Cataracts More Likely in Patients Using Insulin, Other Medications

An investigation analyzing the Centers for Disease Control and Prevention database — National Health and Nutrition Examination Survey (NHANES) — demonstrates relationships between 8 therapeutic agents and cataracts, according to a report published in the American Journal of Ophthalmology. The research on associations between systemic drugs and risk for cataracts has been inconsistent, in part due to a limited array of medications examined. 

Investigators merged data from the NHANES for the years 1999 to 2008. After excluding participants whose records were missing information, the cohort numbered 14,931, with an average age of 56.8±0.2 years. Cataracts were defined as a lens needing surgery, and prescriptions were queried for the 30 days before the survey. Drugs used by fewer than 0.5% of individuals, and ophthalmic agents were excluded.

Of 329 drug categories, the first statistical model found 19 positively associated with cataracts — after correction for demographic factors of gender, age, ethnicity, income, and education level. The logistic regression model also identified 1 category negatively associated; sex hormone combinations for women (P =.010). The 19 showing positive correlations comprised glucocorticoids, minerals/electrolytes, and medications indicated for groups of illnesses.

After a second model adjusted for comorbidities and smoking, several categories persisted; 7 positive and 1 negative association, including the following agents:

1. Minerals and electrolytes (P =2.12×10^-4) or (P =.000212)
2. SSRI antidepressants (P =.008)
3. Tricyclic antidepressants (P =.001)
4. Insulin (P =9.41×10^-5)
5. Calcium channel blockers (P =.010)
6. Loop diuretics (P =1.06×10^-5)
7. Group III antiarrhythmic agents (P =.004)
8. Sex hormone combinations for women (P =5.98×10^-5), protective

All 8 categories exhibited dose-response correlations. Investigators caution that despite statistical correction for comorbid diseases, the impact on cataract formation may still be related to residual confounding factors; for example high intraocular pressure with antidepressants.

Sensitivity analysis for drugs prescribed in 1% or more of the sample revealed that all except group III antiarrhythmic agents retained a significant association. However, narcotic analgesics showed a sizable connection, odds ratio 1.53 (95% CI, 1.09-2.13; P =.014). Researchers used a second sensitivity model for meds taken for more than 30 days, finding similar results to the first model.

Prior studies have not reported mineral/electrolyte findings, but have discovered risks connected to high serum sodium. Other analyses exploring antihypertensives do not agree on cataract risk. Current investigators speculate calcium channel blockers may affect calcium homeostasis, central to lens transparency. 

“Of note, after further adjustment for smoking status and diabetes mellitus, only the association between insulin use and cataract remained significant, implying that the associations between sulfonylurea or biguanide use and cataract were mainly confounded by diabetes mellitus status,” the study authors explain. They add that insulin may create an “unfolded protein response” and death of lens epithelial cells — or is prescribed for uncontrolled disease.

Limitations of this investigation included lack of records for cataract subtype, duration and severity of underlying conditions, potential patient non-compliance with drug regimen, or index date for surgery and prescription order. Excluded individuals may not have sought medical care. Strengths involved a large, population-based sample and a 2-phase statistical correction.