Inhaled corticosteroid and secondary glaucoma: A meta-analysis of 18 studies

Authors: Ishii M, Horita N, Takeuchi M, Matsumoto H, Ebina-Shibuya R, Hara Y, Kobayashi N, Mizuki N, Kaneko T

Geographical coverage: United States of America (USA), Canada, Finland, United Kingdom (UK), Saudi Arabia, China, Netherlands, Denmark, Turkey, Australia, India, and Croatia

Sector: Corticosteroid

Sub-sector: Risk of glaucoma and intraocular pressure

Equity focus: Not reported

Study population: Patients with glaucoma

Review type: Epidemiology (risk association) review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background

Asthma and chronic obstructive pulmonary disease (COPD) are chronic airway diseases that cause airflow obstruction and inflammation. Inhaled corticosteroids (ICSs) are a key treatment for asthma and asthma-COPD overlap (ACO), and a treatment option for COPD, providing localised anti-inflammatory effects. However, prolonged high-dose ICS use has been associated with potential adverse effects, including ocular conditions such as glaucoma and increased intraocular pressure (IOP). Although newer studies have largely refuted this risk, there is still concern regarding ICS use in children with respiratory conditions.

Objectives

To clarify whether ICS increases the risk of glaucoma and elevates IOP.

Main findings

Overall, the review found no significant evidence supporting a positive association between ICS use and glaucoma or IOP, and suggested that ICS use might be associated with a lower incidence of glaucoma due to indirect effects.

The search identified 623 articles, of which 18 studies comprising 31,665 patients were included in this quantitative analysis. The included studies encompassed a variety of study designs: seven randomised controlled trials (RCTs), four prospective cohort studies, one retrospective cohort study, two nested case–control studies, and four cross-sectional studies. Sample sizes varied significantly, ranging from 22 to 13,312 participants. Of the 18 studies, six were conducted in the United States of America (USA), two in Canada, and one each in Finland, United Kingdom (UK), Saudi Arabia, China, Netherlands, Denmark, Turkey, Australia, India, and Croatia.

The meta-analysis showed no significant differences between ICS users and non-users in the following:
– Crude glaucoma incidence (OR = 0.95; 95% CI: 0.86 to 1.04; P = 0.26)
– Adjusted glaucoma incidence (OR = 0.90; 95% CI: 0.65 to 1.24; P = 0.64)
– Crude prevalence (OR = 1.82; 95% CI: 0.23 to 14.19; P = 0.57)
– Adjusted prevalence (OR = 1.22; 95% CI: 0.50 to 2.96; P = 0.66)
– IOP change during ICS treatment (mean difference [MD] = +0.01 mmHg; 95% CI: −0.19 to 0.20; P = 0.95)
– Single IOP measurement (MD = +0.37 mmHg; 95% CI: −0.24 to 0.97; P = 0.23)

However, a time-to-event analysis for glaucoma development as a secondary endpoint (adjusted hazard ratio = 0.52; 95% CI: 0.28 to 0.96) suggested a potential reverse association between ICS use and glaucoma risk.

Methodology

Two reviewers conducted systematic searches in Medline, Embase, Cochrane, and the Web of Science Core Collection on 10 May 2020 to identify studies comparing the incidence and prevalence of glaucoma between ICS users and non-users. No language restrictions were applied, and non-English articles were translated by external translators. Reference lists of eligible studies and relevant reviews were scanned for additional studies. Authors of original studies were contacted when necessary. Two reviewers also conducted manual searches.
Identified articles were screened, and the methodological quality of included studies was assessed using the Newcastle–Ottawa Scale. Two reviewers independently extracted the data. Findings were synthesised using a random-effects model meta-analysis. A fixed-effects model and the leave-one-out method were used in sensitivity analyses. Heterogeneity was assessed using I² statistics, and publication bias was evaluated using the Begg–Kendall test.

Applicability / External validity

The review did not explicitly discuss the generalisability of its findings.

Geographic focus

The included studies were conducted in a range of countries including the USA, Canada, Finland, UK, Saudi Arabia, China, Netherlands, Denmark, Turkey, Australia, India, and Croatia.

Summary of quality assessment

Overall, there is low confidence in the conclusions drawn from this study. The authors conducted thorough literature searches, avoiding publication and language bias. They used appropriate methods to assess the risk of bias in the included studies and to synthesise the data. However, they did not report the methods used for study screening, and the results were not analysed or reported according to the risk of bias.