A nested case-control study has found an association between current use of SSRIs/venlafaxine and cataracts.

The cohort comprised residents of Quebec aged ≥ 65 years who had undergone coronary revascularisation between 1995 and 2004. Clinical and prescribing data were captured prospectively by linked databases.

Cases were defined as those receiving a new diagnosis of cataract from a consultant ophthalmologist. Ten controls, matched for age and duration of follow-up, were selected randomly for each case. A total of 18,784 cases and 187,840 controls were included. There was a significant increase in cataract risk for current users of SSRIs/venlafaxine, which persisted after adjustment for gender, and use of corticosteroids, statins antihypertensives, or antidiabetic drugs; adjusted rate ratio 1.15 (95% CI: 1.08-1.23).

The risk was highest with fluvoxamine RR 1.39 (95% CI: 1.07-1.80) and venlafaxine RR 1.33 (95% CI: 1.14-1.55). Average time from initiation of antidepressants to diagnosis of cataract was 656 days.

In nested case-control studies, selection of both cases and controls from the same cohort avoids the problem of selection bias, and prospective data collection avoids the problem of recall bias.1 This study has an impressively large sample size, although this may not have been sufficient to provide reliable data on individual antidepressants.

It was not possible to adjust for all known confounders. In particular, smoking increases the risk for both the development of depression and cataracts. Other potential confounders are alcohol use, diet and obesity.2

There are serotonin receptors in the lens, and serotonin inhibits an enzyme thought to be crucial in the maintenance of lens transparency.3

If confirmed, the 15% increased risk of cataract development is a further potential hazard of SSRI use in elderly patients. Other important risks are upper GI bleeding, hyponatraemia and fragility fracture. The absolute risk of serious bleeding is small but clinically significant in patients > 80,4 elderly patients are more likely to develop SSRI-induced SIADH and the risk of fragility fracture may be doubled in patients > 50.5

We need to weigh very carefully the risks against the potential benefits before prescribing SSRIs for elderly patients; for those with mild to moderate depression, a psychosocial intervention may well be preferable.

Dr Phillip Bland