Risk of Ocular Hypertension With Intravitreal Triamcinolone Compared to the Dexamethasone Implant

Abstract

Purpose: To determine whether there is a difference in the risk of ocular hypertension between patients in a large United States healthcare database who received an intravitreal dexamethasone implant and those who received intravitreal (IVT) triamcinolone acetonide injections. Methods: A retrospective cohort study was conducted from 2011 to 2022. Cohorts were created by identifying patients who were treated for the first time with either dexamethasone implants or IVT triamcinolone. The primary outcome was the number of patients requiring treatment for ocular hypertension. Cox proportional hazards regression modeling was performed for risk analyses, and inverse proportional treatment weighting was included. A sensitivity analysis used a new diagnosis of glaucoma or suspected glaucoma as the outcome. Results: In total, 147 (8.8%) of 1665 patients treated with dexamethasone implants and 282 (7.3%) of 3866 patients treated with IVT triamcinolone received medical or surgical therapy for ocular hypertension. In hazards models with inverse proportional treatment weighting of propensity scores, cohorts were found to be balanced in all covariates (standardized mean difference <0.1 for each variable). Cox proportional hazards regression analysis showed no difference in risk of requiring ocular hypertension treatment in the group treated with dexamethasone implants compared with the IVT triamcinolone group in the postinjection period of 1 to 40 days (hazard ratio [HR], 0.77, 95% CI, 0.53–1.12; P = .17). However, analysis after 40 days showed that treatment with dexamethasone implants, as compared with IVT triamcinolone, conferred an increased risk of requiring ocular hypertension treatment (HR, 1.41, 95% CI, 1.08–1.86; P = .01). Sensitivity analyses showed a similar, but nonsignificant risk of progressing to a diagnosis of glaucoma or suspected glaucoma in the dexamethasone implant group compared with the IVT triamcinolone group at 40 days after injection (HR, 1.24, 95% CI, 0.91–1.70; P = .18). Conclusions: Treatment with dexamethasone implants was associated with an increased risk of requiring ocular hypertension treatment when compared with IVT triamcinolone treatment. Clinicians should be aware of this increased risk when counseling patients regarding their treatment options.

Keywords

ocular hypertension dexamethasone implant triamcinolone steroid induced glaucoma

Introduction

Intraocular steroid therapy has been an essential part of the treatment armamentarium for disease processes associated with various retinal vascular and uveitis etiologies. Well-known side effects of steroid therapy include ocular hypertension and cataract formation. Ocular hypertension is defined as an intraocular pressure (IOP) of >25 mm Hg or an increase in IOP of at least 10 mm Hg from baseline. Steroid-induced glaucoma has been postulated to occur as a result of decreased outflow at the trabecular meshwork due to physical impedance, toxic effects on trabecular meshwork cells, and upregulation of the myocilin gene.¹ It is estimated that approximately 40% of the United States general population are steroid responders.² Generally, steroid-induced IOP spikes are transient and will reverse with cessation of steroid use but can lead to worsening of glaucomatous optic neuropathy.³ Therefore, it is important for the clinician to be able to identify and counsel those patients at highest risk of developing complications.

Intravitreal (IVT) triamcinolone acetonide and dexamethasone implants are 2 of the most common treatment modalities for intraocular steroid delivery. In phase III clinical studies, the most common adverse effects of the dexamethasone implant were increased IOP, with 32% of patients having a posttreatment IOP of >25 mm Hg and 41.5% having concurrent use of IOP-lowering medication.⁴ Mean IOP increases occurred around 6 weeks. In contrast, in a meta-analysis of 3654 eyes treated with IVT triamcinolone, 32.1% of eyes (95% CI, 28.2–36.3) receiving the 4 mg/0.1 mL dose developed ocular hypertension.² That same meta-analysis also showed that ocular hypertension was observed in 15.3% of eyes (95% CI, 9.2–24.3) following treatment with 0.7 mg dexamethasone implant. Although no direct comparison was made, the dexamethasone implant was thought to be associated with fewer increases in IOP compared with IVT triamcinolone.² Another study, again without directly comparing the 2 steroid therapies, demonstrated results that concurred with those findings, suggesting that dexamethasone implants would have fewer IOP-related issues due to its reduced lipophilic tendency and the lower likelihood of physical obstruction of the trabecular meshwork from triamcinolone crystals.¹

Despite their regular use in clinical practice, the relative adverse effect profiles of IVT triamcinolone and dexamethasone implants remain poorly characterized. The goal of this study was to determine the risk of needing treatment for ocular hypertension after use of IVT triamcinolone compared with a dexamethasone implant.

Methods

Database

De-identified data were abstracted from the Optum Clinformatics Data Mart Database. The database contains all outpatient medical claims (office visits, procedures, and medications given) as well as demographic data and some laboratory values for all patients enrolled in commercial and Medicare Advantage insurance plans. The subset of data available for this study included all patients in the database from January 1, 2009, to June 30, 2022.

Cohorts

For the study, we identified all patients who initiated treatment with either IVT triamcinolone acetonide or a dexamethasone implant after January 1, 2011. The study evaluation period was started in 2011 due to the availability in that year of the dexamethasone implant after US Food and Drug Administration approval. Two cohorts were created based on the first type of treatment received, either IVT triamcinolone or dexamethasone implant. The earliest date of the incident injection was considered the index date. Patients were excluded for having any prior diagnosis of ocular hypertension, suspected glaucoma, or glaucoma or for receiving glaucoma-related procedures or prescriptions prior to the index date (see Supplemental Table 1 for the full list of International Classification of Diseases, Ninth or Tenth Revision codes and Current Procedural Terminology codes used during this study). Patients were also required to have at least 2 years of uninterrupted time within the data set prior to the index date. To further limit the occurrence of secondary glaucoma, all patients before or after the index date who had been diagnosed as having neovascular glaucoma, had received systemic steroids within 90 days of the index date, or had received a fluocinolone or retisert implant were excluded.

Covariates and Inverse Probability of Treatment Weighting

To reduce the effects of confounding due to potential differences in the baseline characteristics between the 2 cohorts, propensity scores were calculated using multivariable logistic regression. Covariates included in the model were age, sex, race, education level, household income, geographic region of the country, index year, smoking status, number of healthcare visits within 1 year prior to the index date, history of uveitis, hypertension, hypercholesterolemia, chronic kidney disease, ischemic heart disease, ischemic stroke, heart failure, peripheral arterial disease, and coronary artery bypass graft, and history of use of steroid drops and sub-Tenon triamcinolone in the 120 days prior to the index date. Inverse probability of treatment weighting with the propensity scores was used in the subsequent analyses.

Outcome of Interest

The primary outcome of interest was progression to ocular hypertension, which was defined as the filling of a prescription for a glaucoma medication or having a glaucoma procedure. Because not all ocular hypertension leads to a diagnosis of glaucoma, a sensitivity analysis anticipating fewer outcomes was used in which the outcome was alternatively defined as having a new diagnosis of glaucoma or suspected glaucoma. Both outcomes have been used previously.^5–7

Statistical Analysis

Cox proportional hazards regression modeling with inverse probability of treatment weighting was performed to compare the IVT triamcinolone and dexamethasone implant cohorts. Censoring occurred at 180 days after the index date, when a patient left the insurance plan, when a patient received the other form of steroid treatment (dexamethasone implant in the IVT triamcinolone cohort, and vice versa), or when a patient received another form of intraocular steroid. Patients were allowed to have repeat treatments of the same medication up to the 180-day censoring limit. During the running of the model, it was found that the proportional hazards assumption was violated at 40 days post-index date. Therefore, time-varying Cox models were used in the final analysis for analyses of both the primary outcome and the sensitivity analysis outcomes, with results reported as the hazard ratio (HR) for requiring treatment for ocular hypertension over the first 40 days and then separately for days after day 40.

All statistical analyses were performed using SAS software version 9.4 (SAS Institute Inc.). Statistically significant differences were defined as P values (2-tailed) less than or equal to .05.

Results

Figure 1 details the distribution of patients included and excluded based on each criterion used to devise our cohorts. Table 1 shows the baseline characteristics of our patients both before and after inverse probability of treatment weighting of propensity scores. The 2 cohorts showed good balance in all covariates (standardized mean difference <0.1 for each variable).

Figure 1.

Flow chart showing the distribution of patients in a large US healthcare database who met the inclusion and exclusion criteria for the study and were assigned to receive either intravitreal dexamethasone implant or IVTA injections. Abbreviations: NVG, neovascular glaucoma; IVTA, intravitreal triamcinolone acetonide.

Table 1.

Baseline Characteristics of Patients in Each Treatment Cohort Before and After Inverse Probability of Treatment Weighting of Propensity Scores.

	Before Weighting			After Weighting
Characteristic	Dexamethasone Implant(n = 1665)	Intravitreal Triamcinolone(n = 3866)	Standardized Mean Difference	Dexamethasone Implant(n = 1665)	Intravitreal Triamcinolone(n = 3866)	Standardized Mean Difference
Age, y, mean (SD)	68.9 (12.5)	70.1 (12.4)	0.1106	69.7 (12.8)	69.7 (12.7)	−0.0065
Sex, %			0.0207			−0.0057
Female	51	52		52	51
Male	49	48		48	48
Race, %			0.2122			0.0572
White	68	74		72	72
Asian	3	3		3	3
Black	11	8		10	9
Hispanic	15	10		11	11
Unknown	4	4		4	4
Education level, %			0.0663			0
Bachelor’s degree plus	13	15		14	14
Less than Bachelor’s degree	57	57		57	57
Less than or equal to HS diploma	27	25		26	26
Unknown	3	3		2	3
Household income, %			0.0412			0.0459
$100K+	21	20		21	21
$40K–$49K	9	9		10	9
$50K–$59K	9	10		9	10
$60K–$74K	12	12		12	12
$75K–$99K	16	15		16	16
<$40K	27	27		27	27
Unknown	6	6		6	6
Geographic location,%			0.2961			0.06
Mountain	17	11		13	13
Northeast	10	9		9	10
Pacific	9	15		13	13
South Atlantic	21	19		20	20
Southern Midwest	21	18		18	19
Upper Midwest	21	28		26	26
Unknown	0	0		0	0
Index year, %			0.5224			0.0676
2011	3	6		5	5
2012	3	10		7	7
2013	5	7		7	7
2014	3	7		6	6
2015	8	11		9	10
2016	12	14		13	14
2017	16	15		15	15
2018	13	9		11	10
2019	11	6		8	8
2020	6	4		5	5
2021	11	7		8	8
2022	10	4		6	6
Hypertension, %	85	82	−0.0756	83	83	0.0067
Hypercholesterolemia, %	87	83	−0.0944	83	84	0.0203
Kidney disease, %			0.3083			0.0215
CKD	44	34		36	37
ESRD	14	9		11	11
No disease	42	57		54	53
Smoking, %	33	30	−0.0628	30	31	0.0176
Ischemic heart disease, %	37	34	−0.0686	35	35	0.0084
Ischemic stroke, %	20	21	0.0239	21	21	0.0008
Heart failure, %	19	17	−0.0553	17	18	0.0125
Peripheral arterial disease, %	25	25	−0.0065	24	25	0.0118
Diabetes mellitus, %	66	54	−0.2326	55	57	0.0406
Intracranial hemorrhage, %	1	1	−0.0181	1	1	−0.0083
Chronic liver disease, %	1	1	−0.0341	1	1%	0.002
Chronic pulmonary disease, %	20	18	−0.0408	18	19	0.0174
Peripheral vascular disease, %	26	26	−0.0008	26	26	0.0062
Atrial fibrillation/flutter, %	16	15	−0.0283	16	15	−0.0231
Congestive heart failure, %	21	19	−0.0437	19	20	0.0109
Myocardial infarction, %	12	13	0.0236	11	12	0.0361
Arrhythmia, %	25	28	0.0527	28	27	−0.0107
Coronary artery bypass graft, %	2	2	−0.0142	2	2	0.0198
Uveitis, %	12	9	−0.1002	10	10	−0.0092
History of ocular steroid use, %			0.2529			0.0709
Dexamethasone	1	2		2	2
Difluprednate	0	0		0	0
Fluorometholone	0	0		0	0
Loteprednol	0	0		0	0
Prednisolone	11	18		14	16
Other	2	5		4	4
None	85	75		79	78
Health care visits in previous year, mean (SD)	12.7 (7.4)	12.1 (7.5)	−0.0959	12.1 (7.2)	12.3 (7.5)	−0.0031

Abbreviations: CKD, chronic kidney disease; ESRD, end-stage renal disease; HS, high school.

A total of 147 (8.8%) of 1665 patients receiving the dexamethasone implant and 282 (7.3%) of 3866 patients receiving IVT triamcinolone injections received medical or surgical therapy for ocular hypertension. A Kaplan-Meier curve for the cumulative probability of requiring ocular hypertension treatment demonstrated a violation of the proportional hazards assumption at 40 days (Figure 2A). The Cox proportional hazards regression analysis for the period of 1 to 40 days showed no significant difference in the risk of requiring ocular hypertension treatment in the dexamethasone implant group compared with the IVT triamcinolone group (HR, 0.77, 95% CI, 0.53–1.12; P = .17) (Table 2). However, analysis after day 40 showed that receiving the dexamethasone implant cohort had an increased risk of requiring ocular hypertension treatment compared with the IVT triamcinolone cohort (HR, 1.41, 95% CI, 1.08–1.86; P = .01).

Figure 2.

Kaplan-Meier curves showing (A) the cumulative probability of requiring ocular hypertension treatment (the primary outcome), and (B) the cumulative probability of the sensitivity outcome of having a new diagnosis of glaucoma or suspected glaucoma, in patients receiving either intravitreal dexamethasone implant or intravitreal triamcinolone acetonide injections.

Table 2.

Time-Varying Hazard Ratio Multivariate Analysis of Primary Outcome and Sensitivity Analysis.

		<=40 days			> 40 days
		N of events	Hazard ratio (95% CI)	P Value	N of events	Hazard ratio (95% CI)	P Value
Primary Outcome	Intravitreal triamcinolone	113	Ref	.2377	169	Ref	.0116
Primary Outcome	Dexamethasone implant	43	0.80 (0.55-1.16)	.2377	106	1.45 (1.10-1.89)	.0116
Sensitivity Analysis	Intravitreal triamcinolone	134	Ref	.2432	149	Ref	.6722
Sensitivity Analysis	Dexamethasone implant	66	1.23 (0.87-1.73)	.2432	72	1.11 (0.80-1.53)	.6722

Sensitivity analysis showed that 136 (8.2%) of 1665 patients receiving the dexamethasone implant and 283 (7.3%) of 3866 patients receiving IVT triamcinolone were given a diagnosis of glaucoma or suspected glaucoma. Figure 2B shows the Kaplan-Meier curve for the cumulative probability of the sensitvity outcome. Cox proportional hazards regression analysis showed that the risk of progression to a diagnosis of suspected glaucoma or glaucoma after 40 days postinjection was similar in the dexamethasone implant group compared with the IVT triamcinolone group, but this did not reach statistical significance (HR, 1.24, 95% CI, 0.91–1.70; P = .18).

Conclusions

Our study analyzed real-world data using a large US administrative medical claims database and showed an increased hazard for requiring ocular hypertension treatment 40 days or later after injection in the dexamethasone implant cohort when compared with the IVT triamcinolone cohort. These results contrast sharply with previous studies that have asserted that ocular hypertension was less common in patients receiving dexamethasone implants.^1,2 Our study confirms the findings of a small study comparing IVT triamcinolone to dexamethasone implants in vitrectomized eyes.⁸

The proportional hazards assumption was violated at 40 days (as shown in Figure 2), thus requiring the Cox regression analysis to be performed with a division at 40 days. While the need to divide the analysis into days 1 to 40 and after day 40 was not preplanned, this timing is consistent with clinical knowledge showing that IOP spikes frequently occur 6 weeks after injection of a dexamethasone implant.⁴ Typical practice following an intravitreal steroid injection is for the patient to be seen between 4 weeks and 8 weeks thereafter. One explanation for why the division may have occurred is that some physicians may have given IOP-reducing medications prophylactically at the time of the injection. Indiscriminate prescribing would create the appearance of different glaucoma medication needs during the first 40 days, as compared with physicians who waited to determine which patients were true steroid responders.

Another possible explanation for the increased risk of ocular hypertension in the dexamethasone implant group may be a longer duration of action or greater steroid potency, which may delay the onset of ocular hypertension when compared with IVT triamcinolone.^8,9 Although these assertions have not been proven in a direct comparison, even if true, it is important to note that no difference in hazard was noted between the 2 steroid treatments during the first 40 days of the analysis. It is likely that if dexamethasone implant were proven to be more potent, a higher rate of ocular hypertension would be expected, similar to the differences in ocular hypertension rates between the topical steroids difluprednate and prednisolone.¹⁰ It seems less likely that the duration of the medication effects could be attributed to the differences seen in our cohorts. If the timing of peak IOP was the primary cause of the difference in hazard found, then it would be expected that the IVT triamcinolone group would have a higher peak IOP than the dexamethasone implant group in the first 40 days, but that was not seen in our results.

In 2 studies in the literature, 6% to 17% of patients with diabetic macular edema (DME) required treatment with topical ocular hypotensives due to a steroid IOP response from a dexamethasone implant.^11,12 In a meta-analysis, 15% to 64% of patients with DME required medical treatment for ocular hypertension secondary to IVT triamcinolone.² A smaller minority of patients who failed medical treatment underwent laser trabeculoplasty or surgical intervention. The rates of patients requiring treatment for ocular hypertension in our study (8.9% of patients in the dexamethasone implant group and 7.3% of patients in the IVT triamcinolone group) were within these bounds, although on the lower side of the reported rates of previously published studies. It is possible that the outcome definition that we used may have impacted our results. Other studies have focused on an elevated IOP of ≥25 mm Hg or an increase in IOP of ≥10 mm Hg over baseline, whereas our study relied on the risk of requiring ocular hypertension treatment as the primary outcome. Some physicians may have found an elevated IOP of >10 mm Hg over baseline in their patients, but because of a healthy-appearing optic nerve and an impending drop in IOP with the dissipation of the steroid effect, they may have considered treatment unnecessary. Of note, our results did not appreciably change whether we used ocular hypertension treatments or glaucoma diagnoses as the outcome.

Limitations to our study design need to be acknowledged. First, due to the de-identified nature of the database, we were not able to chart review patient encounters to verify that the correct medication or diagnosis code was documented appropriately. Also, we are unable to verify IOP measurements or specific glaucoma diagnoses; instead we used the proxy outcome of initiating glaucoma medication or having a glaucoma procedure. Specific IOP measurements would be ideal, and these proxy measures have been used previously.^5,6 In addition, the data were obtained from a single US-based claims database, which may not be fully representative of trends found outside of the US or US patients without insurance. Furthermore, patients were not offered the 2 treatments in a randomized manner; however, the use of inverse probability of treatment weighting helps to balance the cohorts in baseline characteristics and limit the impact of confounding on our study. Lastly, due to the nature of the data set, we were unable to control for the specific indication requiring the steroid injection; however, we did control for history of uveitis and for history of diabetes at baseline, because these can result in ocular hypertension.

Clinicians need to account for all beneficial and potential adverse effects of the treatments they offer their patients. We found that patients who received dexamethasone implants had an increased hazard of requiring treatment for ocular hypertension compared with patients who received IVT triamcinolone. As more local steroid options become available, direct comparison of the benefits and risks between IVT triamcinolone, dexamethasone implant, fluocinolone acetonide implant, suprachoroidal steroid delivery, and other newer agents should continue to be made.

Supplemental Material

sj-docx-1-vrd-10.1177_24741264261447503 – Supplemental material for Risk of Ocular Hypertension With Intravitreal Triamcinolone Compared to the Dexamethasone Implant

Supplemental material, sj-docx-1-vrd-10.1177_24741264261447503 for Risk of Ocular Hypertension With Intravitreal Triamcinolone Compared to the Dexamethasone Implant by Apoorv P. Chebolu, Yineng Chen, Fangming Jin and Brian L. VanderBeek in Journal of VitreoRetinal Diseases

Footnotes

Author Note

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

ORCID iD

Brian L. VanderBeek

Ethical Considerations

The University of Pennsylvania has deemed all projects performed on the Optum database exempt from ethics review due to its de-identified nature. This study follows the Strengthening the Reporting of Observational Studies in Epidemiology guidelines for reporting cohort studies.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: University of Pennsylvania Core Grant for Vision Research (grant 2P30EYEY001583). Additional funding was provided by the Metzger Family, Research to Prevent Blindness, and the Paul and Evanina Mackall Foundation. Funding from each of these sources was received in the form of block research grants. None of the funding organizations had any role in the design or conduct of the study.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

Dr. VanderBeek had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Data can be made available through contracting directly with Optum Insights.

Statement of Informed Consent

Informed consent was waived due to the de-identified nature of the data set.

Supplemental Material

Supplemental material is available online with this article.

References

Sharma

Kuppermann

Bandello

, et al. Intraocular pressure (IOP) after intravitreal dexamethasone implant (Ozurdex) amongst different geographic populations-GEODEX-IOP study. Eye (Lond). 2020;34:1063-1068. doi:10.1038/S41433-019-0616-7

Kiddee

Trope

Sheng

, et al. Intraocular pressure monitoring post intravitreal steroids: a systematic review. Surv Ophthalmol. 2013;58:291-310. doi:10.1016/J.SURVOPHTHAL.2012.08.003

Armaly

. Effect of corticosteroids on intraocular pressure and fluid dynamics: I. The effect of dexamethasone* in the normal eye. Arch Ophthalmol. 1963;70:482-491. doi:10.1001/ARCHOPHT.1963.00960050484010

Boyer

Yoon

Belfort

, et al. Three-year, randomized, sham-controlled trial of dexamethasone intravitreal implant in patients with diabetic macular edema. Ophthalmology. 2014;121:1904-1914. doi:10.1016/J.OPHTHA.2014.04.024

Kim

Addis

Pan

VanderBeek

BL.

Comparative effectiveness of generic latanoprost versus branded prostaglandin analogs for primary open angle glaucoma. Ophthalmic Epidemiol. 2019;26:63-71. doi:10.1080/09286586.2018.1516786

Cui

Gray

VanderBeek

BL.

Repeated intravitreal injections of antivascular endothelial growth factors and risk of intraocular pressure medication use. Graefes Arch Clin Exp Ophthalmol. 2019;257:1931-1939. doi:10.1007/S00417-019-04362-7

Sterling

Hua

Dunaief

Cui

VanderBeek

BL.

Glucagon-like peptide 1 receptor agonist use is associated with reduced risk for glaucoma. Br J Ophthalmol. 2023;107:215-220. doi:10.1136/BJOPHTHALMOL-2021-319232

Wakabayashi

Liu

Anderson

, et al. A comparison of ocular complications after 0.7 mg dexamethasone implant versus 2 mg of intravitreal triamcinolone in vitrectomized eyes. Ophthalmol Retin. 2023;7(6):480-488. doi:10.1016/J.ORET.2022.12.009

Ozkok

Saleh

Sigford

Heroman

Schaal

Comparison of ozurdex and triamcinolone acetonide for refractory cystoid macular edema in retinal vein occlusion. Retina. 2015;35:1393-1400. doi:10.1097/IAE.0000000000000475

10.

Slabaugh

Herlihy

Ongchin

Van Gelder

RN.

Efficacy and potential complications of difluprednate use for pediatric uveitis. Am J Ophthalmol. 2012;153:932-938. doi:10.1016/J.AJO.2011.10.008

11.

Boyer

Faber

Gupta

, et al. Dexamethasone intravitreal implant for treatment of diabetic macular edema in vitrectomized patients. Retina. 2011;31:915-923. doi:10.1097/IAE.0B013E318206D18C

12.

Haller

Kuppermann

Blumenkranz

MS.

Randomized controlled trial of an intravitreous dexamethasone drug delivery system in patients with diabetic macular edema. Arch Ophthalmol. 2010;128:289-296. doi:10.1001/ARCHOPHTHALMOL.2010.21

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.01 MB

0.00 MB