Efficacy of subthreshold micropulse laser photocoagulation therapy versus anti-vascular endothelial growth factor therapy for refractory macular edema secondary to non-ischemic branch retinal vein occlusion

Abstract

Objective

To assess the efficacy of subthreshold micropulse laser photocoagulation (SMLP) therapy versus anti-vascular endothelial growth factor (anti-VEGF) therapy in patients with refractory macular edema (ME) secondary to non-ischemic branch retinal vein occlusion (BRVO).

Methods

This single-center, prospective, nonrandomized, case-control trial involved patients with refractory ME that responded poorly to three or more initial anti-VEGF injections. The patients were examined and divided into two groups according to their chosen treatment: the intravitreal ranibizumab (IVR) group and the SMLP group. Both groups were followed up monthly for 12 months. Therapeutic efficacy and safety were assessed throughout the follow-up period.

Results

The IVR group comprised 49 eyes, and the SMLP group comprised 45 eyes. The improvements in the optical coherence tomography findings and visual acuity were comparable between the two groups at the final follow-up. The total number of injections was significantly lower in the SMLP than IVR group. No serious adverse events occurred during the study period.

Conclusions

SMLP therapy is better for patients with central macular thickness (CMT) of ≤400 μm. For patients with CMT of >400 μm, we advise continuation of anti-VEGF agents to reduce ME followed by application of SMLP therapy when CMT has decreased to ≤400 μm.

Introduction

Macular edema (ME) is a major complication of branch retinal vein occlusion (BRVO) that results in vision-impairing retinopathy. With the widespread use of intravitreal injections containing anti-vascular endothelial growth factor (anti-VEGF) agents, which has been the first-line treatment method for BRVO-induced ME, ¹ patients can achieve a certain degree of vision improvement. In clinical practice, however, we have found that some patients with BRVO-induced ME continue to have persistent, chronic, or recalcitrant ME after a period of regular anti-VEGF therapy. Despite the lack of uniform definition standards, we still refer to these cases as refractory ME. ^{2 , 3} Refractory ME with a poor response to anti-VEGF drugs is not uncommon. In a real-world study, ⁴ 43.7% of patients with BRVO who underwent intravitreal therapy did not achieve complete ME resolution. The RETAIN study ⁵ showed that 50% of eyes receiving anti-VEGF treatment had persistent ME at the end of the study period. Unfortunately, the current clinical approaches to refractory ME are focused on constantly increasing the number of intravitreal injections ⁶ and changing the type of anti-VEGF agents. ⁷ However, studies have shown that multiple intravitreal injections may increase the risk of endophthalmitis and retinal detachment. Additionally, use of an excessive number of anti-VEGF agents will increase the risk of cerebrovascular accidents, especially for patients with cardiovascular disease. ⁸ Therefore, new treatments are needed to compensate for patients with refractory ME who respond poorly to anti-VEGF therapy.

Subthreshold micropulse laser photocoagulation (SMLP) was recently developed and has been applied to the treatment of diabetic ME, ⁹ central serous chorioretinopathy, ¹⁰ ME due to retinal vein occlusion (RVO), ¹¹ and Coats disease. ¹² SMLP achieves its therapeutic effect without causing retinal cell death by exerting heat stress on retinal pigment epithelium (RPE) cells and eliciting their biological response. This improves retinal edema by drainage of subretinal fluid. ^{13 , 14} Several studies have shown that SMLP combined with intravitreal therapy for ME is effective and safe. ¹⁵ However, no study has been specifically designed to assess the best therapeutic approach of SMLP for refractory ME secondary to non-ischemic BRVO. In addition, two questions remain unanswered: Might SMLP therapy lead to resolution of edema and allow patients to discontinue injections if they were to require further injections to control refractory ME? How might SMLP therapy and anti-VEGF therapy be used effectively to treat refractory ME?

This study was performed to compare the efficacy of SMLP therapy versus anti-VEGF therapy among patients with refractory ME secondary to non-ischemic BRVO. The therapeutic effect of SMLP therapy was assessed to facilitate selection of the best treatment option.

Methods

This single-center, prospective, nonrandomized, case-control trial was approved by the Research Ethics Committee of the Affiliated Hospital of Yunnan University (Kunming, Yunnan, China; No. 2022130, June 2022) and registered with the Chinese Clinical Trial Registry (ChiCTR2300067505). The study followed the tenets of the Declaration of Helsinki. The reporting of this study conforms to the STROBE guidelines. ¹⁶

Patient selection

Patients with refractory ME ¹⁷ due to non-ischemic BRVO were examined in the Department of Ophthalmology, Affiliated Hospital of Yunnan University, China. Consecutive patients were invited to participate in the study. Each patient provided written informed consent to participate after they had received an explicit explanation of the purpose of the trial and potential adverse effects of each treatment.

The inclusion criteria were (1) an age of >18 years with a diagnosis of BRVO-induced ME, (2) a history of intravitreal anti-VEGF treatment (only ranibizumab is an on-label treatment for BRVO-induced ME in China) involving three or more injections within the previous 6 months (24 weeks), (3) central macular thickness (CMT) of >250 µm measured using spectral-domain optical coherence tomography (SD-OCT) (Nidek OCT RS-3000; Nidek Co., Ltd., Gamagori, Aichi, Japan), and (4) fundus fluorescein angiography (FFA) showing no signs of ischemia (or ischemia having been ameliorated by conventional laser therapy outside the macular arch).

The exclusion criteria were a history of other underlying retinal vascular diseases or other retinal pathologies, a history of vitreoretinal surgery or cataract surgery within the previous 6 months, and a history of steroid therapy within the previous 6 months. Patients were ineligible if they had retinal thickening due to an epiretinal membrane or vitreomacular traction syndrome.

Each included patient was assigned a unique identification number, and all patient-identifying information was anonymized before the study. Patient data including age, sex, and comorbidities (e.g., blood pressure, liver and renal function) were obtained from the medical records. Each patient then underwent complete ophthalmological examinations at monthly intervals that included best-corrected visual acuity (BCVA), a slit-lamp examination, dilated fundoscopy, OCT angiography (OCTA) (Optovue RTVue XR Avanti; Optovue Inc., Fremont, CA, USA), and SD-OCT (Spectralis; Heidelberg Engineering, Heidelberg, Germany) over a planned follow-up of 12 months. FFA was carried out every 3 months. If the patient’s disease progressed to the ischemia stage during follow-up, they were removed from the study.

Treatment procedures

The patients were divided into two groups based on the treatments they had chosen. In the intravitreal ranibizumab (IVR) group, the patients received anti-VEGF therapy in the form of ranibizumab (Lucentis; Genentech Inc., South San Francisco, CA, USA) at a dose of 1.25 mg/0.05 mL per injection according to protocol. In the SMLP group, the patients were treated with a micropulse laser (IQ 577; IRIDEX Corp., Mountain View, CA, USA) at a 577-nm wavelength via a laser contact lens (Volk Optical, Inc., Mentor, OH, USA). The laser parameters were a 200-µm retinal spot size, 0.2-s exposure, and 5% duty cycle. The micropulse laser was applied in the continuous-wave mode after titrating for each eye. Both groups were followed up monthly for 12 months. In addition, IVR or SMLP treatment was repeated and continued if suggested by two doctors on follow-up visits. In both groups, the therapies were paused if the ME resolved completely.

Evaluation of therapeutic efficacy

For evaluation of therapeutic efficacy, the primary outcome was the CMT at the 12-month examination. The secondary outcomes were the change in BCVA and the number of IVR injections or SMLP procedures needed during the 12-month follow-up, as well as the number of total injections, including the number of injections at baseline and the number of injections during the study.

Evaluation of therapeutic safety

Therapeutic safety was estimated by the change in macular microvasculature, represented by the superficial and deep vascular density (VD), the areas of the superficial and deep foveal avascular zone (FAZ) on OCTA, and the scars found by fundus photography. All adverse events, including serious adverse events, that were detected during the study period were recorded.

Statistical analysis

The sample size was calculated based on the primary outcome (CMT at 12 months). We assumed mean CMTs of 250 μm and 270 μm for the SMLP group and IVR group, respectively, with a common standard deviation of 32 μm. A sample size of 42 eyes per group would achieve 80% power at a two-sided significance level (alpha) of 0.05 using a two-sample t-test. With an expected dropout rate of <10%, 47 eyes per group were required. The sample size calculation was performed using PASS software V.15.0 (NCSS, Kaysville, UT, USA).

The statistical analyses for the evaluation of therapeutic efficacy and safety were performed using SPSS 27.0.0.0 (IBM Corp., Armonk, NY, USA), and the significance level (alpha) was set to 0.05. Because of differences in cognitive levels and educational backgrounds among the patients, BCVA was assessed using a Snellen chart and converted to the logarithm of the minimum angle of resolution (logMAR) units for statistical analysis. Student’s t-test (paired or unpaired depending on the group) was used to evaluate the changes in BCVA, CMT, number of IVR injections, and FAZ area. The chi-square test was applied to compare proportions. Risk factors associated with the treatment of refractory ME were determined using multiple linear regression.

Results

Overall, 94 eyes of 94 patients with a mean age of 53.12 ± 6.52 years (range: 36–72 years) were enrolled in the study. There were 49 eyes in the IVR group and 45 eyes in the SMLP group. All patients were regularly followed up and completed the entire 12-month follow-up. The mean follow-up time was 13.11 ± 2.5 months. FFA showed that all patients’ disease remained in the non-ischemic stage throughout follow-up. There were no statistically significant differences in age, sex, visual acuity, CMT, time from last treatment to enrollment, number of previous anti-VEGF injections, or general condition between the two groups. The patients’ characteristics are shown in Table 1.

Table 1.

Baseline clinical characteristics and parameters in patients with refractory macular edema secondary to non-ischemic branch retinal vein occlusion.

	IVR group (n = 49)	SMLP group (n = 45)	P
Age, years	59.42 ± 10.10	57.06 ± 7.79	0.21*
Sex, male/female	24/25	17/27	0.38†
BCVA, logMAR	0.79 ± 0.20	0.76 ± 0.18	0.47*
CMT, μm	512.71 ± 85.43	490.53 ± 109.27	0.27*
Time from last treatment to enrollment, months	3.22 ± 1.531	3.48 ± 1.37	0.38*
Dyslipidemia	20 (40.81)	21 (46.67)	0.56†
Hypertension	28 (57.14)	26 (57.79)	0.95†
Number of previous anti-VEGF injections	4.04 ± 1.22	3.39 ± 1.26	0.67†
Superficial capillary plexuses
VD, %	19.30 ± 2.44	20.13 ± 2.75	0.12*
FAZ, mm2	0.43 ± 0.074	0.45 ± 0.07	0.37*
Deep capillary plexuses
VD, %	13.48 ± 2.7	12.71 ± 2.43	0.15*
FAZ, mm2	0.76 ± 0.07	0.73 ± 0.06	0.06*

Data are presented as mean ± standard deviation, n, or n (%).

*Student’s t-test, †Chi-square test.

IVR, intravitreal ranibizumab; SMLP, subthreshold micropulse laser photocoagulation; BCVA, best-corrected visual acuity; logMAR, logarithm of the minimum angle of resolution; CMT, central macular thickness; VEGF, vascular endothelial growth factor; FAZ, foveal avascular zone; VD, vascular density.

CMT

In both groups, CMT was significantly better at the final visit than at baseline (P < 0.05) (Figure 1 and Figure 2). At 6 months, CMT was lower in the SMLP than IVR group (267.56 ± 30.695 vs. 300.22 ± 60.381 µm, respectively). However, there was no significant difference between the SMLP and IVR groups at 12 months (263.84 ± 33.955 vs. 274.71 ± 42.086 µm, respectively).

Figure 1.

Time course of changes in BCVA and CMT. (a) BCVA was significantly improved at all time points compared with that at baseline in both groups. (b) CMT was significantly reduced at all time points compared with that at baseline in both groups.

BCVA, best-corrected visual acuity; logMAR, logarithm of the minimum angle of resolution; CMT, central macular thickness; IVR, intravitreal ranibizumab; SMLP, subthreshold micropulse laser photocoagulation; MO, months.

Figure 2.

Optical coherence tomography scans of patients with refractory macular edema secondary to non-ischemic branch retinal vein occlusion. (a–d) Images of a patient who chose IVR therapy. (a, b) At baseline. (c, d) At 12 months after IVR therapy. (e–h) Images of a patient who chose SMLP therapy. (e, g) At baseline and (f, h) At 12 months after SMLP therapy.

IVR, intravitreal ranibizumab; SMLP, subthreshold micropulse laser photocoagulation.

BCVA

In both groups, BCVA was significantly better at the final visit than at baseline (P < 0.05) (Figure 1). However, there was no significant difference between the IVR and SMLP groups at 12 months (0.339 ± 0.11 vs. 0.308 ± 0.10 logMAR, respectively) (Figure 1).

Therapeutic efficacy in the subgroups

For a more detailed evaluation of the treatment effect, the two therapy groups were classified into two subgroups based on CMT of ≤400 or >400 μm by OCT. For CMT, the subgroup with CMT of ≤400 μm showed significantly greater improvement in the SMLP than IVR group at 3, 6, and 9 months. However, the subgroup with CMT of >400 μm had the opposite result at 3 and 6 months (Figure 3). In the subgroup with CMT of ≤400 μm, the BCVA was significantly better in the SMLP than IVR group at 12 months (0.264 ± 0.07 vs. 0.350 ± 0.11 logMAR) (P < 0.05). However, it did not reach statistical significance in the subgroup with CMT of >400 μm at the final visit (Figure 3).

Figure 3.

Time course of changes in BCVA and CMT in different subgroups. (a, b) Changes in BCVA and CMT in the CMT >400 μm subgroup. (a) BCVA was significantly better in the IVR than SMLP group at 6 months. (b) CMT was significantly better in the IVR than SMLP group at 3 and 6 months. (c, d) Changes in BCVA and CMT in the CMT ≤400 μm subgroup. (c) BCVA in the SMLP group was significantly improved at 3, 6, and 12 months compared with the values in the IVR group and (d) CMT in the SMLP group was significantly reduced at 3, 6, and 9 months compared with the values in the IVR group.

BCVA, best-corrected visual acuity; logMAR, logarithm of the minimum angle of resolution; CMT, central macular thickness; IVR, intravitreal ranibizumab; SMLP, subthreshold micropulse laser photocoagulation.

Number of treatments

The mean number of injections in the IVR group was 4.21 ± 0.20 (range, 3–6) during the study course. The total number of injections, including the number of injections at baseline and the number of injections during the study, was 8.04 ± 0.2. In the SMLP group, the patients received 2.11 ± 0.53 (range, 2–4) laser treatments. The total number of injections was 3.39 ± 1.26. The total number of injections was significantly lower in the SMLP than IVR group (P < 0.05).

Change in macular microvasculature

In both groups, no significant differences were found in VD or the FAZ area at baseline versus during follow-up. Furthermore, no significant differences were found between the two groups at the final visit (Figure 4 and Table 2).

Figure 4.

Comparison of the superficial and deep VD and the area of the superficial and deep FAZ in patients with refractory macular edema secondary to non-ischemic branch retinal vein occlusion. (a, b) No significant difference was found in the superficial or deep VD at any time points between the two groups and (c, d) No significant difference was found in the superficial or deep FAZ at any time points between the two groups.

VD, vascular density; FAZ, foveal avascular zone; IVR, intravitreal ranibizumab; SMLP, subthreshold micropulse laser photocoagulation.

Table 2.

Comparison of optical coherence tomography angiography parameters between eyes before and after treatment and between eyes in different groups

	Superficial capillary plexuses						Deep capillary plexuses
	VD (%)			FAZ (mm2)			VD (%)			FAZ (mm2)
	Baseline	12 months	P	Baseline	12 months	P	Baseline	12 months	P	Baseline	12 months	P
IVR	29.30 ± 2.44	29.91 ± 2.53	0.201	0.43 ± 0.07	0.42 ± 0.03	0.773	23.48 ± 2.78	23.61 ± 2.33	0.45	0.76 ± 0.07	0.78 ± 0.08	0.24
SMLP	30.13 ± 2.75	29.15 ± 3.47	0.100	0.45 ± 0.07	0.44 ± 0.06	0.745	22.71 ± 2.43	23.08 ± 1.53	0.40	0.73 ± 0.06	0.76 ± 0.13	0.23
P	0.12	0.22	–	0.37	0.20	–	0.15	0.12	–	0.06	0.37	–

Data are presented as mean ± standard deviation.

IVR, intravitreal ranibizumab; SMLP, subthreshold micropulse laser photocoagulation; FAZ, foveal avascular zone; VD, vascular density.

Adverse events

In the IVR group, 11 patients developed subconjunctival hemorrhage and 5 patients showed a brief increase in intraocular pressure. All patients received appropriate medical treatment that resolved all adverse events within 1 week. No scars were found by fundus photography in the SMLP group. No patient developed significant visual deterioration after SMLP. No other adverse events of any kind were detected during the study period.

Risk factors

The presence of ME at the last follow-up observation was significantly associated with poor blood pressure control, dyslipidemia, and the time from last treatment to enrollment (P < 0.05) (Table 3). Other factors had no significant associations (Table 3).

Table 3.

Risk factors associated with the treatment effect.

Variables	β (95% CI)	P
Treatment group	−0.045	0.612
Time from last treatment to enrollment (months)	0.303	0.002
Dyslipidemia	−0.314	0.001
Hypertension	−0.210	0.034
Age (years)	0.037	0.674
Number of previous anti-VEGF injections	−0.128	0.134

CI, confidence interval; VEGF, vascular endothelial growth factor.

Discussion

In this study, we compared the treatment effects of SMLP versus IVR injections in patients with refractory ME secondary to non-ischemic BRVO. Both treatment strategies achieved structural improvement, a reduction in CMT, and improved BCVA at the final visit.

The patients who received SMLP therapy had the same treatment outcomes as the patients who received an increased number of anti-VEGF injections for refractory ME. The results showed that patients in the SMLP group needed fewer injections than those the IVR group (3.39 ± 1.26 vs. 8.04 ± 0.20, respectively). These results address the first question posed earlier: SMLP therapy could, in practice, lead to the resolution of edema and allow patients to discontinue injections if they were to require further injections to control refractory ME.

Our protocol of SMLP therapy for patients with refractory ME who had a history of intravitreal anti-VEGF after three or more injections within the previous 6 months was similar to the strategy of combination therapy in some studies. ¹⁷ The results indicate that the combination of laser therapy and an anti-VEGF agent is safe and effective for visual gain and anatomic stabilization while requiring fewer injections. In a study by Terashima et al., ¹⁸ the patients received an initial IVR injection and were then divided into an IVR monotherapy group (IVR injections were delivered pro re nata (PRN)) or combination therapy group (IVR + SMLP). Compared with monotherapy, the combination therapy resulted in better visual acuity and a lower rate of ME. However, our study showed no significant differences between the SMLP and IVR groups at the final visit. We believe that the difference in these results might stem from the different inclusion criteria used for the patients. We emphasized the selection of patients with refractory ME, whose CMT was worse than that in the criteria used by Terashima et al. ¹⁸ However, Song et al. ¹⁹ compared a combination treatment (IVR + macular grid laser) with an IVR group, and both groups showed improved BCVA and decreased CMT. The authors suggested that there was no need to combine macular grid photocoagulation in the treatment of ME secondary to BRVO because the combination group did not show a reduced number of ranibizumab injections. ¹⁹ Conversely, we recommend SMLP treatment for patients with refractory ME. One reason for this recommendation is that we prefer SMLP to macular grid photocoagulation from a safety standpoint. In animal experiments, ¹⁴ no laser coagulation spots were observed on fundus photography after subthreshold laser treatment. Likewise, no patients in our study developed retinal scars, hemorrhage, perforations, or neovascularization. Another reason for our recommendation is that SMLP treatment effectively reduced the number of injections in our study. Compared with the SMLP group, the IVR group needed 4.21 ± 0.20 additional injections. This means that patients with refractory ME needed 8.04 ± 0.20 injections in total. A higher number of injections is associated with higher costs and risks.

To answer the second question, “How might the SMLP and anti-VEGF treatment methods be used effectively to treat refractory ME?,” we divided the patients into two subgroups based on CMT of ≤400 or >400 μm. The results showed that SMLP therapy is more effective for patients with CMT of ≤400 μm than for those with CMT of >400 μm. This is consistent with previous results. ²⁰ Two possible reasons for this result are as follows. First, it might be caused by the mechanism by which SMLP therapy exerts heat stress on RPE cells and elicits their biological response to drain subretinal fluid, which improves ME. ²¹ However, high amounts of subretinal fluid may change the distribution of laser energy throughout the retina and RPE. Second, when we chose the 577-nm SMLP setting for safety (because yellow wavelengths are minimally absorbed by macular xanthophylls), we had to accept its lower penetration in severe retinal swelling than that accomplished by an 810-nm diode laser or a 532-nm green laser. Thus, our results suggest that SMLP therapy is more effective for patients with CMT of ≤400 μm. For patients with CMT of >400, however, it might be a better strategy to continue using anti-VEGF agents to reduce ME and then switch to SMLP therapy when the CMT has decreased to ≤400 µm.

The method by which micropulse laser therapy is effectively applied must also be considered. Laser patterns and treatment methods might have important effects on treatment outcomes. Opinions on laser patterns differ because of the lack of a standardized protocol for subthreshold laser treatment. In our study, a 577-nm laser was set at a 200-µm spot diameter, 0.2-s exposure, and 5% duty cycle. However, Terashima et al. ¹⁸ employed a 577-nm laser with a 100-µm spot diameter, 0.2-s exposure, 10% to 15% duty cycle, and 50% to 100% threshold power. Buyru Özkurt et al. ²² used a spot diameter of 100 to 150 µm, 0.2-s exposure, and 5% duty cycle. Although different parameters were used among these studies, neither our study nor other studies reported damage to the retina, suggesting that the use of these parameters is safe. However, because no visible scars are left on the retina, the maximum amount of laser treatment that might be applied to the retina is unknown; this makes it difficult to judge the time points for repeated treatment. Few studies have mentioned the specific time point for SMLP retreatment. Terashima et al. ¹⁸ applied additional SMLP therapy at intervals of at least 3 months, and the mean number of SMLP procedures was 1.4 ± 0.6. By contrast, based on our clinical experience, we preferred SMLP retreatment at intervals of at least 1 month when ME persisted during the follow-up period. Despite our higher number of repeat treatments (2.1 ± 0.5), no retinal damage was found, demonstrating that this method is safe. Although more clinical data are needed to verify the effects of this therapeutic method, our findings suggest that more aggressive treatment might be acceptable when using SMLP for refractory ME due to BRVO.

A recent report indicated that FAZ area enlargement in eyes with RVO reflects macular ischemia, which might be a reliable biomarker of visual impairment. ²³ OCTA has been shown to be a useful imaging tool to evaluate the FAZ and parafoveal capillary networks in eyes with RVO. ²⁴ In the present study, the area of the superficial and deep FAZ and VD measured via OCTA were larger after treatments, but no significant differences were detected. The measured increase might have reflected the spread of vessels as the edema resolved rather than indicating the worsening of ischemia.

We also analyzed the clinical risk factors for a poor response to the treatment of refractory ME. Dyslipidemia and high systemic arterial blood pressure were the prognostic factors for ME persistence. The irregular timing of treatment and follow-up during the COVID-19 pandemic was also a risk factor for patients with refractory ME in this study, as shown previously. ²⁵ Therefore, patients with refractory ME might benefit from controlling systemic conditions, such as blood pressure and lipid levels, with the help of the physician and receiving more regular ophthalmic follow-up treatments.

Notably, we defined refractory ME as a lack of response after three or more ranibizumab injections for two reasons. First, the anti-VEGF agents used in China include ranibizumab, aflibercept, and conbercept. Only ranibizumab was approved for treatment of ME secondary to BRVO in China at the beginning of this study. Conbercept was subsequently approved in March 2023, and aflibercept was still an off-label treatment at the time of this writing. Therefore, we only analyzed the data of patients who were treated by ranibizumab. Second, we chose three ranibizumab injections as the baseline number of injections because the clinical protocol of ranibizumab suggested that participants receive three or more PRN injections for BRVO-induced ME. ²⁶ Three injections are considered the minimum effective number of treatments to evaluate the therapeutic effect of BRVO-induced ME. Therefore, one of our inclusion criteria was a lack of response after three or more ranibizumab injections.

This study had some limitations. First, the nonrandomized study design raises the possibility of selection and information biases; the study was also affected by some practicality issues such as treatment costs and reimbursement policies. Second, it was a single-center study with a small sample size. Therefore, large-scale randomized controlled clinical trials or real-world research data are needed to confirm the reliability of our findings. Another important point is that the study demonstrated the superiority of SMLP therapy in the treatment of refractory ME during a short-term follow-up. More long-term clinical studies are needed to prove the safety and efficacy of SMLP in the treatment of vitreoretinal disease. Notably, we only obtained data from patients using ranibizumab, which may have introduced selection bias. We will collate more data from patients who use different kinds of drugs, including conbercept or steroid implants such as dexamethasone intravitreal implants, in future studies.

Conclusion

Our data showed that SMLP treatment might improve both structure and function in patients with refractory ME secondary to BRVO. Our study also suggested that an earlier switch to SMLP therapy might be a better treatment strategy for patients with refractory ME if the response is poor after three or more initial anti-VEGF injections, especially for patients whose CMT is ≤400 μm.

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

ORCID iD

Xiaoxiao https://orcid.org/0000-0002-8538-7069