Initial healing rates of venous ulcers: are they useful as predictors of healing?

Abstract

Clinical trials that follow venous ulcers to complete healing can be costly because of the prolonged healing time involved. Initial healing rates in venous ulcers are calculated by 2 methods, which are based on a metric using wound area and perimeter. It has been proposed that these rates allow the prediction of complete healing and that they may be useful as surrogate end points for clinical trials. The objective of this study was to compare the 2 proposed methods for calculating initial healing rates and determine their usefulness in predicting the healing of venous ulcers. Venous leg ulcers from patients enrolled in a randomized, double-blind, placebo-controlled study were measured weekly for up to 12 weeks. Their healing status was determined for up to 24 weeks. Initial healing rates were calculated using the 2 proposed methods. The ability of these rates to predict time to complete healing was assessed. Information from 17 patients was available. The initial healing rates, calculated by either method, were quite similar; both methods produced the same median value of 0.046 cm/week in our patients. Five of the patients had negative initial healing rates, which do not allow any prediction of a healing time. Three of 7 patients predicted to heal within 24 weeks failed to do so. One of the 5 patients was predicted to heal at some time after 24 weeks but actually healed within 24 weeks. None of the 5 patients with negative initial healing rates healed within 24 weeks. Initial healing rates, as calculated by either method, have limited utility in describing healing curves and predicting a healing time. This poor predictive ability argues against using these initial healing rates as surrogate end points for clinical trials. The great variability observed in venous ulcer healing curves may limit the development of useful predictive models in this patient population.