Purpose
To evaluate the role of preoperative macular hole (MH) diameter, cataract surgery, and cystoid macular edema (CME) in the reopening of idiopathic macular hole (IMH) after initially successful surgery.
Design
Multicenter, retrospective, comparative case series.
Methods
One hundred and thirty patients (135 eyes) with stage 2, 3, or 4 IMH were included after successful IMH repair at Dijon University Hospital or Nancy University Hospital. The eyes were separated into 4 groups according to the lens status: group 1, vitrectomy in pseudophakic eyes; group 2, vitrectomy and cataract extraction as a combined procedure; group 3, vitrectomy followed by cataract extraction; group 4, vitrectomy on eyes that remained phakic. All patients underwent a complete vitrectomy with retinal inner limiting membrane peeling and intraocular gas tamponade. The main outcome measures were IMH closure rate and best-corrected visual acuity.
Results
Mean duration of symptoms was 8.6 ± 7.6 months. The follow-up was 37 ± 8.6 months (range, 30 to 67 months). The mean preoperative IMH diameter was 454 ± 191 μm. No MH reopened during the follow-up after the initial vitreous surgery. Seven cases of CME were observed (1 in groups 1 and 2; 5 in group 3), and cataract surgery was performed in the 53 patients in group 3 with a mean delay of 11.6 months.
Conclusions
Cataract extraction, CME, or preoperative MH diameter measured by optical coherence tomography were not identified as risk factors for IMH reopening.
Since Kelly and Wendel first reported the successful closure of idiopathic macular holes (IMH) by vitrectomy with gas tamponade, the surgical success rate has improved with the modifications made since these first procedures. The rate of successful closure of macular holes (MH) by pars plana vitrectomy (PPV) has increased to more than 85%.
To improve the closure rate, recent studies have attempted to define tailor-made treatment considering some preoperative characteristics. The IMH diameter measured by optical coherence tomography (OCT) seems to play a major role in defining the indication for internal limiting membrane (ILM) peeling and the need and the duration of postoperative positioning.
However, several postoperative adverse events that could impact long-term visual outcome have also been reported: development or progression of cataract, retinal detachments (RD), elevation of intraocular pressure, and IMH reopening, with rates fluctuating from 0% to 20%.
Many researchers have speculated on the pathogenesis of IMH reopening. The role played by epiretinal membrane (ERM) formation, ILM tractional forces, residual vitreous traction with development of cystoid macular edema (CME), and cataract extraction have been described but remain speculative.
The purpose of this study was to investigate the long-term anatomic and visual results in patients with initial anatomic and OCT-evaluated success after IMH surgery. We focused on the rate of reopening and the potential influence of cataract surgery and preoperative IMH diameter.
Methods
Patients
We conducted a retrospective study in two academic centers on idiopathic full-thickness MHs that had been operated on successfully between January 1, 2002 and December 31, 2005. Consecutive patients with stage 2, 3, and 4 IMHs as defined by Gass and confirmed by OCT (Stratus OCT III; Carl Zeiss Meditec, Dublin, California, USA) were included. Other inclusion criteria were: 1) follow-up lasting at least 30 months and 2) preoperative and postoperative OCT examination. Eyes with high myopia defined as an axial length greater than 27 mm, traumatic MHs, a previous macular surgical procedure and unsuccessful IMH closure with a single surgery, absence of preoperative OCT, and eyes with MHs associated with rhegmatogenous RD at first presentation were excluded.
Preoperative and Postoperative Assessment
Visual acuity (VA) was measured with projected-light Snellen charts converted to logarithm of minimal angle of resolution (logMAR) units for statistical analysis. The preoperative and postoperative routine examinations included best-corrected visual acuity (BCVA), applanation tonometry, biomicroscopy with anterior segment evaluation, careful peripheral retina examination with a wide-angle viewing system with no indentation, and macular OCT assessment before and after the surgery. Before surgery, the largest diameter of the idiopathic IMH was measured with OCT, using 6 radial scans and measurements of the largest diameter recorded for analysis with calipers. The data collected included eye history, age, gender, estimated duration of IMH according to the patient’s history of visual symptoms, lens status, IMH stage, and the follow-up period after surgery.
The primary endpoint of this study was MH reopening after an initial successful surgical procedure. Secondary endpoints aimed to find potential risk factors for IMH reopening among different conditions, such as IMH diameter measured by OCT, stage of IMH, cataract surgery, CME, postoperative capsulotomy, or RD.
To identify the role of cataract surgery in recurrence of IMH, we divided our patients into 4 groups: group 1, vitrectomy in pseudophakic eyes; group 2, vitrectomy and cataract extraction as a combined procedure; group 3, vitrectomy followed by cataract extraction; group 4, vitrectomy in eyes that remained phakic.
Surgical Technique
Vitrectomy was performed similarly by three surgeons (I.H., J.P.B., C.C.G.), using the same surgical procedure : an extensive three-port PPV using 20-gauge instrumentation. A peristaltic pump was used with a maximum vacuum set between 200- and 500-mm Hg, according to the surgeon’s preference, to obtain posterior vitreous detachment. If the posterior hyaloid was attached to the disc (stages 2 and 3) and to the macula (stage 2), it was lifted either with the vitreous cutter probe or with active suction through a blunt cannula. Then the ILM was removed systematically around the IMH using microforceps without indocyanine green or any other dye. Vitrectomy was completed, especially at the vitreous base, with a wide-angle viewing system (Oculus 130; Oculus, Wetzlar-Dutenhofen, Germany). The periphery of the retina was examined carefully with a wide-angle system with indentation. Cryotherapy or argon laser photocoagulation was applied to any tears or suspicious retinal hemorrhages or thinning. Finally, total fluid–air exchange was performed and a nonexpanding mixture of air and SF 6 (20%) was used for pneumatic tamponade in IMHs less than 500 μm, air and C 2 F 6 (17%) in IMHs larger than 500 μm, and air and C 3 F 8 (14%) in IMHs larger than 800 μm. If RD was found, the type of gas was left to the surgeon’s discretion.
In group 2, phacoemulsification combined with vitrectomy was carried out with posterior chamber intraocular implantation. At the end of the surgery, patients received topical antibiotics and corticosteroids. All surgeries were performed with the patient under peribulbar anesthesia. All patients were instructed to remain face down for 15 hours daily for 7 days.
The primary outcome of this study was recurrence of MH depending on cataract extraction. VA was compared among the groups as a secondary outcome. All of the available follow-up data from the time of the initial vitrectomy surgery were included for each eye. Additional variables analyzed in relation to IMH reopening included patient age at vitrectomy, stage and duration of the IMH, and ILM peeling at the time of vitrectomy.
Statistical Analysis
All variables were statistically examined for normality (Kolmogorov-Smirnov test). During the follow-up, VA expressed in logMAR units was compared within the 4 groups of patients using a two-factor, repeated-measure analysis of variance, group × time (preoperative, month 6, postoperative), with repeated measures over time. Separate analyses of variance were used to compare the dependent variables of: 1) the number of RDs, 2) the number of CMEs, and 3) the number of posterior neodymium: yttrium–aluminum–garnet (Nd:YAG) capsulotomies between the 4 groups of patients. Post hoc analyses (Tukey-Kramer) were used to test for differences between pairs when appropriate. The threshold of statistical significance was set at P < .05. The statistical analyses were performed using Statistica software for Windows (StatSoft version 5.1; Statistica, Tulsa, Oklahoma, USA). Data are reported as mean ± standard deviation.
Results
The charts from surgeries performed between January 1, 2002, and December 31, 2005, of 332 consecutive eyes operated on for MH were reviewed. Sixty of these were excluded because the follow-up was less than 30 months. For those with a follow-up longer than 30 months (n = 272), 137 were excluded for no preoperative OCT (n = 79), persistent IMH at 1 month (n = 26), previously vitrectomized eyes (n = 15), traumatic MH (n = 11), and severe myopia (n = 6). Finally, 135 eyes of 130 patients with a mean follow-up of 37 ± 8.6 months (range, 30 to 67 months) were included in this retrospective series; the patient characteristics are reported in Table 1 .
| Group 1 | Group 2 | Group 3 | Group 4 | Overall | |
|---|---|---|---|---|---|
| No. of eyes (% overall) | 22 (16.3) | 49 (36.3) | 53 (39.3) | 11 (8.1) | 135 (100) |
| Mean age (SD), years | 74.5 (6.5) | 70.4 (7) | 64.6 (7.5) | 67.2 (13.3) | 68.1 (8.7) |
| Mean follow-up (range), mos | 35.9 (30 to 48) | 37.4 (30 to 57) | 41.6 (30 to 67) | 34.3 (30 to 52) | 37.3 (30 to 67) |
| Gender, no. (%) | |||||
| Female | 17 (77.3) | 31 (70.4) | 34 (64.2) | 6 (54.5) | 88 (67.7) |
| Male | 5 (22.7) | 13 (29.6) | 19 (35.8) | 5 (45.5) | 42 (32.3) |
| Duration of idiopathic MH, no. (%) | |||||
| <6 mos | 6 (27.3) | 7 (14.3) | 17 (32.1) | 4 (36.4) | 34 (25.2) |
| ≥6 mos | 9 (40.9) | 27 (55.1) | 31 (58.5) | 5 (45.4) | 72 (53.3) |
| Unclear | 7 (31.8) | 15 (30.6) | 5 (9.4) | 2 (18.2) | 29 (21.5) |
| Idiopathic MH clinical stage, no. (%) | |||||
| 2 | 2 (9.1) | 8 (16.3) | 7 (13.2) | 2 (18.2) | 19 (14.1) |
| 3 | 16 (72.7) | 37 (75.5) | 40 (75.5) | 8 (72.7) | 101 (74.8) |
| 4 | 4 (18.2) | 4 (8.2) | 6 (11.3) | 1 (9.1) | 15 (11.1) |
| Idiopathic MH size (μm), no. (%) | |||||
| <400 | 12 (54.5) | 21 (42.9) | 24 (45.3) | 5 (45.5) | 62 (45.9) |
| ≥400 | 10 (45.5) | 28 (57.1) | 29 (54.7) | 6 (54.5) | 73 (54.1) |
| Mean time to cataract extraction (range), mos | — | — | 11.6 (1 to 43) | — | — |
| Gas, no. (%) | |||||
| SF 6 | 17 (77.3) | 40 (81.6) | 42 (79.2) | 8 (72.7) | 107 (79.2) |
| C 2 F 6 | 5 (22.7) | 9 (18.4) | 8 (15.1) | 2 (18.2) | 24 (17.8) |
| C 3 F 8 | 0 | 0 | 3 (5.7) | 1 (9.1) | 4 (3) |
Five patients had bilateral IMH. Eighty-eight patients were women (67.7%). The mean age of the patients was 68.1 ± 8.7 years (range, 47 to 85 years). Before surgery, IMHs had been staged as follows: stage 2, 19 eyes (14.1%); stage 3, 101 eyes (74.8%); and stage 4, 15 eyes (11.1%). Overall, the preoperative mean BCVA was 0.90 logMAR units (range, 0.15 to 1.7 logMAR units; Snellen equivalent, 20/160; Figure ). The postoperative mean BCVA was 0.33 logMAR units (range, 0 to 1.3 logMAR units; Snellen equivalent, 20/40). VA improved significantly ( P < .001) and to a similar extent for all groups ( P > .05). Nevertheless, the kinetics of improvement differed between groups ( P < .01). VA ending values were statistically different from month 6 values for groups 2 and 3 (data not shown), whereas no differences appeared for groups 1 and 4. This is mainly because of the unchanged status of the lens at 6 months and at the end of the observation period in group 1 (pseudophakic eyes) and in group 4 (lens still clear after vitrectomy). Conversely, the lens was extracted during the IMH surgery in group 2 and after a while in group 3, thus leading to a rapid VA change. Of the 135 eyes included, 128 (94.8%) had improved VA after successful IMH closure. In 3 eyes (2.2%), the VA did not change, and in 4 eyes (3%) the VA worsened despite successful IMH closure ( Table 2 ).
| Reason (No.) | Group 1 (n = 22) | Group 2 (n = 49) | Group 3 (n = 53) | Group 4 (n = 11) | Overall (n = 135) |
|---|---|---|---|---|---|
| Macular atrophy | 0 | 0 | 1 | 1 | 2 |
| Persistent CME | 0 | 0 | 0 | 2 | 2 |
| Progression of cataract | 1 | 0 | 0 | 0 | 1 |
| Glaucoma | 0 | 0 | 0 | 1 | 1 |
| RD treated with silicone oil | 0 | 0 | 1 | 0 | 1 |
| Reopening of idiopathic MH | 0 | 0 | 0 | 0 | 0 |
The mean duration of IMH was 8.6 ± 7.6 months (range, 1 to 24 months). Seventy-two eyes (53.3%) had visual symptoms for 6 months or longer, 34 eyes (25.2%) were symptomatic for less than 6 months, and in 29 eyes (21.5%), the duration was unclear because of the lack of data on the patient’s history. The mean size of IMH was 454 ± 192 μm (range, 166 to 1000 μm). There were 62 IMHs smaller than 400 μm and 73 IMHs larger than 400 μm. We observed 17 preoperative retinal tears and 8 preoperative RDs, which were treated during surgery, and 10 RDs were observed during follow-up (7.4%), which were diagnosed at 10.1 months on average. There were no cases of endophthalmitis. During the follow-up, no reopening of a previously closed IMH occurred.
Overall, 113 eyes were phakic at baseline. Twenty-two patients (22 eyes) were pseudophakic at initial presentation (group 1). Forty-four (49 eyes) patients were phakic and underwent a combined procedure (group 2). Fifty-three patients (53 eyes) were operated on for cataract during the follow-up (group 3), whereas only 11 patients (11 eyes) were still phakic at the last follow-up visit (group 4). In group 3, the cataract extraction was performed after a mean delay of 11.6 ± 7.9 months (range, 1 to 43 months). In this group, the mean follow-up after cataract extraction was 29.8 ± 10.8 months (range, 9 to 63 months). Despite these discrepancies, when an analysis of covariance between the groups was performed for the variables VA, IMH stage, IMH duration, MH size, preoperative retinal tears, reopening of IMH, RDs, CME, and improvement of VA, covarying for the age, age did not influence the results between groups.
Cystoid macular edema was observed in 7 cases (5.2%): 1 in group 1, 1 in group 2, 5 in group 3, and none in group 4 ( Table 3 ). Three of them were related to surgical complications. Chronic elevation of intraocular pressure was found in 8 eyes (5.9%): 2 eyes in group 1, 1 in group 2, 5 in group 3, and none in group 4. Twenty-five secondary cataracts were treated using Nd:YAG laser capsulotomy a mean 15 ± 8.7 months after cataract extraction.
