Change in Full-Field ERGs after Macular Translocation Surgery with 360° Retinotomy

Abstract

purpose. One of the methods used in macular translocation (MT) surgery for subfoveal neovascularization is to create a temporary total retinal detachment followed by a 360° retinotomy. The whole retina is then shifted from the original surface of the retinal pigment epithelium (RPE), resulting in an unusual retina and RPE complex. The purpose of this study was to assess retinal function after MT surgery.

methods. Full-field electroretinograms (ERGs) were recorded before and 4 to 8 months (mean, 5.4 months) after MT surgery with a 360° retinotomy in 15 consecutive patients with age-related macular degeneration (10 eyes), high myopia (4 eyes), and polypoidal choroidal vasculopathy (1 eye). Their ages ranged from 57 to 74 years. The angle of rotation of the retina ranged from 18° to 45° (mean ± SE, 30 ± 2°). In addition to the recording of the standard rod and mixed rod–cone ERGs after 30 minutes of dark adaptation, the cone single flash and 30-Hz flicker ERGs were recorded immediately after a light-adapting background was turned on (LA₀) and also after 10 minutes of light adaptation (LA₁₀).

results. The mean amplitude of the full-field ERGs was reduced after surgery by 44% for the rod response, by 24% for the mixed rod–cone b-wave, by 12% and 35% for the cone single-flash b-wave at LA₀ and 30-Hz flicker ERGs at LA₀, respectively. The mean implicit times were delayed by 8 msec for the rod response, by 2 msec for the mixed rod–cone oscillatory potential (OP1), by 4 msec for the cone single-flash b-wave at LA₀, and by 6 msec for the 30-Hz flicker at LA₀.

conclusions. These results demonstrated a functional alteration in both the rod and cone components of the ERGs for the entire retina after MT surgery.

Age-related macular degeneration (AMD) is the leading cause of legal blindness among the elderly in industrialized countries. Laser photocoagulation ^{1 2} and surgical removal of subretinal choroidal neovascularization (CNV) ^{3 4 5} have been effective in reducing the severe loss of vision in patients with the exudative type of AMD. These treatments, unfortunately, damage the underlying retinal pigment epithelium (RPE) and choriocapillaris, which can lead to alterations of retinal function and loss of reading ability after the atrophy of the RPE. 

Selective photocoagulation of the CNV under the fovea by photodynamic therapy is one of the new promising methods; however, its effectiveness has been demonstrated in only a limited number of patients with AMD. ⁶  

Other diseases associated with subretinal CNV are high myopic chorioretinal degeneration, ^{7 8} idiopathic polypoidal choroidal vasculopathy, ⁹ and angioid streaks. The consequences of these diseases resemble those in AMD if the CNV exists under the fovea, although the CNV is generally smaller than that in AMD. 

Foveal or macular translocation (MT) surgery is an operation that moves the fovea from the diseased RPE onto healthy RPE. ^{10 11} This type of surgery has the potential of improving or preserving central visual function in eyes after the removal of the subfoveal CNV. A number of macular translocation case series have been published ^{12 13 14 15 16 17 18 19 20 21 22 23 24 25 26} with various modifications of the original surgical procedures first reported by Machemer and Steinhorst in 1993. ^{25 26} Today, two techniques of macular translocation are performed, and both techniques involve the creation of retinal detachment. One major technique involves a detachment of the entire retina from the RPE by a subretinal infusion of fluid, with a 360° circumferential retinotomy followed by the rotation of the macula. During this meticulous manipulation, some retinal function may be lost, although central visual function may be improved. 

To evaluate the efficacy of this surgical procedure, assessment should be made of not only macular function but also of the whole retina before and after surgery. To date, only subjective assessments have been made with measurements of visual acuity and microscotometry, ^{12 13 14 15 16 17 18 19 20 21 22 23 24 25 26} except for one case report of the use of focal macular electroretinograms (ERGs) to assess the outcome of the MT surgery. ¹⁶  

In the present study, we tested the retinal function of the entire retina by recording full-field ERGs before and after MT surgery. Additionally, ERGs were recorded during the course of light adaptation to examine the changes in the cone system during this process. We show that each component of the full-field ERGs was depressed by different degrees, whereas the visual acuity improved in all eyes. 

Patients and Methods

Patients

Fifteen patients who underwent MT surgery for subfoveal choroidal neovascularization at the Nagoya University Hospital from March through October 2000 were examined by full-field ERGs before and after surgery. These patients were the first 15 of 26 consecutive surgical cases in our institute that we could observe for at least 6 months after surgery. The remaining nine patients were also successfully treated; however, the ERGs have not been performed yet. There have been no serious complications, such as retinal detachment and proliferative vitreoretinopathy, in the 26 patients to date. CNV recurred at the previous foveal location in two patients with AMD and in one patient with high myopia, and all were treated by laser photocoagulation. 

The data from 10 patients with AMD, aged from 62 to 79 years, and one 77-year-old patient with polypoidal choroidal vasculopathy were studied. An additional four patients with high myopic chorioretinal degeneration, aged 61 to 77 years, were also studied. These 15 patients had new or recurrent CNV that involved the geometric center of the foveal avascular zone. 

The preoperative best corrected visual acuity (BCVA) ranged from hand motion to 20/133 (BCVA was measured with the Standard Japanese Visual Acuity Chart and converted to Snellen visual acuity), and the diameter of the CNVs ranged from 0.3 to 2.6 disc diameters (mean, 1.2 ± 0.2). Before the MT surgery, one patient with AMD had been treated by surgical removal of the CNV followed by the photocoagulation to the recurrent CNV. Two other patients with AMD had been treated with radiation therapy of 20 Gy for 10 days, and another patient with AMD had been treated with photocoagulation. 

The angle of rotation of the retina ranged from 18° to 45° (mean ± SE, 30 ± 2°). The follow-up period after the MT surgery in these 15 patients ranged from 6 to 14 months (mean, 10.1 ± 0.6). The follow-up after successful silicone oil removal ranged from 3 to 13 months (mean, 5.3 ± 0.8). 

This research was conducted in accordance with the institutional guidelines of Nagoya University and conformed to the tenets of the World Medical Association’s Declaration of Helsinki. Informed consent was obtained from each patient for the surgery and for the pre- and postoperative ERGs after providing information on other treatment options, including photocoagulation, removal only, and observation alone. 

Surgical Methods

Our technique was a modification of the 360° retinotomy method of Machemer and Steinhorst. ^{25 26} Initially, a lensectomy with preservation of the anterior lens capsule was performed, followed by a complete removal of the vitreous by pars plana vitrectomy with infusion of Ca⁺²- and Mg⁺²-free balanced salt solution (BSS) for 10 minutes (patients 1–8) or with BSS-plus (patients 9–15). Four separate dome-shaped retinal detachments were then created by subretinal infusion of BSS. Fluid–air exchange was then performed, which led to the coalescence of the detachments, followed by a 360° retinotomy at the ora serrata with automated scissors. This was followed by the injection of heavy perfluorocarbon liquid and the rotation of the whole retina around the axis of the optic disc with further injection of the perfluorocarbon liquid during the rotation. The site of the 360° retinotomy and the holes artificially created for detaching the retina were sealed by endophotocoagulation. The total number of photocoagulation spots ranged from 806 to 1348 (mean, 1175 ± 41). After this procedure, an exchange of perfluorocarbon liquid with silicone oil (1000 centistokes) was performed. The operation time ranged from 163 to 280 minutes (mean, 207 ± 9.1). 

After 2 to 3 months, the silicone oil was removed, and an intraocular lens was implanted in patients who requested the implantation. 

Full-Field ERGs

Full-field ERGs were recorded before and 4 to 8 months (mean, 5.4 ± 1.2) after the MT surgery. Repeat ERGs were recorded from one eye at 3 and 8 months in one patient and at 3 and 6 months in another patient. The data at 8 and 6 months were used for the overall analysis. 

Full-field ERGs were recorded after pupil dilatation with 0.5% tropicamide and 0.5% phenylephrine hydrochloride and 30 minutes of dark adaptation. The scotopic (rods) ERGs were elicited by a blue stimulus at an intensity of 5.2 × 10⁻³ cd/sec · m². The rod–cone mixed single flash (bright white) ERGs were elicited by a white stimulus at an intensity of 44.2 cd/sec · m². 

The cone single-flash ERG and the 30-Hz flicker responses were elicited with a white stimulus intensity of 4 cd/sec · m² and 0.9 cd/sec · m², respectively, on a white background of 68-cd/m² luminance. To follow the enhancement of the cone ERGs during the course of light adaptation, these two cone–mediated ERGs were first recorded from the dark-adapted eye immediately after the white background was turned on (LA₀) and again after 10 minutes of light adaptation (LA₁₀). These two sets of responses were used to determine the enhancement of the cone ERGs during light adaptation. ²⁷  

Results

Visual Acuity

The preoperative (abscissa) and postoperative (ordinate) visual acuities are plotted in Figure 1A . The numbers in the symbols correspond to the patient number and are the same in all the graphs (Figs. 1B 1C) and ERGs (Figs. 2 3 4) . The postoperative BCVA ranged from 20/133 to 20/30 in 10 eyes with AMD, 20/200 to 20/25 in the 4 eyes with high myopic chorioretinal atrophy and CNV, and 20/30 in the eye with polypoidal choroidal vasculopathy (Fig. 1A) . 

Full-Field ERGs

The pre- and postoperative amplitude of rod responses are plotted as case numbers in Figure 1B , and the b-wave amplitude of mixed rod–cone response (bright-white-flash ERG) are plotted in Figure 1C . Each waveform of the full-field ERGs for rod response, mixed rod–cone response (bright white flash ERG), cone single flash response at LA₀ and LA₁₀, and 30-Hz flicker response at LA₀ and LA₁₀ for representative 10 cases are shown in Figure 2 3 4 . 

The mean amplitude of the ERGs in the 15 eyes was reduced by 44% for the rod response (P < 0.001), by 22% for the a-wave of the mixed rod–cone response (P < 0.01), by 24% for the b-wave of the mixed rod–cone response (P < 0.01), by 55% for the sum of oscillatory potentials (OP)1, 2, and 3 (P < 0.001), by 12% for the b-wave of the cone single flash (LA₀; P < 0.01), and by 35% for the 30-Hz flicker (LA₀; P < 0.01; Wilcoxon signed-rank test). The reduction in amplitude after MT surgery was statistically significant for all components except for the cone single-flash b-wave (LA₀). The number of photocoagulation spots was correlated with the decrease in rod response after surgery (r = 0.518; P = 0.047). 

The mean implicit times were delayed by 8 msec for the rod response (P < 0.01), by 1 msec for the mixed rod–cone response a-wave (P > 0.05), by 2 msec for the OP1 (P < 0.01), by 4 msec for the cone single-flash b-wave (LA₀; P < 0.01), and by 6 msec for the 30-Hz flicker (LA₀; P < 0.01). The implicit times were significantly delayed for all components, except that of the mixed rod–cone response a- and b-waves after surgery (Wilcoxon signed-rank test; Fig. 5 ). 

The time between the ERG recording and MT surgery was not correlated with the amplitude decrease and delayed implicit time after surgery. In the two eyes with repeat ERG recordings, no remarkable change was found between the first and second set of data. 

In addition to the amplitude reduction and implicit time delay for each ERG component, we noted an interesting phenomenon in the cone-mediated ERGs. It has been shown that the amplitude of the cone-driven ERGs (cone single-flash ERG and 30-Hz flicker ERG) is relatively small when recorded from a dark-adapted eye when the light-adapting field is first turned on (LA₀) but then increases gradually during the course of light adaptation (LA₁₀). ²⁷ The ratio of the cone b-wave amplitude after 10 minutes of light adaptation (LA₁₀) to that immediately after turning the light-adapting field on (LA₀) is reported to be between 1.5 and 2.5 in normal eyes. ²⁷ This ratio for the cone single-flash b-wave in our cases was 1.63 ± 0.07 (mean ± SE) before surgery and 1.25 ± 0.08 after surgery, and for the 30-Hz flicker ERGs, the ratio was 1.20 ± 0.06 before surgery and 0.91 ± 0.08 after surgery. These differences in the ratios before and after surgery were statistically significant (cone single-flash b-wave, P = 0.0022; 30-Hz flicker, P = 0.0097; Wilcoxon-signed rank test). 

Discussion

Our results demonstrated that MT surgery preserved or improved visual acuity in all 15 eyes. However, when assessed by full-field ERGs, the amplitudes of both the rod and cone components of the full-field ERGs were reduced by 12% to 50%, and the implicit times for various components were prolonged. This suggests that the mass function of the retina had been altered by the surgery. 

Histopathologically, Machemer and Steinhorst ²⁵ demonstrated a remarkable intactness of the outer segments of the rods and the pigment epithelial cells by scanning and transmission electron microscopic studies after acute retinal detachment. They stated that the ultrastructural changes were different from those with a slower course after experimental retinal detachment. 

Our findings of reduced retinal function may have resulted partly from the loss of peripheral retinal function caused by the retinotomy and the three to five rows of endophotocoagulationto reattach the peripheral retina. However, the function of the more posterior central retina must be relatively well preserved in the majority of patients as manifested by the improved visual acuity. 

Imai et al. ²⁸ reported on the histologic and electrophysiological findings after limited macular translocation in rabbit eyes. They reported mild morphologic damage to the outer retinal layers with normal morphology of the inner layers, although the results in the rabbit eye may not be comparable to those in the human eye because of the rabbits’ avascular retina. They reported that there was some change in the vertical alignment of the photoreceptors, as well as some loss of the outer segment discs. The reduction of the dark-adapted ERGs suggests a transient reduction in retinal function. In addition, they reported a 70% reduction in the a- and b-waves on the 3rd postoperative day that recovered to a 30% reduction on the 14th postoperative day. In the vascularized retina of primates, good anatomic recovery after experimental retinal detachment has also been demonstrated. ^{29 30} The retinal function in the detached area would be expected to recover better in humans than in experimental rabbits after macular translocation. 

It is well known that cone-mediated ERGs show an increase of amplitude during light-adaptation; however, the mechanism for the increase of amplitude of the ERG components during light adaptation has still not been established. The enhancement has been proposed to arise from changes in the standing potential of the eye, changes of the cone photoreceptors, and cone–rod interaction. In the incomplete type of congenital stationary night blindness, there is an exaggerated increase of the 30-Hz flicker amplitude, ³¹ but the physiological mechanism for this phenomenon has still not been determined. 

In the present study, we noted that the increase in the amplitude of the cone-mediated ERG during light adaptation became significantly less after MT surgery. We found unexpectedly that the mean amplitude of the 30-Hz flicker ERG was smaller after full light adaptation than just after the light-adapting background was turned on, which has never been reported in normal retinas. It is difficult to explain why the enhancement was altered after this surgery, because the exact mechanism for this phenomenon has not been determined. However, we have found that the magnitude of increase was minimal in the central retina and was greater in the peripheral retina using the multifocal ERG technique. ³² This means that the peripheral retina contributes more to the amplitude increase than the central retina. Thus, that the increase of the amplitude of the cone-mediated ERG during light adaptation becomes less after this surgery may indicate a predominant dysfunction of the peripheral retina. 

The flicker ERGs in the light-adapted state was most reduced. This reduction probably resulted from the unusual decrease of the amplitude after the light adaptation, or the failure of an increase in amplitude during light adaptation. 

The MT surgery causes not only peripheral retinal damage (360° retinotomy and photocoagulation), but also the temporary total retinal detachment and shift of the entire retina to a different RPE surface after surgery. Thus, this suggests that several factors may be involved in the decrease in enhancement. 

Severe postoperative complications such as recurrent retinal detachment have been reported in approximately 30% of patients after MT, ¹⁰ although none of our patients had any complications. MT surgery with a 360° retinotomy may be feasible from the standpoint of macular function, if severe complications are technically avoided; however, a certain degree of peripheral retinal function will be depressed. Although this study was conducted in small numbers of patients and they were the first 15 surgeries performed by one experienced surgeon, the results do not appear to be related to the learning of the surgical procedures. 

In this study, the ERGs were recorded after 4 to 8 months after the surgery, and most were recorded between 4 to 6 months, which seems to be a relatively short period after surgery. However, there was no significant correlation between the length of time to ERG testing after surgery and the amplitude decrease or the delayed implicit time. In addition, the two eyes that were tested twice showed very little change in the amplitudes and implicit times. Because retinal function may recover or may even deteriorate after longer periods, additional follow-up is planned. 

In conclusion, a reduction of physiological function in the peripheral retina after MT surgery with a 360° retinotomy was suggested. The cone components of the ERGs recorded after light adaptation were significantly more deteriorated than those recorded immediately after turning the light-adapting field on after dark adaptation. 

 Supported by Grants-in-Aid 12470361 and 11470363 from the Ministry of Education, Science and Culture, Tokyo, Japan.

 Submitted for publication April 18, 2001; revised September 6, 2001; accepted October 8, 2001.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked“ advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Preoperative (abscissa) and postoperative (ordinate) values for the 15 eyes that underwent the MT surgery. The numbers correspond to the patient number and are the same in all the graphs. (A) BCVA; (B) rod b-wave amplitude; and (C) mixed rod–cone response (bright white flash) b-wave amplitude.

Figure 2.

 

Waveform of full-field ERGs for rod (left) and mixed rod–cone (right) response (bright white flash) before and after MT surgery in 10 patients.

Figure 3.

 

Waveform of full-field cone responses recorded from the dark-adapted eye (left) immediately after a light-adapting field (LA₀) and (right) 10 minutes after light adaptation (LA₁₀) was turned on. The ERGs recorded before and after MT surgery in 10 patients are shown.

Figure 4.

 

Waveform of the 30-Hz flicker responses recorded from the dark-adapted eye (left) immediately after a light-adapting field (LA₀) and (right) 10 minutes after the light adaptation (LA₁₀) was turned on. The ERGs recorded before and after MT surgery in 10 patients are shown. The amplitude in both states was markedly reduced after surgery and, in several cases, the amplitude was smaller after full light adaptation than while the eye was not completely light adapted, which has never been reported in normal retinas.

Figure 5.

 

The means (±SEs) of the percentage decrease in the amplitude (left) and the delayed implicit times (right) of each response after MT surgery. *Total amplitudes of oscillatory potentials (OP1 + OP2 + OP3) of mixed rod–cone response (bright white flash ERG) and the implicit time of OP1.