Jpn J Ophthalmol (2014) 58:455–461 DOI 10.1007/s10384-014-0345-1
CLINICAL INVESTIGATION
Long-term outcomes of 3 surgical adjuvants used for internal limiting membrane peeling in idiopathic macular hole surgery Noriko Mochizuki • Teiko Yamamoto Hiroshi Enaida • Tatsuro Ishibashi • Hidetoshi Yamashita
•
Received: 26 November 2013 / Accepted: 11 July 2014 / Published online: 9 September 2014 Ó Japanese Ophthalmological Society 2014
Abstract Purpose Indocyanine green (ICG), an adjuvant used for peeling of the internal limiting membrane (ILM) during vitreous surgery for idiopathic macular hole (MH), has been reported to be toxic, possibly affecting postoperative visual acuity. We compared the long-term outcomes (within 2 years) of brilliant blue G (BBG), ICG, and triamcinolone acetonide (TA). Patients and methods This study involved 97 eyes of 94 patients who underwent vitreous surgery for MH at the Yamagata University Hospital between June 2002 and November 2010. The surgical adjuvants used were BBG for 15 eyes, ICG for 61 eyes, and TA for 21 eyes. We compared the postoperative visual acuities, initial closure rates, final closure rates, and complications of the 3 groups. Results In all 3 groups, the visual acuity significantly improved after surgery. The magnitude of the improvement at 2 years after surgery was significantly better in the BBG group than in the ICG group (Mann–Whitney test, P = 0.020). The postoperative visual acuity did not
N. Mochizuki (&) H. Yamashita Department of Ophthalmology and Visual Sciences, Yamagata University Faculty of Medicine, 2-2-2 Iida-Nishi, Yamagata, Yamagata 990-9585, Japan e-mail: [email protected] T. Yamamoto Kanamecho Yamamoto Eye Clinic, Tokyo, Japan H. Enaida Department of Ophthalmology, Faculty of Medicine, Saga University, Saga, Japan T. Ishibashi Department of Ophthalmology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
significantly differ between the BBG and TA groups (P = 0.627) or between the ICG and TA groups (P =0 .137). Thus, the surgery using BBG resulted in a significantly better outcome in visual acuity than did the surgery using ICG. The 3 groups did not differ in initial or final closure rates or in incidence of complications. Conclusion Analysis of the long-term outcomes of vitreous surgeries provided evidence that BBG is a useful adjuvant for ILM peeling. Keywords Brilliant blue G Indocyanine green Internal limiting membrane peeling Macular hole Triamcinolone acetonide
Introduction In 1991, Kelly and Wendel [1] reported the successful closure of idiopathic macular holes (MH) with vitreous surgery combined with gas tamponade. Since then, this technique has been used in many hospitals and clinics. In earlier procedures, after the posterior vitreous cortex was detached, the air–fluid exchange was performed with an intravitreal injection of sulfur hexafluoride gas (SF6). The success rate in closing the hole with this method was 50 to 80 %. The closure rate was further increased by addition of autologous serum [2] or by peeling of the internal limiting membrane (ILM), the innermost layer of the retina [3]. However, because the ILM is thin and transparent, its removal while still leaving the retinal nerve intact is challenging. Since a report that the closure rate was improved by staining the ILM with indocyanine green (ICG) for better visual identification, ILM peeling has become more reliable [4], leading to further increases in MH closure rates. At present, ICG and triamcinolone
123
456
N. Mochizuki et al.
Table 1 Patients’ backgrounds
BBG
ICG
TA
Number (eyes)
15
61
21
Male
9
31
9
Female
6
30
12
Age, (years)*
68.6 ± 7.4
65.9 ± 8.6
63.2 ± 7.6
0.24
Duration of visual symptoms, (months)*
2.57 ± 2.71
7.34 ± 10.32
7.81 ± 12.64
0.01
MH diameter, (lm)*
923.33 ± 290.75
820.11 ± 303.17
700.71 ± 267.64 \0.001
Stage 2
2 (13.3)
10 (16.4)
9 (42.9)
Stage 3
5 (33.3)
25 (41.0)
5 (23.8)
Stage 4
8 (53.3)
26 (42.6)
7 (33.3)
0.86 ± 0.22 12 (80.0)/3 (20.0)
0.84 ± 0.31 60 (98.4)/1 (1.6)
0.94 ± 0.30 21 (100)/0 (0)
MH stage, n (%)
BBG brilliant blue G, ICG indocyanine green, TA triamcinolone acetonide, MH macular hole * Mean ± SD
Materials and methods Study design and patients This was a retrospective cohort study involving 97 eyes of 94 patients who underwent vitreous surgery for idiopathic MH and were followed up for over 3 months at Yamagata University Hospital between June 2002 and November 2010. The patients comprised 48 men (49 affected eyes) and 46 women (48 affected eyes). They were divided into 3 groups according to the dye used for ILM peeling: the BBG group (15 eyes), the ICG group (61 eyes), and the TA group (21 eyes). No significant differences were detected between any 2 of these 3 groups in terms of the male-tofemale ratio (Fisher exact test, P = 0.62), age (Kruskal– Wallis test, P = 0.24), MH stage (2 to 4) distribution (Fisher exact test, P = 0.14), or preoperative logarithm of
123
0.62
0.14
Preoperative visual acuity, logMAR* Phakic eyes, n (%)/Pseudophakic eyes, n (%)
acetonide (TA) are used for ILM peeling. ICG, however, has been reported to induce the formation of reactive oxygen species when exposed to light, leading to injury of the retinal pigment epithelial cells and decreased retinal function [5]. ICG has also been shown to be histologically toxic, causing loss of retinal pigment epithelial cells [6]. As regards TA, it is not designed to stain the ILM. If TA detaches from the ILM because of water flow or other factors, the ILM flap becomes transparent, possibly making it difficult for the surgeon to identify the flap. To address these limitations of TA, the usefulness of a new dye for staining the ILM, brilliant blue G (BBG), has been widely investigated [7, 8]. In this study, we report the use of BBG as an adjuvant in ILM peeling during MH surgery We also present results of a comparison of the long-term outcomes (within 2 years) of MH surgeries in which BBG was used with those of MH surgeries in which ICG or TA was used.
P
0.42 0.03
Fig. 1 Optical coherence tomography (OCT) of macular hole diameter. The diameter was measured at the base of the hole
the minimum angle of resolution (logMAR) visual acuity (VA) (Kruskal–Wallis test, P = 0.42) (Table 1). The duration of visual symptoms was significantly shorter in the BBG group than in the other 2 groups (Kruskal–Wallis test, P = 0.01). In accordance with a previous report [9], we used optical coherence tomography (OCT) to measure the MH. We measured the largest diameter of the hole, i.e., the diameter at the hole base (Fig. 1). The MH was significantly larger in the BBG group than in the other 2 groups (Kruskal–Wallis test, P \ 0.001). The proportion of phakic eyes before surgery was significantly lower in the BBG group than in the other 2 groups (Fisher exact test, P = 0.03). Surgical technique All patients underwent 3-port pars-plana vitrectomy, mainly with microincisional vitrectomy (25-gauge) under retrobulbar anesthesia. In addition, phacoemulsification and intraocular lens implantation (cataract surgery) were
BBG, ICG, and TA compared in ILM peeling
457
adjusted to complete the surgery. In the BBG group, the fluid was substituted with 1 ml of 100 % SF6 in all eyes. In the ICG group, the fluid was substituted with air in 29 eyes, 20 % SF6 (total substitution) in 9 eyes, 1 ml of 100 % SF6 in 22 eyes, and 12 % C3F8 (total substitution) in 1 eye. In the TA group, the fluid was substituted with air in 14 eyes and with 1 ml of 100 % SF6 in 7 eyes. Although recent studies have shown that the duration of prone positioning time after MH surgery was shortened [11], we instructed the patients to maintain the prone position until closure of the MH was confirmed. This study was approved by the institutional review board of Yamagata University and performed in accordance with the ethical standards of the 1989 Declaration of Helsinki. Fig. 2 Brilliant blue G (BBG)-assisted internal limiting membrane (ILM) peeling for macular hole (MH). After ILM removal, a difference is clearly visible in the retinal surface color between the area from which the ILM was removed and the surrounding area
performed for 5 of the 15 eyes (33.3 %) in the BBG group, 36 of the 61 eyes (59.0 %) in the ICG group, and 10 of the 21 eyes (47.6 %) in the TA group. For stage 2 and 3 cases, TA (Kenacort-A; Bristol-Myers Squibb, New York, NY, USA) was used for posterior vitreous separation. The supernatant was removed using the method reported by Nishimura et al. [10], which involves a liquid filter and a balanced salt solution (BSS) (BSS Plus; Alcon, Fort Worth, TX, USA). Then, a suspension of TA (8 mg/ml) was prepared with 5 ml of BSS. ILM peeling was performed as follows. In the ICG group, ICG was used at a concentration of 0.125 %. The prepared ICG solution was injected by spraying over the posterior pole while avoiding the MH. This was immediately followed by washing with BSS. After confirmation that the ILM was stained green, a scaffold was created on the ILM with a V-lance. ILM forceps were then used to peel the ILM for approximately 3 disc diameters around the MH. In the TA group, TA (0.2–0.3 ml) was applied thinly to the posterior pole, which was followed by the ILM peeling. For the BBG group, BBG was made into a 0.25 mg/ml solution with BSS. It was passed through a filter and injected by spraying over the posterior pole, while avoiding the hole. This was followed immediately by washing with BSS. After confirmation that the ILM was stained blue, the ILM was peeled in the same way as that described for the ICG group (Fig. 2). After the ILM removal, air–fluid exchange was performed. If needed, the fluid was substituted with gas. The wound was then closed, and the intraocular tension was
Methods of evaluation The surgery outcomes of the different adjuvants were compared by determining the VA and confirming the closure rates by means of OCT. In the cases in which BBG was used, the presence/absence of atrophy of the retinal pigment epithelium was checked by fluorescein angiography (FA) 6 months after the surgery. Statistical analysis Decimal VA was converted into logMAR units. The logMAR units before surgery and at 3 months, 6 months, 1 year, 1.5 years, and 2 years after surgery were compared between 2 groups and analyzed using the Wilcoxon signedrank test. The initial and final closure rates and the presence/absence of complications were tested using the Fisher exact test. Multivariate regression analysis was conducted using the available VA data of 69 eyes (BBG: 14 eyes; ICG: 42 eyes; TA: 13 eyes) until 2 years after surgery. The magnitude of VA improvement at 2 years after surgery (logMAR VA at 2 years minus preoperative logMAR VA) was the dependent variable; the independent variables were sex, age, duration of visual symptoms, MH diameter, MH stage, preoperative logMAR VA, tamponade agents, cataract extraction during MH surgery, cataract extraction after MH surgery, timing of cataract extraction (pseudophakia preoperatively, cataract extraction during MH surgery, cataract extraction after MH surgery, phakia during follow-up), and adjuvants. The magnitude of VA improvement at 2 years after surgery was compared among the 3 groups using the Kruskal–Wallis test, and then between 2 groups (BBG vs TA, BBG vs ICG, or TA vs ICG) using the Mann–Whitney test. All statistical analyses were conducted with PASW Statistics 18 software (IBM Corporation, Armonk, NY,
123
458
USA). Probability values of less than 0.05 were considered significant.
Results The intensity of the chromatic response of the ILM to BBG (0.25 mg/ml) was comparable to that of the ILM response to 0.125 % ICG. Once peeling of the ILM began in the BBG group, the peeled area became clearly distinguishable from the remaining area, allowing the subsequent peeling steps to be performed smoothly (Fig. 2). The area affected by the epimacular membrane was not positively stained. Figure 3 shows the time course of VA changes in each adjuvant group. In the BBG group, the mean logMAR VA before surgery was 0.86 ± 0.22 (n = 15), and after surgery 0.40 ± 0.25 (n = 15) at 3 months, 0.37 ± 0.20 (n = 15) at 6 months, 0.33 ± 0.20 (n = 15) at 1 year, 0.21 ± 0.18 (n = 14) at 1.5 years, and 0.16 ± 0.14 (n = 14) at 2 years. A significant improvement in VA relative to the preoperative acuity was noted at 3 months after surgery (Wilcoxon signed-rank test, P \ 0.001), and this improved acuity was maintained for 2 years (P \ 0.001). In the ICG group, the mean logMAR VA before surgery was 0.84 ± 0.31 (n = 61), and after surgery 0.59 ± 0.36 (n = 61) at 3 months, 0.57 ± 0.36 (n = 61) at 6 months, 0.44 ± 0.35 (n = 60) at 1 year, 0.37 ± 0.30 (n = 53) at 1.5 years, and 0.35 ± 0.28 (n = 42) at 2 years. A significant improvement in VA relative to the preoperative acuity was noted 3 months after surgery (P \ 0.001), and this improvement was maintained for 2 years (P \ 0.001). In the TA group, the mean logMAR VA before surgery was 0.94 ± 0.30 (n = 21), and after surgery 0.44 ± 0.34 (n = 21) at
Fig. 3 Visual acuity changes after vitrectomy in the brilliant blue G (BBG), indocyanine green (ICG), and triamcinolone acetonide (TA) groups. The circles represent the BBG group; the squares, the ICG group; and the triangles, the TA group. In all 3 groups, the visual acuity improves significantly after surgery. (*Wilcoxon signed-rank test, P \ 0.05)
123
N. Mochizuki et al.
3 months, 0.41 ± 0.37 (n = 21) at 6 months, 0.33 ± 0.34 (n = 21) at 1 year, 0.30 ± 0.29 (n = 17) at 1.5 years, and 0.28 ± 0.30 (n = 13) at 2 years. A significant improvement in VA relative to the preoperative acuity was noted 3 months after the surgery (P \ 0.001), and this improvement was maintained for 2 years (P \ 0.001). The magnitude of VA improvement at 2 years after the surgery was 0.69 ± 0.22 in the BBG group, 0.49 ± 0.26 in the ICG group, and 0.66 ± 0.29 in the TA group. The initial closure rate was 86.7 % (n = 15) in the BBG group, 86.9 % (n = 61) in the ICG group, and 90.5 % (n = 21) in the TA group; no significant intergroup difference was detected (Fisher exact test, P = 1.00). In 1 case of the ICG group, because the patient did not agree to further surgery, the MH could not be closed. In all of the other cases, the MH was closed. No significant differences in the final closure rates were detected between any 2 of the BBG group (100 %, n = 15), ICG group (98.4 %, n = 61), or TA group (100 %, n = 21) (Fisher exact test, P = 1.00). With respect to the intraoperative complications, a retinal tear in the peripheral retina occurred in 2 eyes of the BBG group, 11 eyes of the ICG group, and 2 eyes of the TA group. In these cases, sufficient vitreous dissection was conducted intraoperatively, and laser photocoagulation was applied to surround the break completely. No significant differences were found in terms of the presence/absence of retinal tear development between any 2 of the 3 groups (Fisher exact test, P = 0.67). After surgery, no patient from any group developed retinal detachment, intraocular hypotension, intraocular hypertension, or endophthalmitis. Reopening of the MH occurred in 1 eye (1.6 %) of the ICG group. The probabilities of MH reopening in the 3 groups did not differ significantly (Fisher exact test, P = 1.00). During the 2 years of followup, cataract surgery was performed in 4 of the 15 eyes (26.7 %) of the BBG group, 9 of the 61 eyes (14.8 %) of the ICG group, and 3 of the 21 eyes (14.3 %) of the TA group (Fisher exact test, P = 0.51). Postoperative FA in the BBG group revealed no abnormalities, such as atrophy of the retinal pigment epithelium or leakage of the dye. The preoperative logMAR VA was found to correlate significantly with the postoperative VA (P \ 0.001; standardized partial regression coefficient = 0.63; Table 2). Sex, age, duration of the visual symptoms, MH diameter, MH stage, preoperative logMAR VA, tamponade agents, timing of the cataract extraction, and adjuvants were not significantly correlated with the postoperative VA. A comparison of the magnitude of VA improvement at 2 years after surgery among the 3 groups indicated significant intergroup differences (Kruskal–Wallis test, P = 0.04). The probability values as determined by the Mann–Whitney test for the differences in VA were as follows: P = 0.02 between the BBG and ICG groups,
BBG, ICG, and TA compared in ILM peeling
459
Table 2 Correlation of the independent variables with postoperative visual acuity Standardized partial regression coefficient
P value
Sex
0.178
0.090
Age
-0.176
0.172
Duration of visual symptoms
-0.190
0.105
MH diameter
-0.071
0.555
MH stage
-0.077
Preoperative visual acuity, logMAR Tamponade agent
0.634
0.483 \0.001*
0.034
0.783
Timing of cataract extraction
-0.155
0.221
Adjuvant
-0.088
0.522
* Multiple linear regression analysis, P \ 0.05
P = 0.63 between the BBG and TA groups, and P = 0.14 between the ICG and TA groups. Thus, the use of BBG resulted in a significantly better time course of VA improvement when compared with the ICG group, whereas the time courses of VA improvement did not significantly differ between the BBG and TA groups or between the ICG and TA groups.
Discussion The diameter and duration of symptoms of MH have been listed as factors affecting the postoperative VA [12, 13]. In this study, the MH diameter, length of time until the surgery, and proportion of phakic eyes before surgery differed among the individual patients. To evaluate the influence of these factors, we performed multivariate regression analysis of the available VA data of 69 eyes for 2 years after surgery. The preoperative VA was found to correlate significantly with the postoperative VA. The longer the period between diagnosis and surgery or the larger the MH diameter became, the poorer the improvement in postoperative VA was. However, the influence of these factors was not significant, suggesting that they do not affect the evaluation of the usefulness of individual adjuvants. Adjuvants conventionally used for ILM peeling include ICG and TA. However, as regards ICG, culture of human retinal pigment epithelial cells in the presence of ICG coupled with light exposure was reported to decrease the mitochondrial activity associated with cell survival rate [5]. In addition, cases of visual field defects after ILM staining with ICG were attributed to the influence of light [14]. In rat eyes, ICG induced histologic disorders, including destruction of the retinal lamellar structure and loss of
retinal pigment epithelial cells [6]. The toxicity of ICG has also been reported to be concentration-dependent. In view of these findings, countermeasures to prevent these disorders, such as avoidance of the use of concentrated ICG and of prolonged retinal exposure to light, appear to be essential to ILM peeling. TA is a steroid that is insoluble in water. The Kenacort A solution is slightly acidic (pH 5–7) and contains preservatives, such as benzyl alcohol. Such preservatives can affect the retina and lens [15]. For this reason, TA is usually used after removal of as much of the solvent as possible and subsequent dissolving in BSS. Complications arising from the use of TA include increased intraocular tension, cataract progression, and onset of aseptic [16] or bacterial endophthalmitis. TA has been used for the treatment of various retinal and vitreous diseases. Its use in ILM peeling during surgery for MH was first reported by Kimura et al. [17] and then by Horio et al. [18]. TA remains in the MH floor for 2 months or longer after surgery, but did not result in any particularly hazardous effects [19]. However, the surgeon occasionally cannot identify the stump of the freed ILM if the TA particles are moved away from the membrane surface by the aqueous humor flow or other factors. The present study was designed to compare the safety and effectiveness of 3 adjuvants, BBG, ICG, and TA, in terms of VA, MH closure rates, intraoperative/postoperative complications, intensity of chromatic responses, simplicity of ILM peeling, and histologic and functional toxicities. In all groups, the VA was significantly improved at 3 months after surgery, and the improvement was maintained for 2 years. This improvement was more marked in the BBG group than in the ICG group. The ICG group, however, included patients whose MH did not close until the end of the study and in whom the MH recurred. The patients with nonclosure of the MH were followed up for 1 year, at which time the final VA was 0.05. In the patients with recurrent MH, the MH recurred after a 1.5year follow-up, but the final VA was determined to be 0.9 just before the recurrence. These MH may have affected the VA outcomes. In pig eyes, histopathologic and electrophysiologic evaluations 6 weeks after subretinal injection of ICG, BBG, or TA indicated that ICG evidently injures the retina both histologically and functionally, while BBG and TA do not [20], which is consistent with the results of this study. In that previous study, for patients with an MH or a macular pucker, ILM sections peeled off after staining with trypan blue, BBG, bromophenol blue, Chicago blue, or ICG were morphologically compared under a transmission electron microscope. The cells on the retinal side of the ILM stained with ICG were evidently larger in size and number than those of the ILM stained with any other dye,
123
460
indicating that the retina undergoes more marked morphologic changes when stained with ICG than when stained with BBG [21]. When the percentages of IS/OS junction restorations after MH surgery were compared between patients in whom BBG was used for ILM dissection and patients in whom ICG was used, the percentage of the IS/ OS restorations was significantly higher in the BBG group than in the ICG group. The percentage of patients with final corrected VA of 20/20 or more was higher in the BBG group than in the ICG group. In accordance with these observations, it has been reported that early IS/OS junction restoration affects the prognosis of the postoperative VA [22]. Baba et al. [23] compared the best-corrected VA (BCVA) of eyes in which BBG had been used for ILM peeling and those in which ICG had been used for MH care and found that the average BCVA was significantly better in the BBG group than in the ICG group at 3 and 6 months after surgery. These results are similar to those of the present study. Driven by the recent characterization of BBG as a P2RX7 antagonist [24], Notomi et al. [25] reported that photoreceptor apoptosis involves P2RX7 activation with cleavage of caspase-8 and -9 and mitochondrial-nuclear translocation of apoptosis-inducing factor (AIF). Furthermore, photoreceptor apoptosis can be attenuated by BBG, a pharmacologic P2RX7 antagonist that acts by blocking the interaction between extracellular ATP and P2RX7. Pharmacologic inhibition of P2RX7 has been reported to result possibly in neuroprotection of photoreceptors in cases of subretinal hemorrhage [26]. These previous reports support the present study’s observations that the long-term prognosis of VA was better in the BBG group than in the ICG group. In this study, the closure rates and intraoperative/postoperative complications did not differ among the 3 groups. The intensity of the ILM responses to BBG at a concentration of 0.25 mg/ml was comparable to that of the ILM responses to 0.125 % ICG, and the membrane was stained well with both dyes. In the TA group, the TA was sometimes washed away during the ILM peeling, requiring additional application of TA. In terms of the simplicity of ILM peeling, BBG seems to be equivalent to ICG, and TA, slightly inferior. To date, no reports of the histologic toxicity of BBG have been published. Reports are available on histologic disorders induced by ICG in cultured human retinal pigment epithelial cells and in rat eyes [5, 6]. According to the data collected here during the long-term (2-year) follow-up, the toxicity of BBG was lower than the toxicity that has been commonly noted for ICG. When the magnitude of the VA improvement at 2 years after surgery was compared among the 3 groups, BBG was the best adjuvant, followed by TA, and then by ICG. Furthermore, BBG allowed for easier ILM peeling than did TA.
123
N. Mochizuki et al.
This was a retrospective study. Some factors in the patients’ backgrounds differed among the 3 adjuvant groups. Our sample size was also relatively limited. Despite these limitations, to the best of our knowledge, this study has demonstrated for the first time that BBG provides a safe modality for staining the ILM and that the intensity of the chromatic response to BBG was not inferior to that of the other 2 dyes. We may therefore conclude that BBG is useful as an adjuvant during ILM peeling. Acknowledgments This work was supported in part by a grant from the Japan Medical Association and by a Grant-in-Aid for Scientific Research (#24592673) from the Japanese Ministry of Education, Culture, Sports, Science and Technology. Conflicts of interest N. Mochizuki, None; T. Yamamoto, None; H. Enaida, Patent (Kyushu Univ.); T. Ishibashi, Patent (Kyushu Univ.); H. Yamashita, None. BBG/Hiroshi Enaida and Tatsuro Ishibashi are the inventors of patent application ‘‘Staining composition for staining an ophthalmic membrane’’ US7731941/EP1819366/4200222JP. Kyushu University holds the patent. Kyushu University licensed the patent of BBG to the bio-venture company, Aqumen Biopharmaceuticals NA, Inc. (Fukuoka, Japan). Eventually, the license was granted to the Dutch Ophthalmic Research Center International BV (D.O.R.C.) from Aqumen.
References 1. Kelly NE, Wendel RT. Vitreous surgery for idiopathic macular holes: results of a pilot study. Arch Ophthalmol. 1991;109:654–9. 2. Kusaka S, Sakagami K, Kutsuna M, Ohashi Y. Treatment of fullthickness macular holes with autologous serum. Jpn J Ophthalmol. 1997;41:332–8. 3. Brooks HL. ILM peeling in full thickness macular hole surgery. Vitreoretinal Surg Technol. 1995;7:2. 4. Kadonosono K, Itoh N, Uchio E, Nakamura S, Ohno S. Staining of internal limiting membrane in macular hole surgery. Arch Ophthalmol. 2000;118:1116–8. 5. Sippy BD, Engelbrecht NE, Hubbard GB, Moriarty SE, Jiang S, Aaberg TM Jr, et al. Indocyanine green effect on cultured human retinal pigment epithelial cells: implication for macular hole surgery. Am J Ophthalmol. 2001;132:433–5. 6. Enaida H, Sakamoto T, Hisatomi T, Goto Y, Ishibashi T. Morphological and functional damage of the retina caused by intravitreous indocyanine green in rat eyes. Graefes Arch Clin Exp Ophthalmol. 2002;240:209–13. 7. Enaida H, Hisatomi T, Goto Y, Hata Y, Ueno A, Miura M, et al. Preclinical investigation of internal limiting membrane staining and peeling using intravitreal brilliant blue G. Retina. 2006;26:623–30. 8. Enaida H, Hisatomi T, Hata Y, Ueno A, Goto Y, Yamada T, et al. Brilliant blue G selectively stains the internal limiting membrane/ brilliant blue G-assisted membrane peeling. Retina. 2006;26:631–6. 9. Matsumiya W, Kusuhara S, Shimoyama T, Honda S, Tsukahara Y, Negi A. Predictive value of preoperative optical coherence tomography for visual outcome following macular hole surgery: effects of imaging alignment. Jpn J Ophthalmol. 2013;57:308–15. 10. Nishimura A, Kobayashi A, Segawa Y, Sakurai M, Shirao E, Shirao Y, et al. Isolating triamcinolone acetonide particles for intravitreal use with a porous membrane filter. Retina. 2003;23:777–9.
BBG, ICG, and TA compared in ILM peeling 11. Isomae T, Sato Y, Shimada H. Shortening the duration of prone positioning after macular hole surgery: comparison between 1-week and 1-day prone positioning. Jpn J Ophthalmol. 2002;46:84–8. 12. Ullrich S, Haritoglou C, Gass C, Schaumberger M, Ulbig MW, Kampik A. Macular hole size as a prognostic factor in macular hole surgery. Br J Ophthalmol. 2002;86:390–3. 13. Ip MS, Baker BJ, Duker JS, Reichel E, Baumal CR, Gangnon R, et al. Anatomical outcomes of surgery for idiopathic macular hole as determined by optical coherence tomography. Arch Ophthalmol. 2002;120:29–35. 14. Tsuiki E, Fujikawa A, Miyamura N, Ogawa T, Kitaoka T. Three cases of nasal visual field defect following indocyanine greenassisted peeling of internal limiting membrane [in Japanese]. Jpn J Clin Ophthalmol. 2004;58:687–90. 15. Hida T, Chandler D, Arena JE, Machemer R. Experimental and clinical observations of the intraocular toxicity of commercial corticosteroid preparation. Am J Ophthalmol. 1986;101:190–5. 16. Roth DB, Chieh J, Spirn MJ, Green SN, Yarian DL, Chaudhry NA. Noninfectious endophthalmitis associated with intravitreal triamcinolone injection. Arch Ophthalmol. 2003;121:1279–82. 17. Kimura H, Kuroda S, Nagata M. Triamcinolone acetonideassisted peeling of the internal limiting membrane. Am J Ophthalmol. 2004;137:172–3. 18. Horio N, Horiguchi M, Yamamoto N. Triamcinolone-assisted internal limiting membrane peeling during idiopathic macular hole surgery. Arch Ophthalmol. 2005;123:96–9. 19. Kuramoto N, Ohashi H, Ozaki S. Deposition of triamcinolone acetonide in the macular hole after vitreous surgery [in Japanese]. Jpn J Clin Ophthalmol. 2006;60:853–6.
461 20. Ejstrup R, Cour MR, Heegaard S, Kiilgaard JF. Toxicity profiles of subretinal indocyanine green, brilliant blue G, and triamcinolone acetonide: a comparative study. Graefes Arch Clin Exp Ophthalmol. 2012;250:669–77. 21. Ricarda GS, Arnd G, Siegfried GP, Anselm K, Christos H. Vital dyes for macular surgery: a comparative electron microscopy study of the internal limiting membrane. Retina. 2009;29:669–76. 22. Fukuda K, Shiraga F, Yamaji H, Nomoto H, Shiragami C, Enaida H, et al. Morphologic and functional advantages of macular hole surgery with brilliant blue G-assisted internal limiting membrane peeling. Retina. 2011;31:1720–5. 23. Baba T, Hagiwara A, Sato E, Arai M, Oshitari T, Yamamoto S. Comparison of vitrectomy with brilliant blue G or indocyanine green on retinal microstructure and function of eyes with macular hole. Ophthalmology. 2012;119:2609–15. 24. Jiang LH, Mackenzie AB, North RA, Surprenant A. Brilliant blue G selectively blocks ATP-gated rat P2X(7) receptors. Mol Pharmacol. 2000;58:82–8. 25. Notomi S, Hisatomi T, Kanemaru T, Takeda A, Ikeda Y, Enaida H, et al. Critical involvement of extracellular ATP acting on P2RX7 purinergic receptors in photoreceptor cell death. Am J Pathol. 2011;179:2798–809. 26. Notomi S, Hisatomi T, Murakami Y, Terasaki H, Sonoda S, Asato R, et al. Dynamic increase in extracellular ATP accelerates photoreceptor cell apoptosis via ligation of P2RX7 in subretinal hemorrhage. PLoS ONE. 2013;8:e53338. doi:10.1371/journal. pone.0053338.
123
CLINICAL INVESTIGATION
Long-term outcomes of 3 surgical adjuvants used for internal limiting membrane peeling in idiopathic macular hole surgery Noriko Mochizuki • Teiko Yamamoto Hiroshi Enaida • Tatsuro Ishibashi • Hidetoshi Yamashita
•
Received: 26 November 2013 / Accepted: 11 July 2014 / Published online: 9 September 2014 Ó Japanese Ophthalmological Society 2014
Abstract Purpose Indocyanine green (ICG), an adjuvant used for peeling of the internal limiting membrane (ILM) during vitreous surgery for idiopathic macular hole (MH), has been reported to be toxic, possibly affecting postoperative visual acuity. We compared the long-term outcomes (within 2 years) of brilliant blue G (BBG), ICG, and triamcinolone acetonide (TA). Patients and methods This study involved 97 eyes of 94 patients who underwent vitreous surgery for MH at the Yamagata University Hospital between June 2002 and November 2010. The surgical adjuvants used were BBG for 15 eyes, ICG for 61 eyes, and TA for 21 eyes. We compared the postoperative visual acuities, initial closure rates, final closure rates, and complications of the 3 groups. Results In all 3 groups, the visual acuity significantly improved after surgery. The magnitude of the improvement at 2 years after surgery was significantly better in the BBG group than in the ICG group (Mann–Whitney test, P = 0.020). The postoperative visual acuity did not
N. Mochizuki (&) H. Yamashita Department of Ophthalmology and Visual Sciences, Yamagata University Faculty of Medicine, 2-2-2 Iida-Nishi, Yamagata, Yamagata 990-9585, Japan e-mail: [email protected] T. Yamamoto Kanamecho Yamamoto Eye Clinic, Tokyo, Japan H. Enaida Department of Ophthalmology, Faculty of Medicine, Saga University, Saga, Japan T. Ishibashi Department of Ophthalmology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
significantly differ between the BBG and TA groups (P = 0.627) or between the ICG and TA groups (P =0 .137). Thus, the surgery using BBG resulted in a significantly better outcome in visual acuity than did the surgery using ICG. The 3 groups did not differ in initial or final closure rates or in incidence of complications. Conclusion Analysis of the long-term outcomes of vitreous surgeries provided evidence that BBG is a useful adjuvant for ILM peeling. Keywords Brilliant blue G Indocyanine green Internal limiting membrane peeling Macular hole Triamcinolone acetonide
Introduction In 1991, Kelly and Wendel [1] reported the successful closure of idiopathic macular holes (MH) with vitreous surgery combined with gas tamponade. Since then, this technique has been used in many hospitals and clinics. In earlier procedures, after the posterior vitreous cortex was detached, the air–fluid exchange was performed with an intravitreal injection of sulfur hexafluoride gas (SF6). The success rate in closing the hole with this method was 50 to 80 %. The closure rate was further increased by addition of autologous serum [2] or by peeling of the internal limiting membrane (ILM), the innermost layer of the retina [3]. However, because the ILM is thin and transparent, its removal while still leaving the retinal nerve intact is challenging. Since a report that the closure rate was improved by staining the ILM with indocyanine green (ICG) for better visual identification, ILM peeling has become more reliable [4], leading to further increases in MH closure rates. At present, ICG and triamcinolone
123
456
N. Mochizuki et al.
Table 1 Patients’ backgrounds
BBG
ICG
TA
Number (eyes)
15
61
21
Male
9
31
9
Female
6
30
12
Age, (years)*
68.6 ± 7.4
65.9 ± 8.6
63.2 ± 7.6
0.24
Duration of visual symptoms, (months)*
2.57 ± 2.71
7.34 ± 10.32
7.81 ± 12.64
0.01
MH diameter, (lm)*
923.33 ± 290.75
820.11 ± 303.17
700.71 ± 267.64 \0.001
Stage 2
2 (13.3)
10 (16.4)
9 (42.9)
Stage 3
5 (33.3)
25 (41.0)
5 (23.8)
Stage 4
8 (53.3)
26 (42.6)
7 (33.3)
0.86 ± 0.22 12 (80.0)/3 (20.0)
0.84 ± 0.31 60 (98.4)/1 (1.6)
0.94 ± 0.30 21 (100)/0 (0)
MH stage, n (%)
BBG brilliant blue G, ICG indocyanine green, TA triamcinolone acetonide, MH macular hole * Mean ± SD
Materials and methods Study design and patients This was a retrospective cohort study involving 97 eyes of 94 patients who underwent vitreous surgery for idiopathic MH and were followed up for over 3 months at Yamagata University Hospital between June 2002 and November 2010. The patients comprised 48 men (49 affected eyes) and 46 women (48 affected eyes). They were divided into 3 groups according to the dye used for ILM peeling: the BBG group (15 eyes), the ICG group (61 eyes), and the TA group (21 eyes). No significant differences were detected between any 2 of these 3 groups in terms of the male-tofemale ratio (Fisher exact test, P = 0.62), age (Kruskal– Wallis test, P = 0.24), MH stage (2 to 4) distribution (Fisher exact test, P = 0.14), or preoperative logarithm of
123
0.62
0.14
Preoperative visual acuity, logMAR* Phakic eyes, n (%)/Pseudophakic eyes, n (%)
acetonide (TA) are used for ILM peeling. ICG, however, has been reported to induce the formation of reactive oxygen species when exposed to light, leading to injury of the retinal pigment epithelial cells and decreased retinal function [5]. ICG has also been shown to be histologically toxic, causing loss of retinal pigment epithelial cells [6]. As regards TA, it is not designed to stain the ILM. If TA detaches from the ILM because of water flow or other factors, the ILM flap becomes transparent, possibly making it difficult for the surgeon to identify the flap. To address these limitations of TA, the usefulness of a new dye for staining the ILM, brilliant blue G (BBG), has been widely investigated [7, 8]. In this study, we report the use of BBG as an adjuvant in ILM peeling during MH surgery We also present results of a comparison of the long-term outcomes (within 2 years) of MH surgeries in which BBG was used with those of MH surgeries in which ICG or TA was used.
P
0.42 0.03
Fig. 1 Optical coherence tomography (OCT) of macular hole diameter. The diameter was measured at the base of the hole
the minimum angle of resolution (logMAR) visual acuity (VA) (Kruskal–Wallis test, P = 0.42) (Table 1). The duration of visual symptoms was significantly shorter in the BBG group than in the other 2 groups (Kruskal–Wallis test, P = 0.01). In accordance with a previous report [9], we used optical coherence tomography (OCT) to measure the MH. We measured the largest diameter of the hole, i.e., the diameter at the hole base (Fig. 1). The MH was significantly larger in the BBG group than in the other 2 groups (Kruskal–Wallis test, P \ 0.001). The proportion of phakic eyes before surgery was significantly lower in the BBG group than in the other 2 groups (Fisher exact test, P = 0.03). Surgical technique All patients underwent 3-port pars-plana vitrectomy, mainly with microincisional vitrectomy (25-gauge) under retrobulbar anesthesia. In addition, phacoemulsification and intraocular lens implantation (cataract surgery) were
BBG, ICG, and TA compared in ILM peeling
457
adjusted to complete the surgery. In the BBG group, the fluid was substituted with 1 ml of 100 % SF6 in all eyes. In the ICG group, the fluid was substituted with air in 29 eyes, 20 % SF6 (total substitution) in 9 eyes, 1 ml of 100 % SF6 in 22 eyes, and 12 % C3F8 (total substitution) in 1 eye. In the TA group, the fluid was substituted with air in 14 eyes and with 1 ml of 100 % SF6 in 7 eyes. Although recent studies have shown that the duration of prone positioning time after MH surgery was shortened [11], we instructed the patients to maintain the prone position until closure of the MH was confirmed. This study was approved by the institutional review board of Yamagata University and performed in accordance with the ethical standards of the 1989 Declaration of Helsinki. Fig. 2 Brilliant blue G (BBG)-assisted internal limiting membrane (ILM) peeling for macular hole (MH). After ILM removal, a difference is clearly visible in the retinal surface color between the area from which the ILM was removed and the surrounding area
performed for 5 of the 15 eyes (33.3 %) in the BBG group, 36 of the 61 eyes (59.0 %) in the ICG group, and 10 of the 21 eyes (47.6 %) in the TA group. For stage 2 and 3 cases, TA (Kenacort-A; Bristol-Myers Squibb, New York, NY, USA) was used for posterior vitreous separation. The supernatant was removed using the method reported by Nishimura et al. [10], which involves a liquid filter and a balanced salt solution (BSS) (BSS Plus; Alcon, Fort Worth, TX, USA). Then, a suspension of TA (8 mg/ml) was prepared with 5 ml of BSS. ILM peeling was performed as follows. In the ICG group, ICG was used at a concentration of 0.125 %. The prepared ICG solution was injected by spraying over the posterior pole while avoiding the MH. This was immediately followed by washing with BSS. After confirmation that the ILM was stained green, a scaffold was created on the ILM with a V-lance. ILM forceps were then used to peel the ILM for approximately 3 disc diameters around the MH. In the TA group, TA (0.2–0.3 ml) was applied thinly to the posterior pole, which was followed by the ILM peeling. For the BBG group, BBG was made into a 0.25 mg/ml solution with BSS. It was passed through a filter and injected by spraying over the posterior pole, while avoiding the hole. This was followed immediately by washing with BSS. After confirmation that the ILM was stained blue, the ILM was peeled in the same way as that described for the ICG group (Fig. 2). After the ILM removal, air–fluid exchange was performed. If needed, the fluid was substituted with gas. The wound was then closed, and the intraocular tension was
Methods of evaluation The surgery outcomes of the different adjuvants were compared by determining the VA and confirming the closure rates by means of OCT. In the cases in which BBG was used, the presence/absence of atrophy of the retinal pigment epithelium was checked by fluorescein angiography (FA) 6 months after the surgery. Statistical analysis Decimal VA was converted into logMAR units. The logMAR units before surgery and at 3 months, 6 months, 1 year, 1.5 years, and 2 years after surgery were compared between 2 groups and analyzed using the Wilcoxon signedrank test. The initial and final closure rates and the presence/absence of complications were tested using the Fisher exact test. Multivariate regression analysis was conducted using the available VA data of 69 eyes (BBG: 14 eyes; ICG: 42 eyes; TA: 13 eyes) until 2 years after surgery. The magnitude of VA improvement at 2 years after surgery (logMAR VA at 2 years minus preoperative logMAR VA) was the dependent variable; the independent variables were sex, age, duration of visual symptoms, MH diameter, MH stage, preoperative logMAR VA, tamponade agents, cataract extraction during MH surgery, cataract extraction after MH surgery, timing of cataract extraction (pseudophakia preoperatively, cataract extraction during MH surgery, cataract extraction after MH surgery, phakia during follow-up), and adjuvants. The magnitude of VA improvement at 2 years after surgery was compared among the 3 groups using the Kruskal–Wallis test, and then between 2 groups (BBG vs TA, BBG vs ICG, or TA vs ICG) using the Mann–Whitney test. All statistical analyses were conducted with PASW Statistics 18 software (IBM Corporation, Armonk, NY,
123
458
USA). Probability values of less than 0.05 were considered significant.
Results The intensity of the chromatic response of the ILM to BBG (0.25 mg/ml) was comparable to that of the ILM response to 0.125 % ICG. Once peeling of the ILM began in the BBG group, the peeled area became clearly distinguishable from the remaining area, allowing the subsequent peeling steps to be performed smoothly (Fig. 2). The area affected by the epimacular membrane was not positively stained. Figure 3 shows the time course of VA changes in each adjuvant group. In the BBG group, the mean logMAR VA before surgery was 0.86 ± 0.22 (n = 15), and after surgery 0.40 ± 0.25 (n = 15) at 3 months, 0.37 ± 0.20 (n = 15) at 6 months, 0.33 ± 0.20 (n = 15) at 1 year, 0.21 ± 0.18 (n = 14) at 1.5 years, and 0.16 ± 0.14 (n = 14) at 2 years. A significant improvement in VA relative to the preoperative acuity was noted at 3 months after surgery (Wilcoxon signed-rank test, P \ 0.001), and this improved acuity was maintained for 2 years (P \ 0.001). In the ICG group, the mean logMAR VA before surgery was 0.84 ± 0.31 (n = 61), and after surgery 0.59 ± 0.36 (n = 61) at 3 months, 0.57 ± 0.36 (n = 61) at 6 months, 0.44 ± 0.35 (n = 60) at 1 year, 0.37 ± 0.30 (n = 53) at 1.5 years, and 0.35 ± 0.28 (n = 42) at 2 years. A significant improvement in VA relative to the preoperative acuity was noted 3 months after surgery (P \ 0.001), and this improvement was maintained for 2 years (P \ 0.001). In the TA group, the mean logMAR VA before surgery was 0.94 ± 0.30 (n = 21), and after surgery 0.44 ± 0.34 (n = 21) at
Fig. 3 Visual acuity changes after vitrectomy in the brilliant blue G (BBG), indocyanine green (ICG), and triamcinolone acetonide (TA) groups. The circles represent the BBG group; the squares, the ICG group; and the triangles, the TA group. In all 3 groups, the visual acuity improves significantly after surgery. (*Wilcoxon signed-rank test, P \ 0.05)
123
N. Mochizuki et al.
3 months, 0.41 ± 0.37 (n = 21) at 6 months, 0.33 ± 0.34 (n = 21) at 1 year, 0.30 ± 0.29 (n = 17) at 1.5 years, and 0.28 ± 0.30 (n = 13) at 2 years. A significant improvement in VA relative to the preoperative acuity was noted 3 months after the surgery (P \ 0.001), and this improvement was maintained for 2 years (P \ 0.001). The magnitude of VA improvement at 2 years after the surgery was 0.69 ± 0.22 in the BBG group, 0.49 ± 0.26 in the ICG group, and 0.66 ± 0.29 in the TA group. The initial closure rate was 86.7 % (n = 15) in the BBG group, 86.9 % (n = 61) in the ICG group, and 90.5 % (n = 21) in the TA group; no significant intergroup difference was detected (Fisher exact test, P = 1.00). In 1 case of the ICG group, because the patient did not agree to further surgery, the MH could not be closed. In all of the other cases, the MH was closed. No significant differences in the final closure rates were detected between any 2 of the BBG group (100 %, n = 15), ICG group (98.4 %, n = 61), or TA group (100 %, n = 21) (Fisher exact test, P = 1.00). With respect to the intraoperative complications, a retinal tear in the peripheral retina occurred in 2 eyes of the BBG group, 11 eyes of the ICG group, and 2 eyes of the TA group. In these cases, sufficient vitreous dissection was conducted intraoperatively, and laser photocoagulation was applied to surround the break completely. No significant differences were found in terms of the presence/absence of retinal tear development between any 2 of the 3 groups (Fisher exact test, P = 0.67). After surgery, no patient from any group developed retinal detachment, intraocular hypotension, intraocular hypertension, or endophthalmitis. Reopening of the MH occurred in 1 eye (1.6 %) of the ICG group. The probabilities of MH reopening in the 3 groups did not differ significantly (Fisher exact test, P = 1.00). During the 2 years of followup, cataract surgery was performed in 4 of the 15 eyes (26.7 %) of the BBG group, 9 of the 61 eyes (14.8 %) of the ICG group, and 3 of the 21 eyes (14.3 %) of the TA group (Fisher exact test, P = 0.51). Postoperative FA in the BBG group revealed no abnormalities, such as atrophy of the retinal pigment epithelium or leakage of the dye. The preoperative logMAR VA was found to correlate significantly with the postoperative VA (P \ 0.001; standardized partial regression coefficient = 0.63; Table 2). Sex, age, duration of the visual symptoms, MH diameter, MH stage, preoperative logMAR VA, tamponade agents, timing of the cataract extraction, and adjuvants were not significantly correlated with the postoperative VA. A comparison of the magnitude of VA improvement at 2 years after surgery among the 3 groups indicated significant intergroup differences (Kruskal–Wallis test, P = 0.04). The probability values as determined by the Mann–Whitney test for the differences in VA were as follows: P = 0.02 between the BBG and ICG groups,
BBG, ICG, and TA compared in ILM peeling
459
Table 2 Correlation of the independent variables with postoperative visual acuity Standardized partial regression coefficient
P value
Sex
0.178
0.090
Age
-0.176
0.172
Duration of visual symptoms
-0.190
0.105
MH diameter
-0.071
0.555
MH stage
-0.077
Preoperative visual acuity, logMAR Tamponade agent
0.634
0.483 \0.001*
0.034
0.783
Timing of cataract extraction
-0.155
0.221
Adjuvant
-0.088
0.522
* Multiple linear regression analysis, P \ 0.05
P = 0.63 between the BBG and TA groups, and P = 0.14 between the ICG and TA groups. Thus, the use of BBG resulted in a significantly better time course of VA improvement when compared with the ICG group, whereas the time courses of VA improvement did not significantly differ between the BBG and TA groups or between the ICG and TA groups.
Discussion The diameter and duration of symptoms of MH have been listed as factors affecting the postoperative VA [12, 13]. In this study, the MH diameter, length of time until the surgery, and proportion of phakic eyes before surgery differed among the individual patients. To evaluate the influence of these factors, we performed multivariate regression analysis of the available VA data of 69 eyes for 2 years after surgery. The preoperative VA was found to correlate significantly with the postoperative VA. The longer the period between diagnosis and surgery or the larger the MH diameter became, the poorer the improvement in postoperative VA was. However, the influence of these factors was not significant, suggesting that they do not affect the evaluation of the usefulness of individual adjuvants. Adjuvants conventionally used for ILM peeling include ICG and TA. However, as regards ICG, culture of human retinal pigment epithelial cells in the presence of ICG coupled with light exposure was reported to decrease the mitochondrial activity associated with cell survival rate [5]. In addition, cases of visual field defects after ILM staining with ICG were attributed to the influence of light [14]. In rat eyes, ICG induced histologic disorders, including destruction of the retinal lamellar structure and loss of
retinal pigment epithelial cells [6]. The toxicity of ICG has also been reported to be concentration-dependent. In view of these findings, countermeasures to prevent these disorders, such as avoidance of the use of concentrated ICG and of prolonged retinal exposure to light, appear to be essential to ILM peeling. TA is a steroid that is insoluble in water. The Kenacort A solution is slightly acidic (pH 5–7) and contains preservatives, such as benzyl alcohol. Such preservatives can affect the retina and lens [15]. For this reason, TA is usually used after removal of as much of the solvent as possible and subsequent dissolving in BSS. Complications arising from the use of TA include increased intraocular tension, cataract progression, and onset of aseptic [16] or bacterial endophthalmitis. TA has been used for the treatment of various retinal and vitreous diseases. Its use in ILM peeling during surgery for MH was first reported by Kimura et al. [17] and then by Horio et al. [18]. TA remains in the MH floor for 2 months or longer after surgery, but did not result in any particularly hazardous effects [19]. However, the surgeon occasionally cannot identify the stump of the freed ILM if the TA particles are moved away from the membrane surface by the aqueous humor flow or other factors. The present study was designed to compare the safety and effectiveness of 3 adjuvants, BBG, ICG, and TA, in terms of VA, MH closure rates, intraoperative/postoperative complications, intensity of chromatic responses, simplicity of ILM peeling, and histologic and functional toxicities. In all groups, the VA was significantly improved at 3 months after surgery, and the improvement was maintained for 2 years. This improvement was more marked in the BBG group than in the ICG group. The ICG group, however, included patients whose MH did not close until the end of the study and in whom the MH recurred. The patients with nonclosure of the MH were followed up for 1 year, at which time the final VA was 0.05. In the patients with recurrent MH, the MH recurred after a 1.5year follow-up, but the final VA was determined to be 0.9 just before the recurrence. These MH may have affected the VA outcomes. In pig eyes, histopathologic and electrophysiologic evaluations 6 weeks after subretinal injection of ICG, BBG, or TA indicated that ICG evidently injures the retina both histologically and functionally, while BBG and TA do not [20], which is consistent with the results of this study. In that previous study, for patients with an MH or a macular pucker, ILM sections peeled off after staining with trypan blue, BBG, bromophenol blue, Chicago blue, or ICG were morphologically compared under a transmission electron microscope. The cells on the retinal side of the ILM stained with ICG were evidently larger in size and number than those of the ILM stained with any other dye,
123
460
indicating that the retina undergoes more marked morphologic changes when stained with ICG than when stained with BBG [21]. When the percentages of IS/OS junction restorations after MH surgery were compared between patients in whom BBG was used for ILM dissection and patients in whom ICG was used, the percentage of the IS/ OS restorations was significantly higher in the BBG group than in the ICG group. The percentage of patients with final corrected VA of 20/20 or more was higher in the BBG group than in the ICG group. In accordance with these observations, it has been reported that early IS/OS junction restoration affects the prognosis of the postoperative VA [22]. Baba et al. [23] compared the best-corrected VA (BCVA) of eyes in which BBG had been used for ILM peeling and those in which ICG had been used for MH care and found that the average BCVA was significantly better in the BBG group than in the ICG group at 3 and 6 months after surgery. These results are similar to those of the present study. Driven by the recent characterization of BBG as a P2RX7 antagonist [24], Notomi et al. [25] reported that photoreceptor apoptosis involves P2RX7 activation with cleavage of caspase-8 and -9 and mitochondrial-nuclear translocation of apoptosis-inducing factor (AIF). Furthermore, photoreceptor apoptosis can be attenuated by BBG, a pharmacologic P2RX7 antagonist that acts by blocking the interaction between extracellular ATP and P2RX7. Pharmacologic inhibition of P2RX7 has been reported to result possibly in neuroprotection of photoreceptors in cases of subretinal hemorrhage [26]. These previous reports support the present study’s observations that the long-term prognosis of VA was better in the BBG group than in the ICG group. In this study, the closure rates and intraoperative/postoperative complications did not differ among the 3 groups. The intensity of the ILM responses to BBG at a concentration of 0.25 mg/ml was comparable to that of the ILM responses to 0.125 % ICG, and the membrane was stained well with both dyes. In the TA group, the TA was sometimes washed away during the ILM peeling, requiring additional application of TA. In terms of the simplicity of ILM peeling, BBG seems to be equivalent to ICG, and TA, slightly inferior. To date, no reports of the histologic toxicity of BBG have been published. Reports are available on histologic disorders induced by ICG in cultured human retinal pigment epithelial cells and in rat eyes [5, 6]. According to the data collected here during the long-term (2-year) follow-up, the toxicity of BBG was lower than the toxicity that has been commonly noted for ICG. When the magnitude of the VA improvement at 2 years after surgery was compared among the 3 groups, BBG was the best adjuvant, followed by TA, and then by ICG. Furthermore, BBG allowed for easier ILM peeling than did TA.
123
N. Mochizuki et al.
This was a retrospective study. Some factors in the patients’ backgrounds differed among the 3 adjuvant groups. Our sample size was also relatively limited. Despite these limitations, to the best of our knowledge, this study has demonstrated for the first time that BBG provides a safe modality for staining the ILM and that the intensity of the chromatic response to BBG was not inferior to that of the other 2 dyes. We may therefore conclude that BBG is useful as an adjuvant during ILM peeling. Acknowledgments This work was supported in part by a grant from the Japan Medical Association and by a Grant-in-Aid for Scientific Research (#24592673) from the Japanese Ministry of Education, Culture, Sports, Science and Technology. Conflicts of interest N. Mochizuki, None; T. Yamamoto, None; H. Enaida, Patent (Kyushu Univ.); T. Ishibashi, Patent (Kyushu Univ.); H. Yamashita, None. BBG/Hiroshi Enaida and Tatsuro Ishibashi are the inventors of patent application ‘‘Staining composition for staining an ophthalmic membrane’’ US7731941/EP1819366/4200222JP. Kyushu University holds the patent. Kyushu University licensed the patent of BBG to the bio-venture company, Aqumen Biopharmaceuticals NA, Inc. (Fukuoka, Japan). Eventually, the license was granted to the Dutch Ophthalmic Research Center International BV (D.O.R.C.) from Aqumen.
References 1. Kelly NE, Wendel RT. Vitreous surgery for idiopathic macular holes: results of a pilot study. Arch Ophthalmol. 1991;109:654–9. 2. Kusaka S, Sakagami K, Kutsuna M, Ohashi Y. Treatment of fullthickness macular holes with autologous serum. Jpn J Ophthalmol. 1997;41:332–8. 3. Brooks HL. ILM peeling in full thickness macular hole surgery. Vitreoretinal Surg Technol. 1995;7:2. 4. Kadonosono K, Itoh N, Uchio E, Nakamura S, Ohno S. Staining of internal limiting membrane in macular hole surgery. Arch Ophthalmol. 2000;118:1116–8. 5. Sippy BD, Engelbrecht NE, Hubbard GB, Moriarty SE, Jiang S, Aaberg TM Jr, et al. Indocyanine green effect on cultured human retinal pigment epithelial cells: implication for macular hole surgery. Am J Ophthalmol. 2001;132:433–5. 6. Enaida H, Sakamoto T, Hisatomi T, Goto Y, Ishibashi T. Morphological and functional damage of the retina caused by intravitreous indocyanine green in rat eyes. Graefes Arch Clin Exp Ophthalmol. 2002;240:209–13. 7. Enaida H, Hisatomi T, Goto Y, Hata Y, Ueno A, Miura M, et al. Preclinical investigation of internal limiting membrane staining and peeling using intravitreal brilliant blue G. Retina. 2006;26:623–30. 8. Enaida H, Hisatomi T, Hata Y, Ueno A, Goto Y, Yamada T, et al. Brilliant blue G selectively stains the internal limiting membrane/ brilliant blue G-assisted membrane peeling. Retina. 2006;26:631–6. 9. Matsumiya W, Kusuhara S, Shimoyama T, Honda S, Tsukahara Y, Negi A. Predictive value of preoperative optical coherence tomography for visual outcome following macular hole surgery: effects of imaging alignment. Jpn J Ophthalmol. 2013;57:308–15. 10. Nishimura A, Kobayashi A, Segawa Y, Sakurai M, Shirao E, Shirao Y, et al. Isolating triamcinolone acetonide particles for intravitreal use with a porous membrane filter. Retina. 2003;23:777–9.
BBG, ICG, and TA compared in ILM peeling 11. Isomae T, Sato Y, Shimada H. Shortening the duration of prone positioning after macular hole surgery: comparison between 1-week and 1-day prone positioning. Jpn J Ophthalmol. 2002;46:84–8. 12. Ullrich S, Haritoglou C, Gass C, Schaumberger M, Ulbig MW, Kampik A. Macular hole size as a prognostic factor in macular hole surgery. Br J Ophthalmol. 2002;86:390–3. 13. Ip MS, Baker BJ, Duker JS, Reichel E, Baumal CR, Gangnon R, et al. Anatomical outcomes of surgery for idiopathic macular hole as determined by optical coherence tomography. Arch Ophthalmol. 2002;120:29–35. 14. Tsuiki E, Fujikawa A, Miyamura N, Ogawa T, Kitaoka T. Three cases of nasal visual field defect following indocyanine greenassisted peeling of internal limiting membrane [in Japanese]. Jpn J Clin Ophthalmol. 2004;58:687–90. 15. Hida T, Chandler D, Arena JE, Machemer R. Experimental and clinical observations of the intraocular toxicity of commercial corticosteroid preparation. Am J Ophthalmol. 1986;101:190–5. 16. Roth DB, Chieh J, Spirn MJ, Green SN, Yarian DL, Chaudhry NA. Noninfectious endophthalmitis associated with intravitreal triamcinolone injection. Arch Ophthalmol. 2003;121:1279–82. 17. Kimura H, Kuroda S, Nagata M. Triamcinolone acetonideassisted peeling of the internal limiting membrane. Am J Ophthalmol. 2004;137:172–3. 18. Horio N, Horiguchi M, Yamamoto N. Triamcinolone-assisted internal limiting membrane peeling during idiopathic macular hole surgery. Arch Ophthalmol. 2005;123:96–9. 19. Kuramoto N, Ohashi H, Ozaki S. Deposition of triamcinolone acetonide in the macular hole after vitreous surgery [in Japanese]. Jpn J Clin Ophthalmol. 2006;60:853–6.
461 20. Ejstrup R, Cour MR, Heegaard S, Kiilgaard JF. Toxicity profiles of subretinal indocyanine green, brilliant blue G, and triamcinolone acetonide: a comparative study. Graefes Arch Clin Exp Ophthalmol. 2012;250:669–77. 21. Ricarda GS, Arnd G, Siegfried GP, Anselm K, Christos H. Vital dyes for macular surgery: a comparative electron microscopy study of the internal limiting membrane. Retina. 2009;29:669–76. 22. Fukuda K, Shiraga F, Yamaji H, Nomoto H, Shiragami C, Enaida H, et al. Morphologic and functional advantages of macular hole surgery with brilliant blue G-assisted internal limiting membrane peeling. Retina. 2011;31:1720–5. 23. Baba T, Hagiwara A, Sato E, Arai M, Oshitari T, Yamamoto S. Comparison of vitrectomy with brilliant blue G or indocyanine green on retinal microstructure and function of eyes with macular hole. Ophthalmology. 2012;119:2609–15. 24. Jiang LH, Mackenzie AB, North RA, Surprenant A. Brilliant blue G selectively blocks ATP-gated rat P2X(7) receptors. Mol Pharmacol. 2000;58:82–8. 25. Notomi S, Hisatomi T, Kanemaru T, Takeda A, Ikeda Y, Enaida H, et al. Critical involvement of extracellular ATP acting on P2RX7 purinergic receptors in photoreceptor cell death. Am J Pathol. 2011;179:2798–809. 26. Notomi S, Hisatomi T, Murakami Y, Terasaki H, Sonoda S, Asato R, et al. Dynamic increase in extracellular ATP accelerates photoreceptor cell apoptosis via ligation of P2RX7 in subretinal hemorrhage. PLoS ONE. 2013;8:e53338. doi:10.1371/journal. pone.0053338.
123
