<p><b>OBJECTIVE</b>To explore the influence of dermal defect and fat dome structure destruction in burn wounds on the formation of hyperplastic scar.</p><p><b>METHODS</b>Fifty two wounds in 24 burn patients with deep partial thickness burn indicating tangential excision in the extremities were enrolled in the study, and they were divided into three groups according to the extent of exposure of dermal fat granules, i.e. A (without fat exposure), B (with little fat exposure) and C (with much fat exposure) groups. These three groups were subdivided into A1 (without grafting), A2 (grafting with razor thin skin), B1 (without grafting), B2 (with razor thin skin grafting), C1 (without grafting) and C2 (with split-thickness skin grafting) groups, with 9 wounds in each group. The dermal depth and exposure rate of the fat granules in each group were measured and analyzed by KS400 photography analysis apparatus. The follow-up conditions of the scars 6 months after operation were evaluated with Vancouver remark system by Vancouver score assessment.</p><p><b>RESULTS</b>There was obvious difference in the dermal depth and exposure rate of the fat granules among all the groups (P < 0.05 or 0.01). The fat exposure rate was positively correlated with the extent of the dermal defect (gamma = 0.554, P < 0.05). The Vancouver score in group A was lower than that in B and C groups (P < 0.05), while that in B1 group (3.714 +/- 2.498) was evidently higher than that in other groups (P < 0.01). The scar score was lowered when the wounds were grafted with the dermis with its thickness similar to the depth of the defect, The scar score was increased along with the elevation of fat exposure rate (P < 0.05).</p><p><b>CONCLUSION</b>There was a positive correlation between the degree of dermal defect and that of hyperplastic scar after burns. The disruption of fat dome structure might also be an important factor in the scar development.</p>