Methods

The medical charts of all patients who underwent vitrectomy by one experienced surgeon (N.O.) between October 1990 and December 2008 for an idiopathic full-thickness MH were reviewed, and informed consent was obtained from all patients. Eyes with previous vitreous surgery, cystoid macular edema (CME) from any cause, and traumatically induced MH were excluded.

All patients underwent a complete ophthalmic examination including best-corrected visual acuity (BCVA) (measured with the manifest refraction using a standard Japanese vision chart), fundus photography, and slit-lamp biomicroscopy with a contact lens. Fluorescein angiography was performed when indicated clinically. A diagnosis of a full-thickness MH was made by slit-lamp biomicroscopy. After 1998, optical coherence tomography (OCT) was used to confirm the presence of a full-thickness MH.

The data recorded from the medical charts were: age, gender, laterality, BCVA, lens status, stage of the MH, duration of symptoms, MH diameter in disc diameters, axial length (AL), refractive error (RE), surgical procedures, intraoperative and postoperative complications, and length of the follow-up. The RE for pseudophakes was the precataract extraction refraction.

The surgery consisted of a three-port pars plana vitrectomy (PPV) with removal of the posterior cortical vitreous layer for stage 2 and stage 3 MH, removal of any epiretinal membranes (ERM), and fluid–gas exchange. After closing the wound, 1.5 ml 100% sulfur hexafluoride gas was injected into the vitreous with a 30-gauge needle, and an equal amount of air was withdrawn from the midvitreous. A posterior vitreous detachment was induced by suction with a back flush needle, and the posterior hyaloid remaining on the retina was removed by a back flush needle under passive aspiration. Since 1993, triamcinolone acetonide (TA) has been used routinely to ensure that the posterior hyaloid is off or to assist in creating a posterior vitreous detachment. The patients were instructed to maintain a face-down position for at least 1 week after the surgery.

Cataract extraction was performed as a combined procedure with the vitrectomy in all patients older than 40 years. Cataract extraction was performed by phacoemulsification with posterior chamber intraocular lens implantation.

Internal limiting membrane peeling was performed on all eyes beginning in March 1998. The ILM was stained with 0.1% indocyanine green (ICG) from April 2000 and with TA from June 2003. Cryotherapy, argon laser photocoagulation, or both were applied to all intraoperative tears. Adjunctives, such as serum, transforming growth factor-β, and platelets, were not used.

Patients were examined before surgery and after surgery on day 1, 2 weeks, and 1, 3, and 6 months. Thereafter, they were examined every 3 to 6 months. At each visit, a complete clinical examination including BCVA, slit-lamp biomicroscopy with a contact lens, indirect ophthalmoscopy, and fundus photography were performed.

A successful anatomic closure was determined after the gas bubble had resolved (<50%) and careful examination of the macula could be made. Anatomic success was defined as not only a flattening of the MH, but also a disappearance of the edges of the MH.

The purpose of this study was to identify the long-term incidence of a reopening of an MH after vitrectomy with and without ILM peeling and to determine any factors associated with the reopening. Because the reopening of the MH in the ILM-off group was rare, we evaluated the factors related to the reopening only in eyes in the ILM-on group.

The visual acuities were measured with the standard Japanese charts in decimal units that then were converted to the logarithm of the minimal angle of resolution units for the statistical analyses. Categorical variables were analyzed using the Chi-square test, and numerical data were analyzed using paired t tests. The life table method was used to examine the long-term reopening rate with the reopening of the MH as an endpoint. The demographics of the patients with a reopening of the MH were compared with those with a nonreopened MH. Multivariate regression analyses were used to analyze the association between the reopening of the MH and clinical parameters. P < .05 was accepted as statistically significant.