The aim of the present study was to observe the effect of calcitonin gene-related peptide (CGRP) on wound contraction in both denervated and normal areas. A total of 100 Wistar rats, each of which had a 2 x 2 cm full-thickness skin defect on the back, were divided into two main control groups and six corresponding experimental groups in which 10 x 10(-9) M and 10 x 10(-12) M synthetic rat CGRP were given intraperitoneally and intradermally. Contraction was assessed weekly with planimetry, and mean surface areas (mm2 +/- SD) of related groups were compared. Complete closure took 4 weeks for the normal control group and 8 weeks for the denervated control group (p < 0.05). The 10 x 10(-12) M CGRP with both types of application showed a decrease in the length of time for complete closure to 3 weeks in the normal experimental groups (p < 0.05), but complete closure still took 4 weeks in normal groups in which 10 x 10(-9) M CGRP was given (p < 0.05). Any dosage of CGRP given intraperitoneally showed no change in the closure period in the denervated experimental groups (p > 0.05). CGRP showed a trophic effect on healing by an increased rate of contraction in the rat model. However, the neural supply to the wound area seemed to be intact because of the necessity of axonal transfer of CGRP.