Pain evaluation in the intensive care unit Observer-reported faces scale compared with self-reported visual analog scale

Background and Objectives. The visual analog scale (VAS) is a simple and sensitive mean of pain assessment. The faces scale is also a simple, self-reporting method for children. Facial signs of pain have not been used to assess pain in postoperative adult patients in the intensive care unit (ICU). Methods. Fifty patients undergoing esoph ageal cancer surgery by a thoracoabdominal procedure were studied. Epidural opioids, such as morphine or buprenorphine, combined with bupivacaine were administered during and after surgery. Pain measurement was performed by a physician in the ICU using the self-reported VAS 0.5, 1, 2, 4, and 6 hours after tracheal extubation and thereafter every 4 hours during the stay in the ICU. A nurse who was unaware of the patients' VAS scores assessed facial expression as a measure of pain intensity using a five-grade faces scale immediately before pain evaluation by VAS. The VAS was rescaled into five discrete units that would match the five faces scale scores. Weighted kappa statistics were used to establish a relative level of agreement between the five-grade VAS and faces scale. Results. Good agreement was found between the five-grade VAS and the faces scale 30 minutes and 1 hour after tracheal extubation (weighted kappa values .67 and .62, respectively). The VAS and faces scales were measured 7-13 times per patient during the stay in the ICU, and 518 observations were collected. Although moderate agreement was found between the ave-graded VAS and faces scale for all pairs of observation: (weighted kappa values .54), less agreement was found between them in patients with moderate pain. In addition, the calculated mean differences between the five-graded VAS and faces scale differed significantly between patients. Conclusion. The faces scale may be useful for pain evaluation in the ICU.