 10%). No difference was observed in TFdi after an interruption of sedatives (mean difference 2.3%, 95% CI-1.6% to 6.2%, p= 0.24). The median length of sedative interruption was 15 minutes (IQR 10 to 20 min). There was no significant difference in Tdi when comparing measurements taken on pressure support and assist control modes. A short interruption of sedatives resulted in an increase in Tdi. The effect of sedatives on diaphragm thickness measured by ultrasound should be considered in future studies assessing longitudinal changes in diaphragm thickness.