Abstract
Purpose:
The purpose of this article is to examine the efficacy of macular hole repair in eyes with coexisting dry age-related macular degeneration (AMD).
Methods:
A retrospective analysis was performed of charts of 25 patients (27 eyes) diagnosed with mild to moderate dry AMD who underwent macular hole repair via 25-gauge pars plana vitrectomy between 2014 and 2016. Data of interest included anatomic failure rates, complication rates, and best-corrected visual acuity (BCVA) preoperatively, and at 1 month, 3 months, 6 months, and 12 months postoperatively. When available, data at each patient’s most recent visit were also analyzed.
Results:
Macular hole repair resulted in a statistically significant (P < .05) visual improvement postoperatively, with BCVA increasing from 20/141 preoperatively to 20/33 1 year postoperatively. Mean BCVA at most recent visit was 20/41. Mean duration of follow-up was 13 months (range, 1-39 months). One of 27 (3.7%) macular holes failed to close after vitrectomy. One of 27 eyes (3.7%) progressed from dry to wet AMD. Four of 27 additional eyes (18.5%) were noted to have worsening of their AMD on exam over the course of follow-up.
Conclusions:
Macular hole repair in patients with coexisting dry AMD leads to a significant improvement in visual performance and has a low risk of failure or complication.
Introduction
Macular holes (MHs) are a cause of significant central vision loss. However, the emergence of surgical techniques allowing for the repair of MHs has led to high rates of anatomic and visual success. 1 Works spanning several years of follow-up have shown that successful hole closure (herein defined as “anatomic success”) can lead to significant improvements in best-corrected visual acuity (BCVA), with 1 such study of patients undergoing MH repair demonstrating a mean increase in visual acuity from 20/129 preoperatively to 20/56 at last visit. 2
Few studies have attempted to elucidate whether coexisting disorders with the potential to limit central vision, such as dry age-related macular degeneration (dAMD), have a significant negative effect on MH repair outcomes. Because dAMD and MHs both affect the macula, many retinal specialists may be reluctant to perform MH repair in a patient presenting with both of these disorders. These concerns, while understandable, may be unfounded. An extensive search of the literature reveals that pars plana vitrectomy (PPV) does not necessarily increase the rate of progression of AMD and may in fact have a protective effect. 3 There has been at least 1 reported case of soft drusen regression following MH surgery. 4 Furthermore, Berinstein et al 5 have shown that successful full-thickness MH closure in patients with coexisting significant (intermediate or large-size) drusen can lead to significant improvements in BCVA, though it should be noted that a significant proportion of their patient cohort required reoperation because of initial failure of hole closure. While prior works such as these are encouraging, other studies have shown a link between choroidal neovascularization and MH surgery. 6,7 Herein, we present our own retrospective evaluation of outcomes in patients with preexisting dAMD who underwent MH repair.
Methods
Approval for this study was obtained from the institutional review board of the University of Alabama at Birmingham. Our study adhered strictly to the tenets of the Declaration of Helsinki. Informed consent was not required for this study given its retrospective nature and our strict use of deidentified data that do not place the privacy of our patients at stake. We retrospectively evaluated the records of 25 patients (27 eyes) with a history of category 2 or 3 dAMD as defined by the Age-Related Eye Disease Study (AREDS) and a diagnosis of idiopathic full-thickness MH, repair of which took place between 2014-2016 at the Callahan Eye Foundation Hospital of the University of Alabama at Birmingham.
All procedures were performed by 1 of 4 experienced retinal surgeons. Patients underwent a 3-port 25-gauge trocar placement. A complete vitrectomy was performed with elevation of the hyaloid in all cases. Internal limiting membrane (ILM) forceps were used to peel the ILM around the hole in all cases. Indocyanine green (ICG) dye was placed on the macula and removed prior to ILM peeling. ICG dye was used to lightly stain the ILM and ensure its complete removal during MH repair. Twenty-one eyes had SF6 gas placed, while the remaining 6 eyes underwent C3F8 gas tamponade. Decisions regarding the type of gas used during MH repair were made at the sole discretion of the operating surgeon. The choice of gas used had no effect on MH closure or visual outcomes. All patients were requested to maintain a face-down position for a total of 72 hours while awake and were not permitted to sleep on their back for 10 days.
Charts were analyzed for multiple variables, including age, sex, gender, ethnicity, left vs right eye, lens status, and preoperative and postoperative BCVA. Additionally, instances of MH repair failure, progression of AMD (herein defined as the appearance of new drusen or choroidal neovascularization or an increase in the size of preexisting drusen), and any complications attributable to MH repair were also noted. Clinical examinations as well as preoperative and postoperative optical coherence tomography (OCT) imaging were used to evaluate the progression or stability of AMD within our patient cohort. Neither preoperative nor postoperative fundus photos or fluorescein angiography were acquired from patients.
Primary outcome measures included measures of BCVA at 1 month, 3 months, 6 months, 12 months, and most recent visit postoperatively, in addition to measures of BCVA preoperatively. Secondary outcomes included rates of hole repair failure, progression of AMD, and surgical complications. Statistical analysis was carried out using Microsoft Excel 365. Measures of visual acuity were converted to logarithm of the minimum angle of resolution (logMAR) values for the purpose of statistical analysis. Statistical significance (P < .05) was assessed using the Student t test.
Results
The baseline characteristics of our patient cohort are listed in Table 1. Of the 27 eyes evaluated in this study, 6 were male and 22 were pseudophakic. The mean duration and size of the MHs seen in our patient cohort were 3.8 months (range, 1-10 months) and 400 µm (range, 250-600 µm), respectively. The mean age of patients on the day of MH repair was 74.4 ± 8.3 years (range, 62-89 years). Mean duration of follow-up was 13 ± 10.8 months (range, 1-39 months). Of the 5 eyes that were phakic, 4 (80%) had cataract surgery during follow-up, and cataract extraction was recommended to the other. Mean time to cataract surgery or recommendation of cataract surgery was 10.5 ± 7.1 months (range, 3-18 months) after MH repair had taken place.
Baseline Characteristics of Eyes Prior to Vitrectomy for Macular Hole Repair.
a Age expressed as mean ± SD (range).
One of 27 eyes (3.7%) was noted to have progressed to wet AMD (ie, displayed clear evidence of new neovascularization on eye exam) over the course of follow-up, with another 4 eyes (14.8%) noted to have worsening of their dAMD (increase in quantity/size of drusen on eye exam or OCT imaging) documented in their records. One of 27 MHs (3.7%) failed to close after initial repair—documentation of additional surgery was not noted in this particular patient’s chart. One patient was noted to have developed a complex macula on retinal detachment with multiple tears and vitreous debris approximately 1 month postoperatively, requiring reoperation.
Mean BCVA for the entire patient cohort progressively improved over the course of follow-up (Figure 1, Table 2), from 0.847 logMAR (20/141 Snellen equivalent) preoperatively to 0.551 (20/71), 0.456 (20/57), 0.506 (20/64), 0.225 (20/33), and 0.315 (20/41) at 1 month, 3 months, 6 months, 12 months, and most recent visit postoperatively. In comparison to preoperative BCVA, all changes in postoperative BCVA over the course of follow-up were statistically significant (P < .05) for the entire patient cohort. Furthermore, of the 9 eyes with data available beyond 1 year of follow-up, 7 (77.8%) exhibited an improvement in vision (defined as >2 lines of improvement in Snellen acuity), while the remaining 2 eyes remained stable (defined as remaining within 2 Snellen lines of preoperative BCVA).
Visual representation of trends in best-corrected visual acuity (BCVA) after macular hole repair. Data from the entire patient cohort (n = 27 eyes) is displayed. BCVA improved significantly throughout the course of follow-up. logMAR indicates logarithm of the minimum angle of resolution; mo, month; pre-op, preoperative; y, year.
Trends in BCVA Over Follow-up, Full Patient Cohort.
Abbreviations: BCVA, best-corrected visual acuity; logMAR, logarithm of the minimum angle of resolution.
Values are expressed as mean logMAR ± SD (mean Snellen equivalent) unless otherwise specified.
Visual outcomes were also analyzed based on gender, lens status, and preoperative BCVA (either less than 20/50 or greater than or equal to 20/50). Findings are outlined in Table 3. Phakic patients and male patients did not develop a significant improvement in BCVA within the first 6 months after surgery, but did do so thereafter. Notably, patients with a good preoperative visual acuity (defined herein as greater than or equal to 20/50) did not develop a statistically significant improvement in BCVA throughout the entire follow-up period.
Subgroup Analysis of Eyes Based on Gender, Lens Status, and Preoperative Best-Corrected Visual Acuity.
a Statistically significant improvement relative to preoperative best-corrected visual acuity within each group (P < .05).
Discussion
To the best of the authors’ knowledge, our study is the first since that of Berinstein and colleagues 5 in 2000 investigating outcomes of MH repair in patients with coexisting dAMD. This is likely a result of the relative rarity with which such patients present. Only 32 of 944 (3.4%) patients in Berinstein’s study diagnosed with idiopathic MH were identified as having intermediate or large-size drusen on eye exam over a 4-year period from 1994 to 1998. 5 In our own study, we were able to identify only 27 eyes over a 3-year period presenting both with dAMD and MH. Given that our practice on average performs 90-100 MH repairs per year, this equates to a rate of 9.3% at best.
In comparison to the study by Berinstein et al 5 of patients with intermediate or large-sized drusen undergoing MH repair, our results demonstrate it is possible for patients with dAMD undergoing MH repair to achieve a significant improvement in BCVA status post-PPV while simultaneously achieving a high rate of anatomic success long term. Berinstein and colleagues had an initial repair failure rate of 24%, with 75% of these patients requiring reoperation, while our failure rate was only 3.7%. Furthermore, complications attributable to surgery or the progression of AMD were few and far between, with just 1 patient developing a retinal detachment and only 1 patient progressing to wet AMD from dAMD. Berinstein et al postulated that their relatively low rates of anatomic success could have been secondary to the interference of drusen with the resorption of subretinal fluid surrounding MHs by the retinal pigment epithelium, and also posited that drusen formation may impede macular reattachment after PPV. 5
Our own results are consistent with more recent studies that have been able to achieve anatomic success rates of 89% or more, indicating that the presence of dAMD may not negatively affect MH repair outcomes status post PPV. 8,9 As discussed previously, these findings may be due to a protective effect that vitrectomy has on the progression of AMD. 3,4 Roller and colleagues 3 suggested that PPV can exert this effect via elimination of adhesion and traction at the vitreoretinal surface and clearance of inflammatory mediators from the vitreoretinal interface. However, we must also note that our higher anatomic success rate may be attributable to improvements in surgical techniques and postoperative care, as several years have elapsed since the publication of the Berinstein study. Additionally, a more in-depth analysis of our data reveals other factors potentially affecting patients with MHs and comorbid dry AMD that may need to be investigated in future works.
Several studies have shown that MHs appear to be more common in women than men, particularly with advancing age. 10 -12 The same is true of our study, with women accounting for 20/27 eyes (74.1%) examined herein, roughly a 3:1 female:male ratio. Though the exact mechanism by which this phenomenon manifests itself is not known, it is thought that postmenopausal women are more prone to developing posterior vitreous detachments (and therefore, MHs) as a result of the effects of decreasing levels of estrogen in the vitreous. 13 The data illustrate that men do not appear to benefit significantly from MH repair via PPV in the short term, but do appear to do so in the long term. Conversely, throughout follow-up, women benefitted significantly from MH repair early on in their postoperative course. However, it should be noted that women tended to fare worse than men in the long term, with the caveat that the sample size of the male cohort is quite small (n = 5 eyes) and limited by loss to follow-up, making it difficult to draw a concrete conclusion regarding the effect of gender on outcomes of MH repair in patients with dAMD in the long-term.
Similar relationships were observed between patients who were phakic and pseudophakic at the time of surgery (Table 3). However, it should be noted that both in the male and phakic subsets of patients, the mean preoperative BCVA was significantly higher than in their female and pseudophakic counterparts, making it more unlikely to achieve a statistically significant improvement in BCVA in the first few months following surgery in these 2 subgroups. In addition, given that vitrectomy accelerates cataract progression 14 and that 4 of 5 initially phakic patients in our study required cataract surgery or had surgery recommended to them by the end of follow-up, it makes sense that visual outcomes would be limited in the short term (ie, prior to cataract extraction) relative to pseudophakic patients. Phakic patients were also noted to have outperformed pseudophakics in the long term. While it is possible that lens status may play a role in long-term visual performance, it is once again difficult to draw a concrete conclusion regarding the effects of phakia and pseudophakia on patient outcomes given the small sample size of this cohort.
Patients with a good preoperative BCVA prior to surgery (ie, ≥20/50) did not achieve a significant improvement in postoperative BCVA over the course of follow-up. As mentioned previously, it is far more difficult to achieve a significant increase in BCVA in such patients given that they had less room for improvement than their counterparts with a far worse baseline BCVA. However, the lack of objective visual benefit as measured by BCVA does not necessarily preclude such patients from undergoing surgery for MH repair. One should also keep in mind that symptoms of MHs such as metamorphopsia or a perceived decrease in one’s quality of life due to deficiencies in vision are not necessarily assessed adequately by evaluating a patient’s BCVA alone. 15 While BCVA is a very useful and objective parameter that can be used to assess patients’ outcomes and need for surgery, it is not, and should not, be the sole criteria on which a physician decides whether a patient would benefit from vitrectomy for MH repair.
Our study is constrained by the usual limitations inherent in retrospective reviews and our lack of a control group. While our data do highlight interesting trends among several subgroups, we were limited in our ability to draw strong conclusions from these findings because of the relatively small sample size of our patient cohort. Furthermore, as mentioned above, use of BCVA alone to assess patient outcomes can have its own shortcomings. Ideally, future studies should make use of other testing modalities to assess symptoms associated with MHs such as metamorphopsia and central visual field defects.
Patients with idiopathic full-thickness MHs and coexisting category 2 or 3 dAMD can benefit significantly from MH repair via PPV, irrespective of lens status or gender. Visual outcomes would likely not be as favorable in eyes with advanced AMD (ie, category 4) in the form of central geographic atrophy or prior choroidal neovascularization. Future studies should examine in more detail the effects of these 2 factors as well as preoperative BCVA on outcomes in this patient population after surgical repair.
Footnotes
Ethical Approval
The institutional review board of the University of Alabama at Birmingham approved this study. This study adhered strictly to the tenets of the Declaration of Helsinki.
Statement of Informed Consent
Informed consent was not required for this study given its retrospective nature and our strict use of deidentified data that do not place the privacy of our patients at stake.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.