Evaluation of the optimum time for direct application of fibroblast growth factor to human traumatic tympanic membrane perforations.

Objective: The aim of this study was to determine the optimum time for direct application of basic fibroblast growth factor (bFGF) on large traumatic tympanic membrane perforations (TMPs).

Study Design: Prospective clinical study.

Setting: Tertiary University Hospital.

Methods: Ninety-three patients, with traumatic TMPs greater in extent than 25% of the entire tympanic membrane, were randomized into observation and bFGF-treated groups (~0.2-0.25 mL of bFGF solution was applied directly onto the TM once daily and continued until the perforation closed). Initial visit times were subcategorized into perforation durations of ≤3 and >3 days, thereby rendering two subgroups, as follows: A and B in the observation group; and C and D in the bFGF-treated group. The closure rate and mean closure time were evaluated after 6 months.

Results: Eighty-six patients were finally analyzed. After 6 months, the bFGF-treated group exhibited a significantly higher total closure rate (97.8 versus 82.5%, p < 0.05) and a shorter mean closure time (12.5 ± 3.4 versus 34.0 ± 5.9 days, p < 0.05) compared with the spontaneous healing group. In addition, in the observation group, visiting time was not associated with differences in closure rate (p > 0.05) and mean closure time (p > 0.05), between the A and B subgroups. Similarly, in the bFGF-treated group, visiting time was not associated with differences in closure rate (p > 0.05) between the C and D subgroups. However, the D subgroup was characterized by significantly shortened mean closure time compared with the C subgroup (p < 0.05).

Conclusions: This study shows the beneficial effect of bFGF on human traumatic large TMPs when applied after the 3rd day post-injury had passed (i.e. during the proliferative stage of wound healing). The procedure can not only significantly shorten closure time but can also reduce both the clinical administration duration and occurrence of side-effects associated with bFGF.