Background: The aims of this prospective study were to analyze the predictors of postoperative sleep disturbance after esophagectomy for cancer and to identify patients at risk for postoperative hypnotic administration.
Methods: Sixty two consecutive patients who underwent cancer-related esophagectomy were enrolled in this study from May 2011 to February 2012. Data about perioperative management, postoperative complications, ICU stay, and vasopressor, hypnotic, and painkiller administration were retrieved. The EORTC QLQ-C30 was used and global quality of life (QL2 item) and sleep disturbance (SL item) were the primary endpoints. Univariate and multivariate analyses were performed.
Results: Postoperative request of hypnotics independently predicted bad quality of life outcome. Sleep disturbance after esophagectomy was independently predicted by the duration of dopamine infusion in the ICU and the daily request of benzodiazepines. Even in this case, only sleep disturbance at diagnosis revealed to be an independent predictor of hypnotic administration need. ROC curve analysis showed that sleep disturbance at diagnosis was a good predictor of benzodiazepine request (AUC = 73%, P = 0.02).
Conclusions: The use of vasopressors in the ICU affects sleep in the following postoperative period and the use of hypnotics is neither completely successful nor lacking in possible consequences. Sleep disturbance at diagnosis can successfully predict patients who can develop sleep disturbance during the postoperative period.