We would like to comment on the manuscript entitled “Utility of basic fibroblast growth factor in the repair of blast-induced total or near-total tympanic membrane perforations: A pilot study” by Lou et al. . The authors investigated whether basic fibroblast growth factor (bFGF) could be used to repair blast-induced total or near-total tympanic membrane perforations (TMPs) . The authors concluded that direct application of bFGF was a promising minimally invasive alternative to conventional tympanoplasty, affording a comparable success rate. The closure rate was higher than that of spontaneous healing. Edge inversion did not affect healing outcomes. The use of bFGF in this setting would have immediate therapeutic applications for military personnel stationed in isolated environments who develop blast-induced TMPs . This is an interesting paper and constitutes excellent work. The authors first report that topical bFGF repaired blast-induced TMPs; the success rate was comparable to that of tympanoplasty . In addition, bFGF treatment is simple; there is no need for hospitalization and military personnel can return to their duties rapidly. Unfortunately, none of the inclusion criteria, or the pre- or post-operative auditory records, are adequately described. Blast-induced TMPs are accompanied by sensorineural hearing loss. The authors should have carefully explored whether bFGF treatment improved such trauma-induced hearing loss.

The authors simply write (in the Materials and Methods section): “The inclusion criteria were: a traumatic TMP involving ≥ 75% of the entire tympanic membrane (TM) and caused by a blast explosion , regular follow-up, and no middle ear infection at the time of the hospital visit.” and “The TM was treated once daily by the direct application of 0.10–0.15 mL (2–3 drops) of recombinant bovine bFGF solution (21000 IU/5 mL). The surface and edge of the TM were kept slightly moist. The edge of the perforation was not approximated, and no scaffolding material was used. The first bFGF application was performed by a physician. Thereafter, the drops were applied daily at home by the patients themselves, as instructed, until complete closure of the perforation was confirmed by a physician. In patients with profound sensorineural hearing loss, intravenous steroids and vasodilators were administered simultaneously for 7–10 days.”. The authors did not explain whether patients with myringosclerosis of the perforation edges were included in the study. Myringosclerosis is a negative prognostic factor in terms of myringoplastic outcomes. Ayache et al. found that myringosclerosis at the perforation edge increased the failure rate of fat-plug myringoplasty . Hakuba et al. found that failure of myringoplasty closure upon use of an atelocollagen and bFGF correlated significantly with myringosclerosis. In a study of the use of FGF-2 to repair traumatic TMPs, Lou et al. earlier used multivariate logistic regression analysis to show that histories of chronic otitis media and myringosclerosis were significant risk factors for treatment failure. The cited authors should clarify, in the future, whether myringosclerosis at the perforation edge affected FGF-2-induced repair of blast-induced TMPs. Also, in their current work , the authors did not describe how they managed the inverted edges. Blast-induced TMPs usually have inverted edges and fractured eardrum epithelia, both of which may cause middle-ear cholesteatoma . Thus, processing of inverted edges and the fractured epithelium is vital prior to FGF-2 treatment.

The authors write, in the Results section: “The average TMP size (17 patients) was 92.8 ± 3.6%. Six months later, 16 of the (94.1%) TMPs had completely closed; the mean time to closure was 28.4 ± 10.9 days (range, 15–47 days).”, and “The air-bone gaps pre- and post-treatment were 35.0 ± 8.2 and 10.5 ± 4.9 dB, respectively, in the 16 patients in whom TMP closure was achieved. This improvement was statistically significant according to both the t -test (P = 0.031) and Wilcoxon signed rank test (P = 0.028). Pure tone audiometry bone conduction thresholds improved in 5 cases with mixed loss concerning the low and middle frequencies, only one case with mixed loss concerning the high frequencies did not improve. In addition, mild otorrhea was observed in 11 patients.” The success rate (94.1%) was indeed high (comparable to that of tympanoplasty; 2–4) and the improvement in the air-bone gap excellent, but the authors did not measure pure-tone bone-conduction thresholds. Most blast-induced TMPs are accompanied by sensorineural or mixed hearing loss . It is very important to measure changes in the bone-conduction threshold. If FGF-2 indeed improves the bone-conduction thresholds of perforations accompanied by sensorineural or mixed hearing loss, FGF-2 repair of blast-induced total or near-total TMPs would be promising. Many experimental studies have shown that topical FGF-2 facilitates the recovery of injured facial nerves and damaged brain neurons ; other studies found that bFGF protects cochlear neurons and hair cells from acoustic trauma . Sekiya et al. showed that therapeutic bFGF application ameliorated trauma-induced cochlear nerve degeneration. Recent technological advances in deafness treatment, such as cochlear and auditory brainstem implants, in combination with neurotrophic and/or growth factor therapies, may be of great potential benefit for patients with hearing loss. However, Yamasoba et al. , in an experimental study, found that neither exogenous FGF-1 nor FGF-2 ameliorated noise-induced hair cell damage. The FGFs are thus not good candidates for auditory hair cell protection in vivo. Future studies should measure changes in bone-conduction thresholds before and after FGF-2 treatment of blast-induced total or near-total TMPs.