Microincision vitreous surgery (MIVS) using a 27-, 25-, or 23-gauge vitreous surgery system is regarded to be less invasive than surgeries using a conventional 20-gauge vitreous surgery system.1–3 MIVS has been shown to be effective for the early recovery of visual function. Furthermore, the recent increased use of MIVS has led to expanded indications for vitreous surgery in patients with vitreoretinal disease.
Epiretinal membrane (ERM), which is also known as macular pucker or cellophane maculopathy, is a common disorder that affects the vitreoretinal interface. ERM surgery, including pars plana vitrectomy (PPV) and peeling, has been commonly performed for more than 30 years.4 Due to the diagnostic advances achieved when using optical coherence tomography (OCT), vitreous surgery for ERM is now being performed in many patients who already have a good visual acuity (VA).
The majority of studies investigating the use of PPV for the removal of ERM have reported low rates of complications and high rates of visual recovery.1–3 However, most of the results reported are from studies of MIVS for ERMs that only have relatively short follow-up periods of usually no more than 12 months.5–10 Thus, the long-term results for ERM removal when using MIVS have yet to be conclusively established.11
The current study investigated the long-term course of the VA and central retinal thickness (CRT) in patients who were followed for at least 3 years after initial vitreous surgery for ERM.
Patients and Methods
This retrospective, consecutive, comparative case series study examined 43 eyes of 43 patients who underwent 23- or 25-gauge MIVS for ERM at the Jikei University School of Medicine Hospital between July 2010 and June 2013.
In 37 cases (86%), cataract surgery was simultaneously performed in conjunction with the surgery for ERM. The mean age of the patients was 66.0 years ± 9.0 years, with the surgical group consisting of 24 females and 19 males. The mean axial length was 24.12 mm ± 1.72 mm for the affected eye. The mean preoperative logarithm of the minimum angle of resolution (logMAR) VA was 0.30 ± 0.24. The preoperative logMAR VA was under 0 in 5% of the cases. The mean preoperative CRT was 429.0 μm ± 86.2 μm.
In 27 cases (62.8%), the preoperative disruption of the outer retinal layer was visualized on the OCT image as a loss of integrity of the external limiting membrane (ELM) band, the ellipsoid zone (EZ), and the interdigitation zone (IZ).
VA was measured using a Landolt chart at a distance of 5 m. For the data analyses, visual acuities in decimal were converted to logMAR for all calculations.
We examined patients every 6 months starting from 1 year after surgery.
All study protocols were approved by the ethics committee of the Jikei University School of Medicine.
All patients underwent surgery under retrobulbar anesthesia. Phacoemulsification and intraocular lens (IOL) implantation were performed in all of the phakic cases, with the vitreoretinal procedures performed following the implantation of the IOL. Cataract surgery was performed via a 2.4-mm sclerocorneal incision, with the IOL inserted through the same incision. Vitreous surgery was conducted using the Accurus Surgical System 800CS (Alcon Surgical, Fort Worth, TX) with a 23- or 25-gauge system. A Resight fundus imaging system (Carl Zeiss Meditec AG, Jena, Germany) was used. Endoillumination was delivered using a xenon-based illumination system (Xenon BrightStar; DORC, Zuidland, The Netherlands). After the creation of three ports, vitrectomy was performed. If there was no posterior vitreous detachment present in the patients, this was induced artificially. While monitoring the macula with a magnifying floating contact lens, the ERM was grasped using intraocular forceps from either above or below the fovea and then detached from the temporal side. Although the internal limiting membrane (ILM) was not routinely peeled during the procedure, this was performed in 18 patients (41.8%). In addition, indocyanine green (ICG)-assisted ILM peeling was performed in 12 cases (27.9%). ICG was used when ILM peeling was difficult without ICG. While simultaneously applying pressure to the sclera, the parts of the vitreous body around the ports were then removed. After withdrawal of the three ports, the scleral incisions were sutured if there was any apparent leakage.
There were no significant complications such as iatrogenic macular hole, postoperative vitreous hemorrhage, retinal detachment, or endophthalmitis that affected the visual outcome during and after the operations. Statistical analysis was performed using software programmed by Hisae Yanai (Statcel-3; OMS Publication, Saitama, Japan) in conjunction with Excel (Microsoft, Redmond, WA). Values of P less than .05 were considered statistically significant. Comparisons of the data were performed using the analysis of variance (ANOVA)+Bonferroni/Dunn method, Mann-Whitney U test, and the t-test, as appropriate. Data correlations were analyzed using the Pearson's correlation and Spearman rank correlation tests.
There was significant improvement observed in the VA at 3 months after surgery compared with baseline, with the improved results maintained for 5 years (ANOVA+Bonferroni/Dunn method; P < .05). The average visual gain was 0.23 ± 0.32. In 23 cases (53%), the logMAR VA was under 0 at 3 years after surgery. The best VA was determined in each of the cases between 1 month and 5 years (84% of the cases were over 1 year), with an average time required to achieve the best VA of 1.7 ± 1.4 months. The mean CRT significantly decreased from 1 month up to 5 years as compared with the CRT before surgery (ANOVA+Bonferroni/Dunn method; P < .05). The lowest mean CRT was observed at 4.5 years, whereas the highest mean VA was observed at 2 years after the MIVS for ERM (Figure 1).
Long-term course of logarithm of the minimum angle of resolution (logMAR) visual acuity (VA) and central retinal thickness (CRT) following vitreous surgery for epiretinal membrane. There was significant improvement observed in the VA at 3 months after surgery as compared with baseline, with the improved results maintained for 5 years (analysis of variance [ANOVA]+Bonferroni/Dunn method; P < .05). The mean CRT significantly decreased from 1 month up to 5 years as compared with the CRT before surgery (ANOVA+Bonferroni/Dunn method; P < .05).
During this 3-year period, there were no significant differences found for the logMAR VA and CRT between the groups using and not using ICG (3-year logMAR VA, 0.11 ± 0.25; 3-year CRT, 339.29 μm ± 38.11 μm, t-test; P = .77 and P = .79, respectively).
Pearson's correlation results indicated that the 3-year logMAR VA was significantly correlated with the pre-logMAR VA (r = 0.34, P = .03) (Figure 2). The Spearman rank correlation showed that the 3-year CRT was significantly correlated with the pre-CRT (r = 0.42, P = .02) (Figure 3). Pearson's correlation also showed that the 3-year CRT was not significantly correlated with the 3-year logMAR VA (r = −0.05, P = .73) (Figure 4).
Correlation of pre-logarithm of the minimum angle of resolution (logMAR) visual acuity (VA) and 3-year logMAR VA. Pearson's correlation results indicated that the 3-year logMAR VA was significantly correlated with the pre-logMAR VA (r = 0.34, P = .03).
Correlation of pre-central retinal thickness (CRT) and 3-year CRT. The Spearman rank correlation showed that the 3-year CRT was significantly correlated with the pre-CRT (r = 0.42, P = .02).
Correlation of pre 3-year logarithm of the minimum angle of resolution (logMAR) visual acuity (VA) and 3-year central retinal thickness (CRT). Pearson's correlation results indicated that the 3-year logMAR VA was not significantly correlated with the 3-year CRT (r = −0.05, P = .73).
In 27 cases (62.8%), OCT showed there was preoperative disruption of the outer retinal layer. The average logMAR VA and CRT in cases found to have outer retinal layer disorder before surgery were 0.46 μm ± 0.24 μm and 462.30 μm ± 104.94 μm, respectively. The average logMAR VA and CRT in cases found to have no outer retinal layer disorder before surgery were 0.20 μm ± 0.23 μm and 414.88 μm ± 67.76 μm, respectively. There was a significantly worse mean logMAR VA in cases found to have outer retinal layer disorder (t-test; P = .003). However, the mean CRT was not significantly worse (t-test; P = .13) in cases found to have outer retinal layer disorder.
At 3 years after the surgery, the number of disruptions of the outer retinal layer had decreased by nine cases (20.9%) (Table 1). Neither the presence nor absence of retinal outer layer disorder had any influence on the CRT at 3 years after the surgery (t-test; P = .37) (Table 2). However, there was a significantly worse mean logMAR VA in cases found to have outer retinal layer disorder before surgery (Mann–Whitney U test; P = .013) (Table 2).
Disruption of Outer Retinal Layer
Influence of Preoperative Disruption of Outer Retinal Layer at 3 Years After Surgery
In the patients who exhibited a preoperative disruption of the outer retinal layer, 41% of these cases were found to have a logMAR VA under 0 at 3 years after surgery. In patients who did not exhibit any preoperative disruption of the outer retinal layer, 75% of these cases had a logMAR VA under 0 at 3 years after surgery.
A total of 20 eyes (46.5%) underwent Nd:YAG capsulotomy, with a mean time for undergoing the procedure of 32.5 months ± 13.0 months (range: 8 months to 55 months). In four eyes (9.3%), the 3-year logMAR VA decreased over 0.2 units as compared with the 1-year VA. In three eyes (7.0%), the 3-year CRT increased over 50 μm as compared with the 1-year CRT. In contrast, in seven eyes (16.2%), the 3-year CRT decreased over 50 μm compared with the 1-year CRT.
There were no complications, such as thinning of the retina (CRT < 200 μm), observed during the 3 years of follow-up after the operation.
A Representative Case
A 58-year-old female patient with no ophthalmological or medical history indicative of eye disease presented with blurred vision and metamorphopsia in her left eye. At presentation, her best-corrected visual acuity (BCVA) was 0.3 logMAR in the left eye. Intraocular pressure was 14 mm Hg, and anterior segment slit-lamp examination was unremarkable in the left eye. Dilated funduscopy revealed ERM in the left eye, confirmed by OCT (CRT = 535 μm) (Figure 5A). The preoperative disruption of the outer retinal layer was visualized on the OCT image as a loss of the IZ.
A representative case of an epiretinal membrane. (A) Preoperative optical coherence tomography (OCT): The preoperative disruption of the outer retinal layer was visualized on the OCT image as a loss of the interdigitation zone (IZ) (white arrow) (logarithm of the minimum angle of resolution [logMAR] best-corrected visual acuity [BCVA]: 0.3, CRT: 535 μm). (B) Postoperative OCT at 1 month: (logMAR BCVA: 0.045, CRT: 390 μm). (C) Postoperative OCT at 3 years: The disruption of the outer retinal layer was not visualized on the OCT image. (logMAR BCVA: 0, CRT: 363 μm).
Cataract and vitreous surgery was conducted using a 25-gauge system. The ERM was removed easily with intraocular forceps. Removal of the ILM was not attempted. At the end of the surgery, intraocular tamponade was not used in the absence of any retinal breaks. One month postoperatively, the ILM was successfully removed. The BCVA was 0.045 log MAR, and the CRT was 390 μm (Figure 5B). At the 3-year follow-up, the logMAR BCVA was 0, and CRT was 363 μm. The disruption of the outer retinal layer was not visualized on the OCT image (Figure 5C).
Although there have been many reports on the postoperative course of ERM, the majority of these have only examined the results of the ERM surgery using MIVS after relatively short follow-up periods of usually no longer than 12 months.5–10 Kim et al. reviewed the records for the VA and CRT in 52 patients with idiopathic ERM who were treated with vitrectomy and who were followed up for longer than 12 months (average: 27 months).12 The findings of this study showed that VA and CRT improvements took more than a year, there-by indicating that long-term follow-up is necessary in these types of patients. In our current study, we found the highest mean postoperative VA occurred at 2 years after surgery, whereas the lowest mean CRT occurred at 4.5 years after surgery. In addition, we found that changes in the morphology and improvement of the visual function continued for more than 3 years after the initial vitreous surgery for ERM. Our results indicate that additional investigations into the changes of the retinal morphology and visual function over a longer period will need to be undertaken in the future.
Previous studies have also reported that a good preoperative VA and the absence of any retinal outer layer disorder are factors that are associated with the visual prognosis after ERM surgery.13,14 During the more than 3-year follow-up in our current study, although we found that the presence or absence of the outer layer of the retina appeared to have no effect on the improvement of the CRT, our results did show that it affected the VA prognosis. Furthermore, we also found that there was a positive correlation between the VA outcome at 3 years after the surgery and the condition of the outer layer of the retina prior to the operation (Figures 2 and 3). There have been many reports of cases in which there was no improvement in the VA when the outer layer disorder of the retina was present prior to the initial surgery, even in patients who underwent long-term follow-ups.13,14 Thus, overall these results suggest that even in patients who undergo a long-term follow-up in order to ensure there is a good visual function prognosis, it is important that surgeries in these patients be performed prior to the appearance of any retinal outer layer disorders.
All vitrectomies in our current study were done using MIVS. During the postoperative follow-up periods between the first and third years, the VA worsened in a total of six eyes (14%). However, there were no complications, such as thinning of the retina, observed during the 3 years of follow-up after the operation. Whereas one of the cases was considered to be caused by an ERM recurrence, all of the other cases with worsening of the VA were thought to be due to the progression of the after-cataract. For long-term VA management and postoperative maintenance of the ERM, follow-up and management of the anterior ocular segment are important and need to be carefully considered. Thus, the selection of an IOL with a low occurrence rate of after-cataracts and the performance of posterior capsulotomy at the time of the vitreous surgery appear to be useful in helping to prevent complications after the initial procedure.15
The limitations of our present study include the fact that it was a retrospective review and there were variations of the surgical technique utilized during the procedures.
Because VA and CRT changes were shown to occur over a long period after the initial surgery for ERM, this demonstrates that longer observation periods are necessary in these types of patients. Disorders of the outer layer of the retina that are present before surgery can influence the VA outcomes, with the changes occurring at times long after the initial surgery.
- Sayed KM, Naito T, Farouk MM, et al. Twenty five-gauge sutureless vitrectomy versus 20-gauge vitrectomy in epiretinal membrane surgery. J Med Invest. 2012;59(1–2):69–78. doi:10.2152/jmi.59.69 [CrossRef]
- Haas A, Seidel G, Steinbrugger I, et al. Twenty-three-gauge and 20-gauge vitrectomy in epiretinal membrane surgery. Retina. 2010;30(1):112–116. doi:10.1097/IAE.0b013e3181b32ebf [CrossRef]
- Sandali O, El Sanharawi M, Lecuen N, et al. 25-, 23-, and 20-gauge vitrectomy in epiretinal membrane surgery: A comparative study of 553 cases. Graefes Arch Clin Exp Ophthalmol. 2011;249(12):1811–1819. doi:10.1007/s00417-011-1752-3 [CrossRef]
- Machemer R. The surgical removal of epiretinal macular membranes (macular puckers) (author's transl). Klin Monbl Augenheilkd. 1978;173(1):36–42.
- Grewing R, Mester U. Results of surgery for epiretinal membranes and their recurrences. Br J Ophthalmol. 1996;80(4):323–326. doi:10.1136/bjo.80.4.323 [CrossRef]
- Garweg JG, Bergstein D, Windisch B, Koerner F, Halberstadt M. Recovery of visual field and acuity after removal of epiretinal and inner limiting membranes. Br J Ophthalmol. 2008;92(2):220–224. doi:10.1136/bjo.2007.131862 [CrossRef]
- Hikichi T, Matsumoto N, Ohtsuka H, et al. Comparison of one-year outcomes between 23- and 20-gauge vitrectomy for preretinal membrane. Am J Ophthalmol. 2009;147:639–643. doi:10.1016/j.ajo.2008.10.009 [CrossRef]
- Konstantinidis L, Berguiga M, Beknazar E, Wolfensberger TJ. Anatomic and functional outcome after 23-gauge vitrectomy, peeling, and intravitreal triamcinolone for idiopathic macular epiretinal membrane. Retina. 2009;29(8):1119–1127. doi:10.1097/IAE.0b013e3181ac23da [CrossRef]
- Wong JG, Sachdev N, Beaumont PE, Chang AA. Visual outcomes following vitrectomy and peeling of epiretinal membrane. Clin Exp Ophthalmol. 2005;33(4):373–378. doi:10.1111/j.1442-9071.2005.01025.x [CrossRef]
- Kwok AKh, Lai TY, Yuen KS. Epiretinal membrane surgery with or without internal limiting membrane peeling. Clin Exp Ophthalmol. 2005;33(4):379–385. doi:10.1111/j.1442-9071.2005.01015.x [CrossRef]
- Shimada H, Nakashizuka H, Hattori T, Mori R, Mizutani Y, Yuzawa M. Double staining with brilliant blue G and double peeling for epiretinal membranes. Ophthalmology. 2009;116(7):1370–1376. doi:10.1016/j.ophtha.2009.01.024 [CrossRef]
- Kim J, Rhee KM, Woo SJ, Yu YS, Chung H, Park KH. Long-term temporal changes of macular thickness and visual outcome after vitrectomy for idiopathic epiretinal membrane. Am J Ophthalmol. 2010;150(5):701–709. doi:10.1016/j.ajo.2010.05.037 [CrossRef]
- Inoue M, Morita S, Watanabe Y, et al. Inner segment/outer segment junction assessed by spectral-domain optical coherence tomography in patients with idiopathic epiretinal membrane. Am J Ophthalmol. 2010;150(6):834–839. doi:10.1016/j.ajo.2010.06.006 [CrossRef]
- Shimozono M, Oishi A, Hata M, et al. The significance of cone outer segment tips as a prognostic factor in epiretinal membrane surgery. Am J Ophthalmol. 2012;153(4):698–704, 704.e1. doi:10.1016/j.ajo.2011.09.011 [CrossRef]
- Sato S, Inoue M, Kobayashi S, Watanabe Y, Kadonosono K. Primary posterior capsulotomy using a 25-gauge vitreous cutter in vitrectomy combined with cataract surgery. J Cataract Refract Surg. 2010;36(1):2–5. doi:10.1016/j.jcrs.2009.07.049 [CrossRef]
Disruption of Outer Retinal Layer
|Preoperation||27 (62.8%)||16 (37.2%)|
|3 Years After the Operation||9 (20.9%)||34 (79.1%)|
Influence of Preoperative Disruption of Outer Retinal Layer at 3 Years After Surgery
|Preoperative Disruption of Outer Retinal Layer||Mean logMAR VA||Mean CRT|
|Presence||0.21 ± 0.28*||334.9 ± 48.7|
|Absence||–0.001 ± 0.11||348.6 ± 48.2|