Findings from a recent study published in JAMA Ophthalmology evaluated the association between the use of curcumin-based nutritional supplements (CBNS) and the risk of the development or progression to age-related macular degeneration (AMD).
Give me some background.
Curcumin is the main component of turmeric (Curcuma longa), a flowering plant in the ginger family commonly used in South Asia that is known to possess anti-inflammatory and antioxidant properties.
Previous studies have demonstrated that curcumin can protect the retinal pigment epithelium (RPE) from oxidative stress and damage induced by inflammation.
Now talk about the study.
In this retrospective cohort study, investigators aggregated data from the electronic health record (EHR) research network TriNetX from 2003 to 2024.
Propensity score matching (PSM) was performed to control for baseline demographics and medical comorbidities—notably, patients without AMD were included in the study before PSM, including those taking and not taking CBNS.
Then: Patients with no history of AMD were stratified by instances of CBNS prescription records, while patients with a history of early nonexudative AMD were stratified by the number of CBNS prescription records.
Main outcomes measures?
These included the relative risk (RR) of developing:
- Nonexudative AMD
- Exudative AMD
- Advanced nonexudative AMD
- Geographic atrophy (GA)
- Blindness
- Requiring intravitreal anti-vascular endothelial growth factor (VEGF) therapy
Findings?
In total, investigators included 66,804 patients (mean age: 64.9 years, 66.1% female) taking CBNS and 1,809,440 patients (mean age: 67.0 years, 55.2% female) not taking CBNS.
Compared to matched patients not taking CBNS, in patients without a history of AMD aged 50 years or older, CBNS use was associated with lower rates of developing:
- Nonexudative AMD (RR: 0.23, 95% confidence interval [CI] 0.21-0.26, P<0.001)
- Advanced nonexudative AMD or GA (RR: 0.11, 95% CI 0.07-0.17, P<0.001)
- Exudative AMD (RR: 0.28, 95% CI 0.24-0.32, P<0.001)
- Blindness (RR: 0.46, 95% CI 0.36-0.59, P<0.001)
- Requiring intravitreal anti-VEGF therapy (RR: 0.15, 95% CI 0.13-0.17, P<0.001)
These results were consistent in patients aged 60 and 70 years or older.
Anything else?
In participants with early nonexudative AMD: Subsequent instances of CBNS prescription records were associated with lower rates of developing advanced nonexudative AMD or GA (RR: 0.58, 95% CI 0.41-0.81, P<0.001) when compared with matched patients without a CBNS prescription record.
Of note: CBNS did not affect the risk of developing exudative AMD.
Expert opinion?
A commentary also published in JAMA Ophthalmology highlighted that the magnitude of the effect of CBNS (RR 0.23) was roughly equal to the protective effect of not smoking (RR 0.26).
Commentary author Brian VanderBeek, MD, MPH, MSCE, added that the protective association observed with CBNS may have occurred due to bias in observational studies related to patient behaviors called the healthy user effect.
Tell me more.
The healthy user effect suggests that people who are health conscious (i.e., take vitamin supplements, get regular checkups, take prescriptions as directed, etc.) typically fare better healthwise than those who do not or are unable to.
As such, patients who have sufficient disposable income to buy supplements have a higher chance of being able to afford other healthy lifestyle adjustments and are more likely to be motivated to partake in activities that improve their health.
I’m sensing a but …
Despite potential confounding from the healthy user effect, the magnitude of CBNS’ effect was sufficient to indicate that a true reduction in the risk of AMD progression may still have occurred.
Any other limitations?
These included:
- The potential for International Classification of Disease (ICD)-10 coding errors
- An inability to determine the influence of confounding factors
Take home.
These findings suggest that a reduced risk of developing AMD or progression to later stages was linked to subsequent use of CBNS—though confounding factors could account for the associations noted in this study.
As such: Further research to validate these findings and evaluate the safety and potential pharmacoprotective mechanisms of CBNS and AMD are warranted.