Using the IRIS Registry, investigators examined changes in IOP after intravitreal injections of anti-VEGF drugs.

Study design

Researchers identified 23,776 unique patients who received either bevacizumab, aflibercept, or ranibizumab out of a database comprising 34 million patients in the United States. Patients were divided into groups that received at least 12, 18, or 25 injections of an anti-VEGF agent. Fellow untreated eyes were used for comparison.

Outcomes

Across all groups and after a minimum of 1 year of follow-up, the IOP decreased from baseline by a mean of 0.9 mm Hg in treated eyes compared with a mean decrease of 0.2 mm Hg in fellow untreated eyes.

Clinically significant IOP increases (sustained rise of at least 6 mm Hg to an IOP greater than 21 mm Hg) occurred in 1.9%, 2.8%, and 2.8% of eyes on aflibercept, ranibizumab, and bevacizumab, respectively. This was significantly higher than untreated fellow eyes treated with bevacizumab or ranibizumab, but not with aflibercept. Investigators hypothesize that this may be attributable to aflibercept’s affinity for placental growth factor, which could impact trabecular meshwork function.

Limitations

Some potential shortcomings of this study include its retrospective nature and the variability in clinical documentation and IOP measurement. The authors did not assess other direct factors of glaucoma risk or progression, such as visual field, use of IOP-lowering drugs, or thickness of the retinal nerve fiber layer.

Clinical significance

Big database studies are well powered to detect small differences, with intrinsic attributes that can make them more representative of real-world practices.

This large study using data from the IRIS Registry ties intravitreous anti-VEGF injections to a small yet statistically significant decrease in IOP over time. A small proportion of patients, about 2.6% on average, experienced a sustained rise in IOP on these drugs, compared with 1.5% of untreated fellow eyes. This rise, however, was not seen in aflibercept-treated eyes.