Combination Therapy With Dexamethasone Intravitreal Implant and Macular Grid Laser in Patients With Branch Retinal Vein Occlusion FRANCESCO PICHI, CLAUDIA SPECCHIA, LUCIA VITALE, ANDREA LEMBO, MARIACHIARA MORARA, CHIARA VERONESE, ANTONIO P. CIARDELLA, AND PAOLO NUCCI PURPOSE: To test a combination of dexamethasone intravitreal implant with macular grid laser for macular edema in patients with branch retinal vein occlusion (BRVO). DESIGN: Prospective interventional, randomized, multicenter study. METHODS: Patients with macular edema secondary to BRVO underwent an Ozurdex intravitreal implant at baseline. After 1 month, patients were randomly assigned to 2 study groups. Patients in Group 1 were followed up monthly and retreated with Ozurdex implant whenever there was a recurrence of macular edema or a decrease in best-corrected visual acuity (BCVA). In Group 2 patients macular grid laser was performed between weeks 6 and 8. After that, patients were followed up and retreated as for Group 1. RESULTS: In Group 1 at 4 months, mean BCVA was 0.49 ± 0.35 logMAR and central retinal thickness (CRT) was 391 ± 172 mm; both improved significantly at 6 months, to 0.32 ± 0.29 logMAR and 322 ± 160 mm, respectively. In Group 2, CRT was reduced significantly to 291 ± 76 mm at 4 months, and BCVA improved to 0.25 ± 0.20 logMAR. At the final visit, BCVA was 0.18 ± 0.14 logMAR and mean CRT was 271 ± 44 mm. The number of Ozurdex implants at 4 months was 12 of 25 (48%) in Group 1 patients vs 3 of 25 (12%) in Group 2 patients (P [ .012). At 6 months 3 of 25 patients (12%) in Group 1 vs 0 of 25 (0%) in Group 2 (P [ .23) were retreated. CONCLUSIONS: The combination of Ozurdex implant and macular grid laser is synergistic in increasing BCVA and lengthening the time between injections. (Am J Ophthalmol 2014;157:607–615. Ó 2014 by Elsevier Inc. All rights reserved.)
I
NTRAVITREAL
PHARMACOTHERAPY
FOR
MACULAR
edema induced by branch retinal vein occlusion (BRVO) has benefited from the recent development of
Accepted for publication Nov 13, 2013. From San Giuseppe Hospital, University Eye Clinic, Milan, Italy (F.P., L.V., A.L., P.N.); Department of Molecular and Translational Medicine, University of Brescia, Brescia, Italy (C.S.); and Sant’Orsola-Malpighi Hospital, Ophthalmology Unit, Bologna, Italy (M.M., C.V., A.P.C.). Inquiries to Francesco Pichi, San Giuseppe Hospital – University Eye Clinic, via San Vittore 12, 20123, Milan, Italy; e-mail: ilmiticopicchio@ gmail.com 0002-9394/$36.00 http://dx.doi.org/10.1016/j.ajo.2013.11.016
Ó
2014 BY
a new treatment: intravitreal implant of slow-release dexamethasone (Ozurdex; Allergan, Irvine, California, USA).1,2 Not only has the use of this product effectively improved the anatomic aspect of the retina by reducing the macular thickness and even in the best cases allowing the recovery of a normal foveal profile, but chiefly it has improved patients’ vision quickly and sustainably. Even if the effect of intravitreal steroidal agents is rapid, we often see a recurrence of the macular edema 3-6 months after complete resolution brought about by the implant. Consequently, repeated injections usually tend to be necessary. The mean number of Ozurdex injections reported in the literature is 4.2 6 0.4 injections per year, and more than 6 consecutive injections are sometimes reported in the reference series.3,4 Laser grid photocoagulation is an evidence-based treatment of the macular edema associated with BRVO. Since publication of the Branch Vein Occlusion Study Group5 results in 1984, in which investigators found a significant mean improvement of 1.33 lines of vision compared with 0.23 line in controls after 3 years of follow-up with level of evidence A, grid laser photocoagulation has been considered the gold standard for treating macular edema in patients with BRVO. As such, macular grid laser has been employed in the standard care arm group of the Standard Care Versus Corticosteroid for Retinal Vein Occlusion (SCORE)-BRVO study, in which the percentage of participants with gain in visual acuity letter score of 15 or more from baseline to month 12 was similar when compared to the 1 mg and 4 mg triamcinolone groups (28.9, 25.6, and 27.2, respectively).2 Based on these reports, grid photocoagulation does have long-term effects on the reduction of macular edema attributable to BRVO, but it is sometimes difficult to perform effective treatment coagulation in eyes that have severe macular edema, and it is often necessary to increase the laser power, increasing the side effects, such as scar expansion, subretinal fibrosis, outer retinal atrophy, and choroidal neovascularization. Because clinicians try to incorporate these studies into clinical practice, it may be possible to combine several treatments to achieve a synergistic effect and decrease the frequency of injection into the eye. The purpose of this prospective interventional study was to determine if a sustained-release corticosteroid injection with macular grid laser can be synergistic, with the hypothesis that an injection of Ozurdex decreases the macular edema,
ELSEVIER INC. ALL
RIGHTS RESERVED.
607
allowing effective laser uptake. The hypothesis is that this combination can provide further improvements in both visual acuity and macular thickness during a 6-month trial, thereby minimizing the number of injections and increasing the duration of effects in BRVO patients with macular edema, in a way that would emulate a physician in a typical office setting.
METHODS FROM SEPTEMBER 1, 2010 TO DECEMBER 31, 2012, ALL CONSEC-
utive patients with macular edema secondary to BRVO were considered for enrollment in this prospective interventional, randomized, multicenter study. The Institutional Review Board (IRB) of the University Eye Clinic of Milan and the IRB of the Ophthalmology Unit of Bologna approved this prospective clinical study on June 15, 2010, and the study was registered at the Ethics Committee of the Multimedica Group–San Giuseppe Hospital, as the coordinating center, under registration number OPH-12-10; the study adhered to the Health Insurance Portability and Accountability Act. All participants signed an informed consent before participation. The study conformed to the tenets of the Declaration of Helsinki. STUDY POPULATION:
Patients were naı¨ve to treatment and showed a macular-involved BRVO with decreased visual acuity and perfused macular edema for at least 3 months, with baseline central retinal thickness (CRT) >300 mm. Exclusion criteria were foveal hemorrhages that had not disappeared after 3 months of observation and ischemic maculopathy detected by fluorescein angiography. Patients with a history of ocular surgery and/or rubeotic or advanced glaucoma, defined as cup-to-disc ratio of 0.8 or worse, were excluded from participating. Patients who had had any previous treatment except for oral medication (eg, aspirin 100 mg) were excluded from this study, as well as patients in whom the underlying cause of edema was suspected to be different from BRVO (eg, diabetes, vitreomacular traction). In bilateral cases, only 1 eye per patient was enrolled in the study.
PRIMARY THERAPY OF DEXAMETHASONE INTRAVITREAL IMPLANT: All patients enrolled in this study under-
went a dexamethasone intravitreal implant 0.7 mg as a primary treatment at their first visit to the University Eye Clinic of Milan and to the Ophthalmology Unit of Bologna. After topical anesthesia with 2% lidocaine, the eye was irrigated with 10% povidone-iodine. The subconjunctival space around the anticipated injection site was then injected with 0.1-0.2 mL of 2% lidocaine. This step was followed by the insertion of the Ozurdex implant into the vitreous cavity through the pars plana using a customized, single-use 22 gauge applicator. Patients were 608
treated with a topical ophthalmic antibiotic 4 times daily starting 3 days before the day of their study procedure and continuing for 3 days after the procedure. Consecutive patients were randomly assigned 1:1 to 2 study groups. GROUP 1: After the first Ozurdex implant, patients in Group 1 were followed up monthly with best-corrected visual acuity (BCVA), ophthalmic examination, and spectraldomain optical coherence tomography (SD OCT) (Spectralis HRA OCT; Heidelberg Engineering, Heidelberg, Germany). Patients received retreatment with another Ozurdex implant 0.7 mg whenever there was a recurrence of macular edema. Recurrence was defined as a decrease of BCVA by 0.3 logMAR or as an increase of CRT by more than 50 mm compared to that in the previous examination. GROUP 2:
In Group 2 patients macular grid laser photocoagulation was performed in the macular area between weeks 6 and 8 after the intravitreal implant, to avoid excessive laser burns. Macular grid was performed under topical anesthesia, with a diode-pumped solid-state laser (GYC-1000; Nidek Co, Ltd, Japan) delivering 2-3 rows of 75 mm spots, 80 mW of intensity at 100 mm apart in the parafoveal region, guided by fluorescein angiography and SD OCT, using Goldmann 3 mirror contact lens. Then, 100-150 mm spots were applied 200 mm apart to the remaining areas of retinal thickening and capillary nonperfusion. The major vascular arcade defined the outer borders of the treated area, and the laser spots were aimed so as to avoid the foveal avascular zone and the papillomacular bundle, and applied until soft whitening of the retina became apparent. After the macular grid laser photocoagulation, the patients were followed up in the same manner as Group 1. When a recurrence of macular edema was detected on SD OCT or when visual acuity decreased by 0.3 logMAR, another Ozurdex implant 0.7 mg was given as a retreatment. Complications related to both treatments were recorded to evaluate the safety and adverse events of this protocol. The presence of nuclear, cortical, and posterior subcapsular lens opacities was measured during the slit-lamp examination using standardized photographs. Intraocular pressure (IOP) was also assessed. STUDY OUTCOMES: The primary outcome was final BCVA; secondary outcomes were final CRT by SD OCT, presence of macular edema, number of dexamethasone intravitreal implants, and the occurrence of any complications by the end of the follow-up period. All patients were evaluated at baseline and at every subsequent visit with BCVA, measured with a decimal visual acuity chart at 5 m, then converted to logMAR units. A change of more than 0.3 logMAR units compared to the previous measurement was defined as either an improvement or deterioration in visual acuity. Visual acuity was considered maintained if the logMAR change was less than 0.3 logMAR units compared to the baseline.
AMERICAN JOURNAL OF OPHTHALMOLOGY
MARCH 2014
TABLE 1. Baseline Characteristics of the Study Patients With Perfused Macular Edema Secondary to Branch Retinal Vein Occlusion Who Received a Baseline Dexamethasone Intravitreal Implant and Were Then Randomized to Asneeded Dexamethasone Implant Retreatment (Group 1) or Macular Grid Photocoagulation 6-8 Weeks After Baseline Followed by Retreatment With Dexamethasone Implant on an As-needed Basis (Group 2) Ozurdex Group 1 N ¼ 25
Ozurdex þ Grid Group 2 N ¼ 25
Sex, n (%) Male 11 (44) 11 (44) Female 14 (56) 14 (56) Age (y), median (range) 68 (53-81) 69 (52-78) Eye, n (%) Right 16 (67) 13 (52) Left 9 (33) 12 (48) Best-corrected visual acuity at 0.62 (0.32) 0.53 (0.21) baseline (logMAR), mean (SD) Best-corrected visual acuity 0.38 (0.30) 0.38 (0.17) 1 month after baseline implant (logMAR), mean (SD) Best-corrected visual acuity 0.23 (0.18) 0.15 (0.14) absolute difference from baseline (logMAR), mean (SD) Central retinal thickness at 466 (91) 426 (109) baseline (mm), mean (SD) 303 (67) 322 (51) Central retinal thickness at 1 month after baseline implant (mm), mean (SD) Central retinal thickness change 33.4 (16.9) 21.4 (14.4) from baseline, % (SD)
Slit-lamp examination, indirect fundus examination, IOP measurement, and SD OCT were performed at baseline visit and then repeated at every follow-up visit. The baseline examination included fluorescein angiography to assess the perfusion status of the macula. Patients in both groups were evaluated monthly for 6 months after the randomization to quantify maximum visual improvement at any given time point, and BCVA and CRT data were reported at 4 and 6 months. Central retinal thickness was tomographically defined as the distance between the inner limiting membrane and the retinal pigment epithelium–choriocapillaris interface on radial lines through the foveal area. The fovea was determined using the patient’s fixation and retinal landmarks. All SD OCT scans were performed with a scan rate of 40 000 A-scans per second in a 4.5 3 6.0-mm area. Subsequent follow-up scans were carried out at the exact location of the original scan using image alignment software. This image analysis allowed for an accurate determination of quantitative and qualitative retinal changes during follow-up.6 To verify the accuracy of the SD OCT findings, the fundus area encompassing the fovea was additionally VOL. 157, NO. 3
scanned using several vertical and horizontal scans. Retinal thickness of each SD OCT image section was obtained with the volume scan mode. Parafoveal retinal thickness was determined using the retinal thickness among the 9 subfields of the Early Treatment of Diabetic Retinopathy Study map, which is composed of 3 rings with radii of 1, 3, and 6 mm, respectively. Macular edema was tomographically defined as more than 300 mm of CRT. STATISTICAL ANALYSIS: Continuous variables have been described by mean values and standard deviation (SD). Absolute differences and percent changes at 4 and 6 months from randomization of BCVA and CRT levels were calculated and compared between treatment groups by using t test with Satterthwaite adjustment to allow for unequal variances. Numbers of Ozurdex implantations at 4 and 6 months from randomization were compared between treatment groups by using the Fisher exact test. All reported P values are 2-sided. A P value
I
NTRAVITREAL
PHARMACOTHERAPY
FOR
MACULAR
edema induced by branch retinal vein occlusion (BRVO) has benefited from the recent development of
Accepted for publication Nov 13, 2013. From San Giuseppe Hospital, University Eye Clinic, Milan, Italy (F.P., L.V., A.L., P.N.); Department of Molecular and Translational Medicine, University of Brescia, Brescia, Italy (C.S.); and Sant’Orsola-Malpighi Hospital, Ophthalmology Unit, Bologna, Italy (M.M., C.V., A.P.C.). Inquiries to Francesco Pichi, San Giuseppe Hospital – University Eye Clinic, via San Vittore 12, 20123, Milan, Italy; e-mail: ilmiticopicchio@ gmail.com 0002-9394/$36.00 http://dx.doi.org/10.1016/j.ajo.2013.11.016
Ó
2014 BY
a new treatment: intravitreal implant of slow-release dexamethasone (Ozurdex; Allergan, Irvine, California, USA).1,2 Not only has the use of this product effectively improved the anatomic aspect of the retina by reducing the macular thickness and even in the best cases allowing the recovery of a normal foveal profile, but chiefly it has improved patients’ vision quickly and sustainably. Even if the effect of intravitreal steroidal agents is rapid, we often see a recurrence of the macular edema 3-6 months after complete resolution brought about by the implant. Consequently, repeated injections usually tend to be necessary. The mean number of Ozurdex injections reported in the literature is 4.2 6 0.4 injections per year, and more than 6 consecutive injections are sometimes reported in the reference series.3,4 Laser grid photocoagulation is an evidence-based treatment of the macular edema associated with BRVO. Since publication of the Branch Vein Occlusion Study Group5 results in 1984, in which investigators found a significant mean improvement of 1.33 lines of vision compared with 0.23 line in controls after 3 years of follow-up with level of evidence A, grid laser photocoagulation has been considered the gold standard for treating macular edema in patients with BRVO. As such, macular grid laser has been employed in the standard care arm group of the Standard Care Versus Corticosteroid for Retinal Vein Occlusion (SCORE)-BRVO study, in which the percentage of participants with gain in visual acuity letter score of 15 or more from baseline to month 12 was similar when compared to the 1 mg and 4 mg triamcinolone groups (28.9, 25.6, and 27.2, respectively).2 Based on these reports, grid photocoagulation does have long-term effects on the reduction of macular edema attributable to BRVO, but it is sometimes difficult to perform effective treatment coagulation in eyes that have severe macular edema, and it is often necessary to increase the laser power, increasing the side effects, such as scar expansion, subretinal fibrosis, outer retinal atrophy, and choroidal neovascularization. Because clinicians try to incorporate these studies into clinical practice, it may be possible to combine several treatments to achieve a synergistic effect and decrease the frequency of injection into the eye. The purpose of this prospective interventional study was to determine if a sustained-release corticosteroid injection with macular grid laser can be synergistic, with the hypothesis that an injection of Ozurdex decreases the macular edema,
ELSEVIER INC. ALL
RIGHTS RESERVED.
607
allowing effective laser uptake. The hypothesis is that this combination can provide further improvements in both visual acuity and macular thickness during a 6-month trial, thereby minimizing the number of injections and increasing the duration of effects in BRVO patients with macular edema, in a way that would emulate a physician in a typical office setting.
METHODS FROM SEPTEMBER 1, 2010 TO DECEMBER 31, 2012, ALL CONSEC-
utive patients with macular edema secondary to BRVO were considered for enrollment in this prospective interventional, randomized, multicenter study. The Institutional Review Board (IRB) of the University Eye Clinic of Milan and the IRB of the Ophthalmology Unit of Bologna approved this prospective clinical study on June 15, 2010, and the study was registered at the Ethics Committee of the Multimedica Group–San Giuseppe Hospital, as the coordinating center, under registration number OPH-12-10; the study adhered to the Health Insurance Portability and Accountability Act. All participants signed an informed consent before participation. The study conformed to the tenets of the Declaration of Helsinki. STUDY POPULATION:
Patients were naı¨ve to treatment and showed a macular-involved BRVO with decreased visual acuity and perfused macular edema for at least 3 months, with baseline central retinal thickness (CRT) >300 mm. Exclusion criteria were foveal hemorrhages that had not disappeared after 3 months of observation and ischemic maculopathy detected by fluorescein angiography. Patients with a history of ocular surgery and/or rubeotic or advanced glaucoma, defined as cup-to-disc ratio of 0.8 or worse, were excluded from participating. Patients who had had any previous treatment except for oral medication (eg, aspirin 100 mg) were excluded from this study, as well as patients in whom the underlying cause of edema was suspected to be different from BRVO (eg, diabetes, vitreomacular traction). In bilateral cases, only 1 eye per patient was enrolled in the study.
PRIMARY THERAPY OF DEXAMETHASONE INTRAVITREAL IMPLANT: All patients enrolled in this study under-
went a dexamethasone intravitreal implant 0.7 mg as a primary treatment at their first visit to the University Eye Clinic of Milan and to the Ophthalmology Unit of Bologna. After topical anesthesia with 2% lidocaine, the eye was irrigated with 10% povidone-iodine. The subconjunctival space around the anticipated injection site was then injected with 0.1-0.2 mL of 2% lidocaine. This step was followed by the insertion of the Ozurdex implant into the vitreous cavity through the pars plana using a customized, single-use 22 gauge applicator. Patients were 608
treated with a topical ophthalmic antibiotic 4 times daily starting 3 days before the day of their study procedure and continuing for 3 days after the procedure. Consecutive patients were randomly assigned 1:1 to 2 study groups. GROUP 1: After the first Ozurdex implant, patients in Group 1 were followed up monthly with best-corrected visual acuity (BCVA), ophthalmic examination, and spectraldomain optical coherence tomography (SD OCT) (Spectralis HRA OCT; Heidelberg Engineering, Heidelberg, Germany). Patients received retreatment with another Ozurdex implant 0.7 mg whenever there was a recurrence of macular edema. Recurrence was defined as a decrease of BCVA by 0.3 logMAR or as an increase of CRT by more than 50 mm compared to that in the previous examination. GROUP 2:
In Group 2 patients macular grid laser photocoagulation was performed in the macular area between weeks 6 and 8 after the intravitreal implant, to avoid excessive laser burns. Macular grid was performed under topical anesthesia, with a diode-pumped solid-state laser (GYC-1000; Nidek Co, Ltd, Japan) delivering 2-3 rows of 75 mm spots, 80 mW of intensity at 100 mm apart in the parafoveal region, guided by fluorescein angiography and SD OCT, using Goldmann 3 mirror contact lens. Then, 100-150 mm spots were applied 200 mm apart to the remaining areas of retinal thickening and capillary nonperfusion. The major vascular arcade defined the outer borders of the treated area, and the laser spots were aimed so as to avoid the foveal avascular zone and the papillomacular bundle, and applied until soft whitening of the retina became apparent. After the macular grid laser photocoagulation, the patients were followed up in the same manner as Group 1. When a recurrence of macular edema was detected on SD OCT or when visual acuity decreased by 0.3 logMAR, another Ozurdex implant 0.7 mg was given as a retreatment. Complications related to both treatments were recorded to evaluate the safety and adverse events of this protocol. The presence of nuclear, cortical, and posterior subcapsular lens opacities was measured during the slit-lamp examination using standardized photographs. Intraocular pressure (IOP) was also assessed. STUDY OUTCOMES: The primary outcome was final BCVA; secondary outcomes were final CRT by SD OCT, presence of macular edema, number of dexamethasone intravitreal implants, and the occurrence of any complications by the end of the follow-up period. All patients were evaluated at baseline and at every subsequent visit with BCVA, measured with a decimal visual acuity chart at 5 m, then converted to logMAR units. A change of more than 0.3 logMAR units compared to the previous measurement was defined as either an improvement or deterioration in visual acuity. Visual acuity was considered maintained if the logMAR change was less than 0.3 logMAR units compared to the baseline.
AMERICAN JOURNAL OF OPHTHALMOLOGY
MARCH 2014
TABLE 1. Baseline Characteristics of the Study Patients With Perfused Macular Edema Secondary to Branch Retinal Vein Occlusion Who Received a Baseline Dexamethasone Intravitreal Implant and Were Then Randomized to Asneeded Dexamethasone Implant Retreatment (Group 1) or Macular Grid Photocoagulation 6-8 Weeks After Baseline Followed by Retreatment With Dexamethasone Implant on an As-needed Basis (Group 2) Ozurdex Group 1 N ¼ 25
Ozurdex þ Grid Group 2 N ¼ 25
Sex, n (%) Male 11 (44) 11 (44) Female 14 (56) 14 (56) Age (y), median (range) 68 (53-81) 69 (52-78) Eye, n (%) Right 16 (67) 13 (52) Left 9 (33) 12 (48) Best-corrected visual acuity at 0.62 (0.32) 0.53 (0.21) baseline (logMAR), mean (SD) Best-corrected visual acuity 0.38 (0.30) 0.38 (0.17) 1 month after baseline implant (logMAR), mean (SD) Best-corrected visual acuity 0.23 (0.18) 0.15 (0.14) absolute difference from baseline (logMAR), mean (SD) Central retinal thickness at 466 (91) 426 (109) baseline (mm), mean (SD) 303 (67) 322 (51) Central retinal thickness at 1 month after baseline implant (mm), mean (SD) Central retinal thickness change 33.4 (16.9) 21.4 (14.4) from baseline, % (SD)
Slit-lamp examination, indirect fundus examination, IOP measurement, and SD OCT were performed at baseline visit and then repeated at every follow-up visit. The baseline examination included fluorescein angiography to assess the perfusion status of the macula. Patients in both groups were evaluated monthly for 6 months after the randomization to quantify maximum visual improvement at any given time point, and BCVA and CRT data were reported at 4 and 6 months. Central retinal thickness was tomographically defined as the distance between the inner limiting membrane and the retinal pigment epithelium–choriocapillaris interface on radial lines through the foveal area. The fovea was determined using the patient’s fixation and retinal landmarks. All SD OCT scans were performed with a scan rate of 40 000 A-scans per second in a 4.5 3 6.0-mm area. Subsequent follow-up scans were carried out at the exact location of the original scan using image alignment software. This image analysis allowed for an accurate determination of quantitative and qualitative retinal changes during follow-up.6 To verify the accuracy of the SD OCT findings, the fundus area encompassing the fovea was additionally VOL. 157, NO. 3
scanned using several vertical and horizontal scans. Retinal thickness of each SD OCT image section was obtained with the volume scan mode. Parafoveal retinal thickness was determined using the retinal thickness among the 9 subfields of the Early Treatment of Diabetic Retinopathy Study map, which is composed of 3 rings with radii of 1, 3, and 6 mm, respectively. Macular edema was tomographically defined as more than 300 mm of CRT. STATISTICAL ANALYSIS: Continuous variables have been described by mean values and standard deviation (SD). Absolute differences and percent changes at 4 and 6 months from randomization of BCVA and CRT levels were calculated and compared between treatment groups by using t test with Satterthwaite adjustment to allow for unequal variances. Numbers of Ozurdex implantations at 4 and 6 months from randomization were compared between treatment groups by using the Fisher exact test. All reported P values are 2-sided. A P value
