Figure. The cumulative incidence of macular hole (MH) bilaterality can be described by a hyperbolic function, y = 2.6 + 29.8 x/(130.1 + x), with R2 = 0.99. Curve-fit analysis showed that the estimated risk of MH in the fellow eye was 12.0% at 5 years and 16.9% at 10 years.
Kumagai K, Ogino N, Hangai M, Larson E. Percentage of Fellow Eyes That Develop Full-Thickness Macular Hole in Patients With Unilateral Macular Hole. Arch Ophthalmol. 2012;130(3):393-394. doi:10.1001/archopthalmol.2011.1427
Author Affiliations: Shinjo Ophthalmologic Institute (Drs Kumagai and Ogino) and Miyazaki Prefectural Nursing University (Dr Larson), Miyazaki, and Department of Ophthalmology and Visual Sciences, Graduate School of Medicine, Kyoto University, Kyoto (Dr Hangai), Japan.
Patients with a unilateral macular hole (MH) have an increased risk of developing an MH in the fellow eye.1- 4 However, to our knowledge, the incidence of developing an MH in the fellow eye has not been analyzed in a large cohort of eyes after macular hole surgery. The purpose of this study was to determine the probability of developing a full-thickness MH in the fellow eyes of patients with a unilateral MH.
A retrospective longitudinal study of 1082 patients with a unilateral, idiopathic, full-thickness MH who underwent vitrectomy by one of us (N.O.) between October 1990 and December 2010 was conducted. All of the patients were confirmed to have a unilateral full-thickness MH at the initial visit by dilated indirect slitlamp biomicroscopy. Patients with any other fundus diseases or history of ocular trauma or surgery in either eye were excluded.
Kaplan-Meier life-table analysis was used to estimate the risk of developing an MH in the fellow eye. In addition, the cumulative incidence of bilateral MHs was fit to a hyperbolic function: G = Gmax × T /(Tm + T), where the visual gain (G) was defined as the preoperative best-corrected visual acuity minus postoperative best-corrected visual acuity in logMAR units; the maximum visual gain (Gmax) was defined as the preoperative best-corrected visual acuity minus final best-corrected visual acuity in logMAR units; the average visual gain was plotted as a function of the postoperative time (T) in months; and Tm was defined as the postoperative time required to reach one-half Gmax. This equation was found earlier to describe the recovery of visual acuity after treatment of different macular diseases.5
There were 394 men and 688 women in the study. The mean (SD) age at the initial surgery was 64.2 (8.3) years (range, 21-95 years). The mean (SD) follow-up period was 71.8 (49.6) months (range, 6-246 months).
Nine hundred sixty patients (88.7%) remained with a unilateral MH (unilateral group) and 122 patients (11.3%) developed an MH in the fellow eye (bilateral group). The sex distribution, age at onset in the first eye, and axial length in the first eye were not significantly different between the unilateral and bilateral groups (Table).
We defined the interval between the onset of the first MH and that in the second eye as the bilateral interval. If the second eye developed an MH within 1 month of onset in the first eye, the bilateral interval was set to 0. The mean (SD) bilateral interval among all patients was 26.1 (28.0) months (range, 0-122 months). The difference in the mean bilateral interval between men and women was not significant (P = .38). The age at onset of an MH in the first eye and its axial length were not significantly correlated with bilateral interval.
The risk of the fellow eye developing an MH estimated by the Kaplan-Meier method was 11.6% at 5 years and 16.7% at 10 years. The cumulative incidence of bilaterality can be described by the following hyperbolic function: y = 2.6 + 29.8 x/(130.1 + x), with R2 = 0.99 (Figure). Curve-fit analysis showed that the estimated risk of the fellow eye developing an MH was 12.0% at 5 years and 16.9% at 10 years.
Earlier retrospective studies reported that the incidence of developing an MH in the fellow eye with or without a posterior vitreous detachment was 22% for a mean follow-up of 57 months (37 patients)1 and 13% within 48 months (340 patients).2 Ezra et al3 reported that the incidence of developing an MH in the fellow eye without a posterior vitreous detachment (114 patients) was 15.6% at 5 years by Kaplan-Meier analysis. Although the long-term incidence of developing an MH in the fellow eye may depend on the patient demographic characteristics and vitreoretinal interface features, our large-scale study showed that the cumulative incidence of bilaterality was well fit by a hyperbolic function. The findings of the curve-fit analysis suggested that the estimated risk was 21.9% at 20 years and 24.5% at 30 years, although these estimates will have to be confirmed by longer longitudinal studies. Because the appearance of the vitreoretinal interface in spectral-domain optical coherence tomographic images is associated with the risk of developing an MH in the fellow eye,6 further studies are required to determine the long-term risk in the fellow eye based on spectral-domain optical coherence tomographic features.
Correspondence: Dr Kumagai, Shinjo Ophthalmologic Institute, 889-1 Mego Shimokitakata-machi, Miyazaki, Japan 880-0035 (firstname.lastname@example.org).
Author Contributions: Dr Kumagai had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Financial Disclosure: None reported.