Safety and Acceptability of Patient-Administered Sedatives During Mechanical Ventilation.

Only 37 patients were enrolled, and randomization was unbalanced with 17 in the PST-DEX group and 20 in usual care. The trial was a small, unblinded pilot, so it cannot show that patient-administered sedation reduces anxiety or improves clinical outcomes. Patients had to be awake enough to follow commands, communicate, be CAM-ICU negative, and have enough hand strength to press the device, so results do not apply to many early or delirious ventilated patients. Usual care was not standardized beyond the participating ICUs' order sets, and the study did not collect data on daily sedative reductions or spontaneous breathing trials for the two groups. Acceptability was measured with investigator-created questions, and only 13 of 17 PST-DEX patients answered. 59% of PST-DEX patients also received supplemental midazolam, fentanyl, or hydromorphone, so the contribution of self-administered dexmedetomidine alone cannot be separated. The delirium difference, 0 PST-DEX patients versus 4 usual-care patients with p = .058, was a post-hoc finding and needs confirmation. No post-ICU physical or mental outcomes were assessed, and RN satisfaction was not evaluated. Two infusion pump deviations occurred, including one pump that did not recognize the medication barcode.

Do not conclude that patient-administered dexmedetomidine alone reduces anxiety or is safe for all ventilated patients. Anxiety showed no clear change over the 5-day study, 29% of PST-DEX patients had blood pressure or heart rate events, and 59% also received midazolam, fentanyl, or hydromorphone.