A comparison of pain measures in newborn infants after cardiac surgery.

Accurate pain assessment tools to evaluate pain in critically ill neonates in the postoperative period are lacking. Therefore, we compared a number of potentially useful indices of pain in critically ill neonates following cardiac surgery. Eighty-one full-term infants were studied during the first 48 postoperative hours and the following indices were measured: heart rate, mean arterial blood pressure, heart-rate variability, urinary and plasma cortisol, and 4 composite pain measurement scales: Children's and Infants' Postoperative Pain Scale (CHIPPS), CRIES, COMFORT, and Premature Infant Pain Profile (PIPP). Regression models were used to investigate relationships between individual pain indices or composite pain assessment scales with respect to procedural intensity and opioid dose and plasma levels. COMFORT score performed best, with a 27% difference in score between procedures causing tissue damage and those that did not (P<0.001). COMFORT score and the high-frequency component of heart-rate variability showed inverse correlations with opioid dose and plasma levels over the first 48 hours postoperatively, but after accounting for clinical variables, only COMFORT score remained significant (eg, 52% of variance in morphine level at 24 hours, P<0.001). The factor structure of the COMFORT score revealed that both behavioural and physiological variables account for a significant proportion of the variance (45% and 15%, respectively; P<0.001). Plasma concentrations of cortisol increased postoperatively but urinary cortisol excretion did not change significantly. Of the pain indices studied, the COMFORT score performed best, with both behavioural and physiological components providing significant contributions.