[Initial results of the reliability and validity of a German-language scale for the quantitative measurement of postoperative pain in young children].

In a previous study it was proved impossible to evaluate the validity of a system modified according to McGrath et al. for measuring postoperative pain in children. Three reasons were postulated for this result. The analgesics used did not modulate the pain in such a manner that the statistical analysis was effective; or the postoperative pain had an acute onset or fluctuated and the regular observations did not allow detection of the pain peaks; or the system used did not measure pain. Therefore, a new and different approach was adopted to evaluate the reliability and validity of a scoring system for the measurement of postoperative pain in small children. MATERIAL. Sixty children classed as ASA I and II and aged between 1 and 5 years were randomly allocated to 2 groups: one group received Tramadol (0.5 mg/kg) after the induction of anesthesia and the other group received placebo. Premedication and anesthesia were standardized. All indications were accepted except for urgent operations and painful diseases. For 1 h after the operation the children were continuously observed by an experienced pediatric anesthetist who did not know what drugs had been given. Every 15 min and when the observer was convinced that the children were in pain and needed analgesics the following items were scaled: wakefulness, wake-up reaction, crying, facial expression, position of the trunk, position of the legs, muscle tone, making contact, verbal communication on being asked about pain, special defense against stimuli and consolability. The items were included in a factor analysis (principal components). The number of the factors was detected using the Scree test. The only items accepted as reliable were those that had a substantial load of at least 0.4 at each measurement. The reliability coefficient was computed with the split-half technique (odd-even). The validity was estimated with a multifactorial analysis of variance with repeated measurements. When the observer was convinced that any child was in pain, Metamizol (15 mg/kg) was administered. Significance was assumed at P less than 0.05. The study was accepted by the ethical committee of the Ruhr University, Bochum. RESULTS. A replication verified the results of the previous study: the factor analysis resulted in a one-factorial solution. Of the 8 items (wakefulness, wake-up reactions, crying, position of the trunk, position of the legs, verbal communications on being asked about pain, special defense against stimuli) only 4 had a sufficiently substantial load on all measurements (crying, facial expression, position of the trunk and position of the legs). (ABSTRACT TRUNCATED AT 400 WORDS)