Outcomes of Selective Laser Trabeculoplasty for Sustained Elevation of Intraocular Pressure Following Intravitreal Anti-VEGF Injections

Abstract

Purpose

To evaluate the efficacy and predictors of success of selective laser trabeculoplasty (SLT) for sustained elevation of intraocular pressure (SE-IOP) associated with anti-vascular endothelial growth factor (VEGF) therapy.

Design

Retrospective observational case series.

Methods

A retrospective chart review was performed for patients who received SLT following one or more anti-VEGF injections at an academic center between January 2013 and June 2022. Exclusionary criteria included history of non-open-angle glaucoma, incisional glaucoma surgery, or SLT more than 24 months after anti-VEGF injections.

Main Outcome Measures

Outcomes included mean IOP and medication use up to 24 months following SLT. Failure was defined as less than 20% IOP reduction, increase in medication use, or need for additional laser or surgical interventions.

Results

Thirty-three eyes of 31 patients were identified. Baseline IOP was 25.1 ± 6.1 mmHg on 2.7 ± 1.3 medications. SLT significantly reduced mean IOP but not medication use at all recorded time frames. At 24 months, mean IOP was 17.5 ± 8.2 mmHg (P = 0.009) on 3.1 ± 0.9 medications (P = 0.313). Success rates were 48.5% at 6 months, 32.3% at 12 months, and 23.6% at 24 months. The frequency of anti-VEGF injections (HR: 1.318; 95% CI: 1.072 to 1.623; P = 0.008) was the only significant predictor of SLT success in the multivariate analysis.

Conclusion

SLT may be an effective adjunctive therapy for SE-IOP associated with anti-VEGF injections. The frequency of anti-VEGF injections prior to SLT may be a key predictor of treatment success.

Introduction

Intravitreal anti-vascular endothelial growth factor (VEGF) injections have become indispensable in the treatment of various retinal diseases, such as age-related macular degeneration, retinal vein occlusion, and diabetic retinopathy. Despite a favorable safety profile, a concerning side effect of anti-VEGF injections is elevation in intraocular pressure (IOP). Elevations in IOP are often transient and resolve spontaneously, but in select eyes, IOP may remain persistently elevated following anti-VEGF injections.¹ According to a meta-analysis, the prevalence of sustained elevation of intraocular pressure (SE-IOP) associated with anti-VEGF injections is estimated to be 4.7%,² underscoring the importance of monitoring IOP in eyes receiving these treatments.

While medications appear to be effective in addressing most cases of SE-IOP associated with anti-VEGF injections,^3–14 more challenging cases requiring laser trabeculoplasty or glaucoma surgery^{6–9,11,12,15} have been reported in the literature. Previous studies also suggest that injected or more frequently injected eyes may be more likely to receive additional glaucoma interventions compared to uninjected or less frequently injected eyes, respectively.^16,17 A commonly performed procedure to lower IOP is selective laser trabeculoplasty (SLT), which has been demonstrated to be effective as a primary or adjunctive therapy for open-angle glaucoma and ocular hypertension. Yet, the existing literature is limited to a few case reports of SLT for the management of SE-IOP associated with anti-VEGF injections.^7,8,10,11 Thus, the purpose of our study was to examine the outcomes of SLT for SE-IOP associated with anti-VEGF injections.

Methods

Sample Selection

A retrospective chart review was performed for patients who received SLT following one or more anti-VEGF injections at a single academic center between January 2013 and June 2022. Patients were screened from the electronic medical records by Current Procedural Terminology (CPT) codes. Eyes were excluded for history of non-open-angle glaucoma or incisional glaucoma surgery or if they received SLT more than 24 months after their last anti-VEGF injection. If an eye received multiple SLT treatments, the first SLT following the initiation of anti-VEGF therapy was referenced. This study was approved by the Institutional Review Board of Yale University and was conducted in accordance with the tenets of the Declaration of Helsinki. As this was a retrospective study, the requirement for individual patient informed consent was waived by the Institutional Review Board. Patient data confidentiality was maintained in accordance with Institutional Review Board policies and all relevant regulations.

Outcome Measures

Primary outcome measures included IOP and glaucoma medication use at 1 week, 1 month, 3 months, 6 months, 12 months, 18 months, and 24 months following SLT. The secondary outcome measure included time to failure, which was defined as less than 20% reduction in IOP without additional medical (increase in medication use over two consecutive visits), laser, or surgical treatment. The baseline IOP was defined as the IOP measured immediately prior to SLT. Follow-up visits were grouped into several time frames (1 to 14, 15 to 60, 61 to 122, 123 to 272, 273 to 456, 457 to 639, 640 to 913 days) outlined by the World Glaucoma Association (WGA). If there were multiple visits within a time frame, data were averaged prior to further analysis. For Kaplan-Meier survival analysis, data from every follow-up visit were independently analyzed. To account for postoperative fluctuations, data from the first week following SLT were not considered for the determination of failure. Eyes were censored following additional laser or surgical treatment, loss to follow-up, or up to 60 months, whichever came first.

Statistical Analyses

Normality of the differences between IOP and medication use prior to and following SLT were examined by the D’Agostino-Pearson test. Accordingly, the change in intraocular pressure was evaluated by the two-tailed paired t-test and the change in medication use was evaluated by the two-tailed Wilcoxon matched-pairs signed rank test. The cumulative probability of survival and the associations between potential predictors and treatment success were analyzed by Kaplan-Meier survival analysis and Cox proportional hazards models. A sandwich variance estimator was applied to account for potential correlation between eyes from individual participants. Variables with P values less than 0.10 in the univariate analysis were included in the multivariate analysis. Statistical analyses were performed with Prism 9.5 (GraphPad Software, La Jolla, CA, USA) and R version 4.3 (R Foundation for Statistical Computing, Vienna, Austria). Continuous variables were reported as mean ± standard deviation (range) and categorical variables were reported as frequencies (percentages). A P value less than 0.05 was considered statistically significant.

Results

Study Sample

The sample selection process is outlined in Figure 1. 47 eyes of 42 patients were initially identified but 14 eyes of 11 patients were excluded for the following reasons: history of neovascular or chronic angle-closure glaucoma (4 eyes); trabeculectomy (4 eyes); and SLT more than 24 months after the last anti-VEGF injection (6 eyes). The final cohort consisted of 33 eyes of 31 patients, all of which were followed up for a minimum of 6 weeks, and none of which had history of other intravitreal injections or implants or anterior segment or vitreoretinal surgery. Based on the records available, 12 of 33 (36.4%) eyes had no preexisting diagnosis of glaucoma, glaucoma suspect, or ocular hypertension, and the mean IOP was 15.3 ± 3.1 (range: 8 to 21) mmHg on 1.0 ± 1.1 (range: 0 to 3) medications prior to anti-VEGF therapy.

Figure 1.

Flowchart of the sample selection process. A total of 47 eyes from 42 patients were identified as having undergone SLT following anti-VEGF injections. After excluding 4 eyes with neovascular or angle-closure glaucoma, 4 with prior trabeculectomy, and 6 with SLT performed more than 24 months after the last injection, 33 eyes from 31 patients remained for analysis.

Baseline Characteristics

The demographic and clinical characteristics of the study sample at the time of SLT are detailed in Table 1. The mean age was 78.2 ± 10.6 (range: 52 to 92) years. Most eyes received aflibercept and/or bevacizumab for wet age-related macular degeneration or retinal vein occlusion and had diagnoses of moderate to severe primary open-angle glaucoma or ocular hypertension. Eyes received mean 14.0 ± 9.3 (range: 1 to 36) anti-VEGF injections over 36.8 ± 23.3 (range: 1.7 to 92.9) months prior to SLT. The mean annual number of anti-VEGF injections between the time of the first anti-VEGF injection and SLT was 5.0 ± 2.3 (range: 0.8 to 10.4) injections. 21 of 33 (63.6%) eyes received their last anti-VEGF injection less than 3 months prior to SLT. During the SLT treatment, eyes received mean 94.1 ± 9.7 (range: 68.0 to 107.0) shots at 0.6 ± 0.1 (range: 0.4 to 0.8) mJ per shot for a total energy of 55.7 ± 10.9 (range: 40.0 to 85.6) mJ over 360 degrees of the angle.

Table 1.

Demographic and Clinical Characteristics of the Study Sample at the Time of SLT

Characteristic	All Patients (n = 31 Patients; 33 Eyes)
Age (years), mean ± SD (range)	78.2 ± 10.6 (52 to 92)
Sex, n (%)
Male	17 (51.5%)
Female	16 (48.5%)
Race, n (%)
White or Caucasian	25 (75.8%)
Black or African American	3 (9.1%)
Asian	1 (3.0%)
Other/Unknown	4 (12.1%)
Lens status, n (%)
Phakic	20 (60.6%)
Pseudophakic	13 (39.4%)
Family history of glaucoma, n (%)	12 (36.4%)
Cup-to-disc ratio, mean ± SD (range)	0.6 ± 0.2 (0.1 to 0.9)
Central corneal thickness (µm), mean ± SD (range)	552.2 ± 34.2 (467 to 613)
Treatment diagnosis, n (%)
OHTN	11 (33.3%)
POAG (mild)	2 (6.1%)
POAG (moderate)	7 (21.2%)
POAG (severe)	8 (24.2%)
POAG (unspecified)	5 (15.2%)
Anti-VEGF therapy
Indication, n (%)
Wet AMD	19 (57.6%)
RVO	8 (24.2%)
DME	6 (18.2%)
Agent, n (%)
Bevacizumab only	10 (30.3%)
Aflibercept only	10 (30.3%)
Mixed#	13 (39.4%)
Total number of injections, mean ± SD (range)	14.0 ± 9.3 (1 to 36)
Time since first injection (months), mean ± SD (range)	36.8 ± 23.3 (1.7 to 92.9)
Mean annual number of injections, mean ± SD (range)	5.0 ± 2.3 (0.8 to 10.4)

Notes: ^#Bevacizumab, aflibercept, and/or ranibizumab.

Abbreviations: SD, standard deviation; OHTN, ocular hypertension; POAG, primary open-angle glaucoma; AMD, age-related macular degeneration; RVO, retinal vein occlusion; DME, diabetic macular edema.

Postoperative Outcomes

The mean IOP and medication use at baseline and follow-up time frames are detailed in Table 2. The mean baseline IOP was 25.1 ± 6.1 (range: 15 to 40) mmHg on 2.7 ± 1.3 (range: 0 to 6) medications. At baseline, 31 of 33 (93.9%) eyes were already on one or more medications. Overall, there was a significant reduction in mean IOP but no significant reduction in mean medication use at all follow-up time frames. The mean IOP was 17.7 ± 4.3 (range: 10.0 to 29.5) mmHg at 6 months (P < 0.001), 18.0 ± 7.1 (range: 10.2 to 38.0) mmHg at 12 months (P < 0.001), and 17.5 ± 8.2 (range: 9.0 to 39.0) mmHg at 24 months (P = 0.009) following SLT. The mean percentage reduction in IOP was 29.5% at 6 months, 28.3% at 12 months, and 30.3% at 24 months following SLT. The mean medication use was 2.9 ± 1.0 (range: 1.0 to 4.5) medications at 6 months (P = 0.102), 2.9 ± 1.2 (range: 0.8 to 5.5) medications at 12 months (P = 0.363), and 3.1 ± 0.9 (range: 1.0 to 4.0) medications at 24 months (P = 0.313) following SLT. There were no reported significant complications from SLT.

Table 2.

Mean Intraocular Pressure and Medication Use at Baseline and Follow-Up Time Frames

Follow-Up	Time Frame (Days)	N	Intraocular Pressure		Number of Medications
			Mean ± SD (Range)	P value	Mean ± SD (Range)	P value
Baseline	–	33	25.1 ± 6.1 (15.0 to 40.0)	–	2.7 ± 1.3 (0.0 to 6.0)	–
Week 1	[1, 14]	28	19.6 ± 5.1 (11.0 to 34.0)	<0.001*	2.9 ± 1.3 (0.0 to 6.0)	0.999
Month 1	[15, 60]	30	18.9 ± 5.0 (10.0 to 36.0)	<0.001*	2.7 ± 1.2 (0.0 to 5.0)	0.500
Month 3	[61, 122]	24	18.7 ± 4.6 (11.0 to 32.5)	<0.001*	2.7 ± 0.9 (1.0 to 4.0)	0.453
Month 6	[123, 272]	22	17.7 ± 4.3 (10.0 to 29.5)	<0.001*	2.9 ± 1.0 (1.0 to 4.5)	0.102
Month 12	[273, 456]	19	18.0 ± 7.1 (10.2 to 38.0)	<0.001*	2.9 ± 1.2 (0.8 to 5.5)	0.363
Month 18	[457, 639]	14	17.2 ± 8.5 (9.5 to 38.0)	0.001*	3.0 ± 1.2 (1.0 to 5.0)	0.180
Month 24	[640, 913]	11	17.5 ± 8.2 (9.0 to 39.0)	0.009*	3.1 ± 0.9 (1.0 to 4.0)	0.313

Notes: Eyes were excluded from calculations only if they received additional laser or surgical treatment or were lost to follow-up. *An asterisk denotes a statistically significant difference (P < 0.05).

The Kaplan-Meier plot of the cumulative probability of success within 24 months of SLT is presented in Figure 2. The overall success rates were 48.5% (95% CI: 30.8% to 64.1%) at 6 months, 32.3% (95% CI: 17.2% to 48.5%) at 12 months, and 23.6% (95% CI: 10.0% to 40.3%) at 24 months following SLT. Overall, 25 of 33 (75.8%) eyes failed. Of 25 eyes that failed, 14 (56.0%) eyes failed due to less than 20% reduction in IOP, 7 (28.0%) eyes failed due to increase in medications, and 4 (16.0%) eyes failed due to additional laser or surgical treatment. Potential predictors of SLT success and their associated hazard ratios from Cox proportional hazards models are detailed in Table 3a and b. The frequency (HR: 1.364; CI: 1.161 to 1.604; P < 0.001) and total number of anti-VEGF injections (HR: 1.049; 95% CI: 1.006 to 1.094; P = 0.026) and sex (HR: 0.469; 95% CI: 0.203 to 1.084; P = 0.076) were the only variables with P values less than 0.10 in the univariate analyses. The frequency of anti-VEGF injections (HR: 1.318; 95% CI: 1.091 to 1.593; P = 0.004) was the only statistically significant predictor of SLT success while holding the total number of anti-VEGF injections (HR: 1.037; 95% CI: 0.992 to 1.083; P = 0.106) and sex (HR: 0.579; 95% CI: 0.238 to 1.408; P = 0.228) constant in the multivariate analysis. A scatter plot of the time to failure and frequency of anti-VEGF injections is presented in Figure 3.

Figure 2.

Kaplan-Meier plot of the cumulative probability of success. Tick marks denote censored eyes. Dashed lines represent 95% confidence intervals.

Table 3.

Cox Proportional Hazards Models with Sandwich Estimators for Potential Predictors of Treatment Failure

(a)
Characteristic+	Variable (Reference)	Univariate Analysis
		Hazards Ratio (95% CI)	P value
Age	Years	1.022 (0.981 to 1.065)	0.293
Sex	Male (female)	0.469 (0.203 to 1.084)	0.076
Lens status	Phakic (pseudophakic)	0.733 (0.339 to 1.583)	0.429
Central corneal thickness	µm	1.000 (0.993 to 1.008)	0.964
Treatment diagnosis	Ocular hypertension (glaucoma)	0.531 (0.188 to 1.498)	0.232
Baseline IOP	mmHg	0.965 (0.893 to 1.043)	0.375
Indication	Wet AMD (RVO) DME (RVO)	0.886 (0.368 to 2.136) 0.457 (0.083 to 2.498)	0.787 0.366
Last injection	≤ 3 months (> 3 months) prior	1.561 (0.632 to 3.856)	0.334
Injection before failure	Yes (no)	0.709 (0.354 to 1.424)	0.334
Injection agent	Bevacizumab only (aflibercept only) Mixed (aflibercept only)	0.409 (0.110 to 1.522) 1.950 (0.768 to 4.950)	0.182 0.160
Total injections#	Injections	1.049 (1.006 to 1.094)	0.026*
Injection frequency#	Injections per year	1.364 (1.161 to 1.604)	< 0.001*
(b)
Characteristic+	Variable (reference)	Multivariate Analysis
		Hazards Ratio (95% CI)	P value
Sex	Male (female)	0.579 (0.238 to 1.408)	0.228
Total injections#	Injections	1.037 (0.992 to 1.083)	0.106
Injection frequency#	Injections per year	1.318 (1.091 to 1.593)	0.004*

Notes: ⁺At the time of SLT, unless otherwise specified. ^#Between time of first anti-VEGF injection and SLT. Variables with P values less than 0.10 in the (a) univariate analysis (bolded) were included in the (b) multivariate analysis (bolded). *An asterisk denotes a statistically significant difference (P < 0.05).

Abbreviations: IOP, intraocular pressure; AMD, age-related macular degeneration; RVO, retinal vein occlusion.

Figure 3.

Scatter plot of the time to failure and frequency of anti-VEGF injections. Squares denote censored eyes.

Discussion

To the best of our knowledge, our study is the first to date to examine the details of outcome of SLT for SE-IOP associated with anti-VEGF injections. We are aware of only the following reports in the existing literature of SLT for the management of SE-IOP. In one case series of 25 eyes with SE-IOP, 3 eyes received SLT or ALT, of which 1 eye required additional medications and 2 eyes required glaucoma surgery.⁷ In another case series of 6 eyes with SE-IOP, 1 of 2 eyes that received SLT required glaucoma surgery.⁸ In a third case series of 21 eyes with SE-IOP, 1 eye was treated with an unspecified type of laser trabeculoplasty.¹¹ Lastly, one study that identified SE-IOP in 13 of 215 eyes reported that one of these 13 eyes initially responded to SLT treatment but later required medications to control IOP.¹⁰

In our exploratory study, SLT was primarily employed as an adjunctive therapy alongside medications to manage uncontrolled IOP in the setting of anti-VEGF injections. During follow-up, we observed substantial reductions in the mean IOP (29.5% at 6 months, 28.3% at 12 months, 30.3% at 24 months), comparable to those reported in prior studies.^18–23 Our absolute success rates—48.5% (95% CI: 30.8–64.1%) at 6 months, 32.3% (95% CI: 17.2–48.5%) at 12 months, and 23.6% (95% CI: 10.0–40.3%) at 24 months—were lower than those typically reported,²⁴ though they fall within the broader range observed in some previous studies.²⁵

The greater overlap between the upper bounds of our confidence intervals and previously reported success rates suggests these differences may reflect variability in study design and patient populations, rather than significant differences in SLT efficacy. For example, the landmark Laser in Glaucoma and Ocular Hypertension Trial (LiGHT) reported 74.2% of eyes achieved target IOP without medications at 36 months,²⁶ but enrolled only treatment naïve eyes, which exhibit better response to SLT,²¹ and did not consider repeat SLT a failure. It also excluded eyes with ocular comorbidities, including retinal ischemia, macular edema, diabetic retinopathy, and age-related macular degeneration,²⁷ all of which commonly warrant anti-VEGF injections. In contrast, the Optimal SLT Trial (OSLT), which included a more clinically representative sample and applied success criteria virtually identical to those of our study (requiring greater than 20% reduction in IOP without additional medical, laser, or surgical intervention), reported comparable 6-month success rates of 41.7% and 58.3% with standard and high energy SLT 360°, respectively.²⁸ Altogether, these findings suggest that SLT may be a reasonable therapeutic option for IOP elevation related to anti-VEGF injections.

Importantly, the long-term efficacy of SLT may be undermined by repeated anti-VEGF injections prior to SLT. In our study, the frequency of anti-VEGF injections prior to SLT was the only statistically significant predictor of SLT success. Each additional annual anti-VEGF injection was associated with a 31.8% increased risk of SLT failure (HR: 1.318; 95% CI: 1.091 to 1.593; P = 0.004), while holding the total number of anti-VEGF injections and sex constant in the multivariate Cox proportional hazards model. Notably, the total number of anti-VEGF injections was not found to be significant in the same multivariate analysis (HR: 1.037; 95% CI: 0.992 to 1.083; P = 0.106) despite being found to be significant in the univariate analysis (HR: 1.049; 95% CI: 1.006 to 1.094; P = 0.026). Our findings parallel those of a large study which found the highest number of anti-VEGF injections over a 6-month period, rather than the total number of anti-VEGF injections, to be the most significantly associated with the increased risk of SE-IOP.²⁹ Our findings also complement those of a case-control study which found a significantly greater associated risk of glaucoma surgery in eyes with 7 or more annual anti-VEGF injections compared to eyes with 3 or less annual anti-VEGF injections.¹⁷ Since glaucoma surgery is generally reserved for patients who have not responded to less invasive glaucoma interventions, such as SLT, the greater associated risk of glaucoma surgery in eyes with more frequent anti-VEGF injections suggested by this particular study may be potentially related to the greater risk of SLT failure with more frequent anti-VEGF injections suggested by our study.

Despite a lack of consensus, multiple studies have found the increased number^4,10,30–36 or decreased interval^5,17,29,37 between anti-VEGF injections to be potential risk factors for SE-IOP. Our findings further suggest that a higher injection frequency may be associated not only with the increased risk of SE-IOP development but also with reduced effectiveness of subsequent SLT intervention. This may reflect the clinical context of patients requiring more frequent injections—possibly due to more aggressive or refractory disease—who have underlying disease-related changes that limit SLT effectiveness. While the clinical significance of these interactions remains uncertain, they highlight important considerations for treatment planning, particularly when evaluating the likelihood of SLT success in patients with high anti-VEGF injection burden.

The pathophysiology of SE-IOP associated with anti-VEGF injections is not fully understood but is likely multifactorial. SLT may help lower IOP by enhancing trabecular outflow and facilitating clearance of obstructive debris. Larger anti-VEGF molecules may be more prone to accumulating within the trabecular meshwork, whereas smaller agents such as ranibizumab may pose a lower risk.^3–13 Additional contributors may include silicone oil microdroplet deposition from syringes and forward displacement of the lens–iris diaphragm in predisposed eyes, such as those with hyperopia or pseudoexfoliation.^1,2,16 However, we were unable to stratify outcomes by specific anti-VEGF agent, given that many patients received more than one type and our sample size was limited, highlighting the need for larger studies to explore this question. Nonetheless, our findings suggest that SLT may represent a useful adjunctive option in these patients, supporting its potential role in addressing the diverse mechanisms of IOP elevation in this setting. Furthermore, the frequency of injections itself may play a more critical role in predicting SLT outcomes. But this hypothesis warrants further investigation in larger, stratified cohorts.

One potential confounding variable unique to our study was that a significant portion of our eyes received anti-VEGF injections throughout their follow-up periods. However, receiving one or more anti-VEGF injections between the time of SLT and subsequent failure was not found to be a significant predictor of SLT success (HR: 0.709; 95% CI: 0.354 to 1.424; P = 0.334) in the univariate analysis. This suggests that the observed trend of IOP rebounding over time was unlikely related to independent incidences of SE-IOP in the context of anti-VEGF injections. While beyond the scope of this study, we also observed that all 3 of 33 (9.1%) eyes that received repeat SLT subsequently failed and required glaucoma surgery, contrary to previous studies suggesting the efficacy of repeat SLT.³⁸ These eyes received between 6.3 to 9.2 annual anti-VEGF injections, so perhaps in frequently injected eyes, there is an interplay by which the mechanisms behind SE-IOP resist the mechanisms promoting the long-term efficacy of SLT.

We did not find other variables, including age, sex, lens status, central corneal thickness, treatment diagnosis, and baseline IOP to be significant predictors of SLT success. Our results are generally consistent with those of previous studies with the exception of baseline IOP.^39–43 Higher baseline IOP may indicate more severe impairment or dysfunction of the trabecular meshwork, and in such cases, SLT may not be sufficiently effective to restore normal trabecular meshwork function. Regarding anti-VEGF therapy, we found no association between the type of anti-VEGF agent or the underlying indication for its use and SLT success. This is noteworthy given that some studies have reported significant differences in the risk of SE-IOP among conditions like age-related macular degeneration, retinal vein occlusion, and diabetic macular edema,^30,44 as well as among agents such as aflibercept, bevacizumab, and ranibizumab.^30,45–47 The clinical relevance of these variables remains uncertain and warrants further investigation.

The results of our study should be interpreted with several caveats in mind. First, based on our selection criteria, we presumed a priori that our sample eyes received SLT for SE-IOP associated with anti-VEGF injections. It is difficult to pinpoint the etiology of elevations in IOP to anti-VEGF injections, and it is possible that elevations in IOP were secondary to other pathological processes, such as the natural progression of glaucomatous disease in eyes with preexisting glaucoma, which several studies have also found to be associated with the increased risk of SE-IOP.^10,44,46,48 Although universal criteria are lacking, SE-IOP generally includes eyes with previously within-normal-limit IOP that experience elevations above 21 mmHg or rises of 5 to 6 mmHg or more from baseline over two or more consecutive visits following 3 or more anti-VEGF injections.² Nearly all eyes in our study experienced SE-IOP as defined in the existing literature, and the exclusion of eyes that do not meet the aforementioned criteria from analyses do not meaningfully impact the statistical significance of the data presented in our study.

Second, we analyzed the frequency of anti-VEGF injections based on the mean annual number of anti-VEGF injections between the time of the first anti-VEGF injection and SLT. This calculated frequency of anti-VEGF injections was not representative of the actual injection frequency by the time of SLT for eyes that had not received anti-VEGF injections for extended periods of time prior to SLT. To minimize false positive selection of eyes that experienced elevations in IOP unrelated to anti-VEGF injections, we excluded eyes that received SLT more than 24 months following the discontinuation of anti-VEGF injections. This cutoff still ensured the adequate inclusion of eyes that, for instance, received SLT to reduce medication burden in the setting of maximally tolerated medical therapy related to SE-IOP associated with anti-VEGF injections. Our reasoning appears to be supported by the results of our univariate analysis, which found no significant difference in the associated risk of SLT failure between eyes that received their most recent anti-VEGF injection within 3 months of, or between 3 to 24 months prior to, SLT (HR: 1.561; 95% CI: 0.632 to 3.856; P = 0.334).

Besides limitations inherent to retrospective observational studies, our study was most restricted by a small sample size. This was expected, as SE-IOP associated with anti-VEGF injections is a relatively rare phenomenon and only a percentage of the eyes requiring treatment receive SLT, perhaps due to the lack of published data for this indication. Thus, we were limited in the scope of our analysis. Consequent univariate screening due to the risk of otherwise overfitting increased the likelihood of residual confounding from omitted-variable bias. Additionally, the heterogeneity and complexity of the patient population receiving anti-VEGF injections posed challenges. Ideally, a control group of fellow uninjected eyes undergoing SLT for open-angle glaucoma or ocular hypertension unrelated to anti-VEGF injections would have been available for comparison.

Conclusion

Our study provides the first insight into the potential efficacy and predictors of SLT success in managing SE-IOP associated with anti-VEGF injections. These findings highlight the importance of periodic IOP monitoring in patients receiving anti-VEGF therapy. SLT may serve as an effective adjunctive treatment, potentially lowering IOP by enhancing trabecular outflow and clearing obstructive debris. The frequency of prior anti-VEGF injections may also be an important predictor of SLT outcomes. While additional studies are needed to clarify the underlying mechanisms and to evaluate outcomes by specific anti-VEGF agents, our results support the role of SLT as a valuable option in this challenging clinical scenario.

Funding Statement

This study was supported in part by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health (T35DK10468 for K.J.). The funding organization had no role in the design or conduct of this research.

Disclosure

R.A. has served on the Scientific Advisory Board for Janssen. The authors report no other conflicts of interest in this work.