We reviewed the records of the patients who underwent a pars plana vitrectomy for idiopathic ERM between January 2004 and February 2008. All of the subjects had a complete ocular examination to exclude possible peripheral retinal breaks or other anomalies which could result in ERM formation. Exclusion criteria included a history of ocular or systemic conditions known to be associated with macular ERM formation, potentially irreversible significant loss of vision, less than seven months of follow-up, and clinical or angiographic evidence of a vascular occlusion or other anomaly involving the macula. Thirty eyes from 30 subjects were included in this study. The patients underwent a standard three port pars plana vitrectomy assisted with triamcinolone and membrane peeling. The procedures were performed by a single surgeon and used high-magnification viewing and intraocular forceps. Because the ERM was peeled without identification of the internal limiting membrane (ILM), the ILM was unintentionally peeled off together with the ERM in some patients. Concomitant cataract surgery was performed if necessary. Preoperative and postoperative examinations, performed at every follow-up visit, included the best-corrected visual acuity using a standard Korean visual acuity chart (Han's chart). The examinations also included a slit-lamp examination, fundus examination, fundus photography, and OCT. Patients followed-up every one or two months until they showed no further changes in visual acuity. After they were stable, they were followed every three to six months according to their postoperative status.

Optical coherence tomography was performed using commercially available equipment (software 3.0; Carl Zeiss Meditec, Dublin, CA, USA). After papillary dilation, measurements of the average thickness of the central 1 mm of the retina before and after surgery were compared. All of the subjects were divided into two groups according to the concomitant cataract surgery. Eighteen eyes underwent a vitrectomy with concomitant cataract surgery; 12 eyes underwent a vitrectomy alone. The visual acuity and foveal thickness was measured for each group, and statistical analyses were carried out.

The statistical analysis was performed using SPSS ver. 12.0 (SPSS Inc., Chicago, IL, USA). All of the comparisons were two-tailed. Means of the two groups were compared using a t-test. Appropriate prognostic parameters were analyzed by Pearson's correlation against the final postoperative visual acuity.

The 30 subjects in this series consisted of seven men and 23 women. They ranged in age from 52 to 78 years (mean, 65.33 years). The follow-up period ranged from seven to 17 months (mean, 8.7 months) (Table 1). The preoperative best-corrected visual acuity in logMAR units was 0.43±0.30 (Table 2). Eighteen eyes underwent concomitant cataract surgery. Twelve eyes had only a vitrectomy without concomitant cataract surgery. Five out of 12 eyes had a previous history of cataract surgery before the diagnosis of ERM.

Visual acuity expressed on a logMAR scale at one month, three months, five months, and seven months after surgery were 0.45±0.40, 0.27±0.36, 0.22±0.35, 0.21±0.35, and 0.20±0.34, respectively. Visual acuity on the Han's chart at one month, three months, five months, and seven months improved by two or more lines of vision, in 30%, 50%, 60%, and 70% of the patients and remained within ±1 line in 47%, 50%, 40%, and 30% of the patients, respectively. The proportion of the subjects who had worsened by two or more lines of vision decreased to 23% after one month and to 0.0% three months after surgery (Table 2). Visual acuity changes were greatest one month and three months after surgery. None of the eyes in either group showed changes in visual acuity seven months after surgery, with one exception. This exception was an eye which had cataract surgery after six months due to a secondary cataract. Visual acuity improved greatly from one to three months after the vitrectomy in patients with idiopathic ERM. Seventy percent of the eyes showed visual improvements up to approximately seven months after the vitrectomy, but not thereafter.

Optical coherence tomography showed that foveal thickness decreased significantly after surgery. The mean foveal thickness was 409.7±107.9 µm before surgery and gradually decreased to 288.6±66.1 µm in the seven months after surgery (Table 2).

Visual improvement was greater in the concomitant cataract surgery group compared to the group who underwent a vitrectomy alone. Visual acuities expressed on a logMAR scale preoperatively and one month, three months, five months, and seven months after surgery were 0.38±0.27, 0.30±0.33, 0.12±0.21, 0.09±0.18, and 0.11±0.20, respectively, in the concomitant cataract surgery group. The values in the vitrectomy only group were 0.52±0.34, 0.65±0.39, 0.47±0.44, 0.42±0.45, 0.37±0.46, and 0.36±0.47, respectively (Table 3).

Visual improvement was statistically significant in the concomitant cataract surgery group at three months and five months after surgery (p<0.05). The proportion of subjects who improved by two or more lines of vision after surgery was significantly greater in the concomitant cataract surgery group compared to those in the vitrectomy only group (Table 3). The foveal thickness before and after surgery showed no significant differences between the two groups (p>0.05).

The parameters which were statistically significantly correlated with the final visual acuity included the preoperative visual acuity (0.683), the preoperative foveal thickness (0.544), and the final foveal thickness (0.643) (p<0.005). Therefore, it may be possible to predict the postoperative visual outcome based on preoperative visual acuity and foveal thickness.

Vitrectomy and membrane peeling for the idiopathic epiretinal membrane has been reported to result in visual improvement in 70-90% of eyes.8,9 In this study, visual improvement by two or more lines of vision was 30% after one month and 70% at the final follow-up. These results are similar to previous results that have been reported in the literature.2-7,10 Combined cataract and ERM surgery is another option for selected patients who have a concurrent cataract. Visual improvement by two or more lines of vision in combined cataract and ERM surgery was 39% after one month and 83% at the final follow up. The visual improvement in the vitrectomy only group was 17% after one month and 50% at the final follow-up. Compared to the vitrectomy only group, visual acuity was statistically better at three months and five months after surgery in the concomitant cataract surgery group. However, there was a borderline statistically significant difference at seven months. Removal of lens opacity could explain the three and five month results. Several patients who had undergone a vitrectomy alone had a cataract surgery several months after the vitrectomy. This could be the reason why the borderline difference was seen after seven months. Because there were no differences in the foveal thicknesses between the concomitant cataract surgery group and the vitrectomy only group, concomitant cataract surgery may have little influence on the foveal thickness. Concomitant cataract surg0 ery could enhance the visual improvement during the early postoperative period by removing lens opacity. However, a vitrectomy and delayed cataract surgery could show similar long-term visual outcomes.

Massin et al.6 reported that foveal thickness decreased significantly after surgery in 55 of 62 cases. Foveal thickness did not return to a normal macular profile. The mean foveal thickness was 326±59 µm after surgery in 42 cases. Postoperative visual acuity did not correlate with preoperative and postoperative central macular thickness. However, there have been several reports that visual acuity does correlate with preoperative and postoperative foveal thickness.10,11

In our study foveal thickness decreased significantly from 409.7±107.9 µm before surgery to 287.9±66.8 µm at the final follow-up after surgery. This is still greater than the normative data. The final visual acuity correlated with preoperative visual acuity (0.683), preoperative foveal thickness (0.544), and final foveal thickness (0.643, p<0.005). Visual acuity improved in proportion to foveal thickness after surgery. We were able to predict the final visual outcome based on the preoperative visual acuity and foveal thickness.

Intra-operative complications could include vitreous hemorrhage, retinal petechiae, retinal surface damage, and peripheral iatrogenjuiic retinal breaks. Postoperative complications could include the recurrence of ERM, accelerated nuclear sclerosis of a crystalline lens, retinal detachment, and visual field disturbances. The most common complication, occurring in 12% to 68% of cases, is accelerated nuclear sclerosis of a crystalline lens.2,12,13 Recurrence occurs in 3% to 12% of cases, and retinal detachment occurs in 3% to 14%.2,12-14 No intraoperative complications were noted in our study. Secondary cataracts developed in four of the seven phakic eyes (57%) that underwent only a vitrectomy. Two of these patients had cataract surgery at six months and 12 months after the vitrectomy, respectively. Two patients did not have cataract surgery. One patient refused to have cataract surgery because of her general condition. An optic neuropathy was identified during follow-up in the other patient. Consequently, we did not recommend cataract surgery. One eye out of the 30 subjects (3%) had a partially persistent ERM, and a recurrence of ERM was seen in one eye out of 30 subjects (3%) six months after surgery. A retinal hole was found in one eye, and it was immediately treated with a barrier laser.

Foveal thickness and visual acuity showed improvements in patients with idiopathic ERM until seven months after the vitrectomy. Preoperative visual acuity, foveal thickness, and final foveal thickness showed a significant correlation with the final visual acuity. Because of this, surgeons may be able to predict the final visual outcome based on the preoperative visual acuity and foveal thickness.