Development of an observational scale for the assessment of postoperative pain in infants.

In a prospective trial in 139 infants ASA classification 1-11 13 observational items were scaled during the first postoperative hour (13 assessments). The items were drawn from the literature and chosen for economic purpose. Factor analyses (Principal component, Kaiser Criterion, Scree-test) were used for the elimination of useless items and for the identification of suitable ones. The discriminative properties of single items and different subsets of items to detect an analgetic demand were tested in discriminant analyses and variance analyses with repeated measurements. Due to insufficient variance four items had to be eliminated: "nasolabial folding", "colour of the face", "sweating of the head", and "muscle tone". The factor analysis if the remaining 9 items resulted in a one factorial solution. Neither the corrected item-scale-correlations nor the inter-item-correlations showed advantageous properties of single items compared with the others. For economic reasons two 5-item scales were chosen for further evaluation in regard to sensitivity, specificity and validity. The items "crying", "facial expression", "positioning of the legs", "positioning of the trunc" and "motoric restlessnes" built the Children's and Infants Postoperative Pain Scale (CHIPPS) wheras an Infants Postoperative Pain Scale (IPPS) contained the items "crying", "facial expression", "positioning of the arms", "positioning of the trunc" and "motoric restlessnes". The latter five items had shown the highest factor loadings. The two systems had a high intern consistency with alpha > 0.90 (p < 0.01) with at least 73% explained variance. Inter-item-correlations and corrected item-scale-correlations showed no differences between the two scales. The discriminant analyses resulted in almost identic data for specificity, sensitivity and predictive values of the IPPS compared with the CHIPPS. There was a significant interaction between repeated measurements and the supply of Piritramide and Ketamine, but not of Midazolam. Concurrent and constructive validation were positive for both systems, using administration of Piritramide as a criterion. For clinical purpose the CHIPPS should be preferred, because it has been proven to be valid in children up to 4 years of age and because controlled data on its sensitivity, specifity, reliability and validity could already be presented.