This cohort study investigates the outcomes of patients with and without age-related macular degeneration (AMD) who were taking curcuma-based nutritional supplements.
Key Points
Question
What are the outcomes of patients with and without age-related macular degeneration (AMD) who are taking curcuma-based nutritional supplements (CBNS)?
Findings
In this cohort study, 66 799 patients without AMD taking CBNS had an associated decreased risk of developing nonexudative AMD, advanced nonexudative AMD, exudative AMD, or blindness when compared with matched patients not taking CBNS.
Meaning
Although confounding factors could account for the associations noted in this study, these findings suggest a potential nutritional supplement might decrease the risk of developing nonexudative AMD, advanced nonexudative AMD, exudative AMD, or blindness, although safety also would need to be determined.
Abstract
Importance
Curcuma-based nutritional supplements (CBNS) are natural anti-inflammatory and antioxidant agents that may confer benefits against age-related macular degeneration (AMD).
Objective
To examine the association between the use of CBNS and the risk of development or progression of AMD.
Design, Setting, and Participants
This was a retrospective cohort study with data collection in June of 2024. Data were gathered from the aggregated electronic health records research network, TriNetX (Cambridge, Massachusetts). Patients without AMD were included in the study before propensity score matching (PSM); these included those taking and not taking CBNS. Patients with no history of AMD were stratified by instances of CBNS prescription records. Patients with a history of early nonexudative AMD stratified by instances of CBNS prescription records were also identified. PSM was performed to control for baseline demographics and medical comorbidities.
Exposures
Patients were stratified by whether or not they were taking CBNS using RxNorm (National Library of Medicine) codes.
Main Outcome Measures
Relative risk (RR) of developing nonexudative AMD, exudative AMD, advanced nonexudative AMD or geographic atrophy (GA), blindness, or requiring intravitreal anti–vascular endothelial growth factor (VEGF) therapy.
Results
A total of 66 804 patients (mean [SD] age, 64.9 [10.1] years; 44 124 female [66.1%]) taking CBNS and 1 809 440 patients (mean [SD] age, 67.0 [9.5] years; 999 534 female [55.2%]) not taking CBNS were included in this study. Among 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% CI, 0.21-0.26; P < .001), advanced nonexudative AMD or GA (RR, 0.11; 95% CI, 0.07-0.17; P < .001), exudative AMD (RR, 0.28; 95% CI, 0.24-0.32; P < .001), blindness (RR, 0.46; 95% CI, 0.36-0.59; P < .001), or requiring intravitreal anti-VEGF therapy (RR, 0.15; 95% CI, 0.13-0.17; P < .001) when compared with matched patients not taking CBNS. Results were consistent among subsets of patients 60 and 70 years or older, respectively. Among patients 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 < .001) when compared with matched patients with early nonexudative AMD without a CBNS prescription record.
Conclusion and Relevance
Results of this cohort study suggest that a reduced risk of developing AMD or progression to later stages of AMD was associated with subsequent use of CBNS. Further investigation to validate these findings, safety, and potential pharmacoprotective mechanisms of CBNS in AMD are suggested.
Introduction
Age-related macular degeneration (AMD), a leading cause of legal blindness in older patients1 has had genetic predisposition, oxidative damage, and chronic inflammation implicated in the pathophysiology of both nonexudative (dry) and exudative (wet) forms of the disease.2 Current pharmacological treatment strategies mitigate progression of the neovascular form through intravitreal anti–vascular endothelial growth factor (VEGF) injection therapy and possibly complement factor inhibition of geographic atrophy (GA) progression.3,4,5 Further, nutritional supplements, such as those used in the Age-Related Eye Disease Study (AREDS), reduced the risk of developing advanced stages of AMD.6,7
Curcumin, a biologically active ingredient of turmeric, can be isolated from the rhizomes of the plant Curcuma longa, consumed as a spice in curry meals, and used as a remedy for various ailments throughout South Asia.8 The pharmacological activities of curcumin, including antioxidant and anti-inflammatory effects, have been purported to mitigate the effects of aging.9,10,11,12,13,14 Antioxidant and anti-inflammatory properties of curcumin are linked to downregulation of free radical production and several inflammatory markers, respectively.15
The association of curcumin with development and progression of AMD has been limited to in vitro and nonhuman exploratory studies.16 This study leverages a large and diverse health research network to investigate potential associations between curcuma-based nutritional supplements (CBNS) and the risk of AMD development and progression.
Methods
Data Source
A retrospective cohort study was conducted using the TriNetX Health Research Network (Cambridge, Massachusetts). This research network aggregates deidentified electronic health record data of more than 107 health care organizations across 15 countries. The most updated data used in this study was collected on June 15, 2024. As this study only contained deidentified patient electronic medical records, the study was exempted from institutional review board approval, and informed consent was not obtained (per compliance with the Health Insurance Portability and Accountability Act). The study adhered to the Declaration of Helsinki and followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.17
Participants and Outcome Measures
This study used a retrospective cohort design, reviewing electronic medical records from January 2003 through June 2024. Patients with no history of AMD who had subsequent instances of CBNS prescription records (referred to as the CBNS cohort) were compared with those who had no history of AMD or a CBNS prescription record (referred to as the control cohort). Patient races and ethnicities included Asian, Black or African American, Hispanic or Latino, White, and unknown. Primary outcome measures were development of nonexudative AMD, advanced nonexudative AMD or GA, exudative AMD, or legal blindness as identified by their respective International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) codes. Legal blindness was defined as best-corrected visual acuity of 20/200 or less in the better eye or visual field that subtended an angle less than 20°.18 Secondary outcomes included subsequent need for intravitreal anti-VEGF injection therapy as identified by the Current Procedural Terminology (CPT) code. Both cohorts were further divided into 3 subgroups of patients 50 years or older, 60 years or older, and 70 years or older.
Observation Period and Exclusions
The observation period was between 6 months after the index event to the date of data collection. For the control cohort, the index event was defined as the patient’s first record of receiving a diagnosis of a cataract. For the CBNS cohort, the index event was defined as the first instance of receiving a diagnosis of a cataract and 3 instances of CBNS prescription codes were both satisfied. Patients with no history of AMD and instances of CBNS prescription codes were identified. A diagnosis of cataract was made within 1 year of a CBNS prescription record in both cohorts. Patients were excluded from their respective cohort if they had 2 years or less of follow-up before their index date; had a history of any of the outcomes of interest, proliferative diabetic retinopathy, diabetic macular edema, retinal vascular occlusions, or retinal edema; or a record of receiving an anti-VEGF injection before the index event. To ensure supplement continuation and adequate time for exposure, patients in the CBNS cohort were required to have 2 additional records of a CBNS prescription code after the index event.
Subgroup Analysis
A separate subgroup analysis of patients with early nonexudative AMD was performed, and patients were stratified based on the subsequent instances of CBNS prescription records or not. Similarly, the index event for the control cohort in the early nonexudative AMD group was defined as the patient’s first instance of early nonexudative AMD diagnosis record, whereas for the CBNS cohort, it was defined as the first instance when the requirements for an early nonexudative AMD diagnosis code and initiation of CBNS were both satisfied. Study groups in the subanalysis were followed up from 6 months after the index event until the date of data collection. Primary outcomes were development of advanced nonexudative AMD or GA and exudative AMD as identified by their respective ICD-10 codes. All patients had at least 6 months of follow-up after the index event. The same exclusion criteria outlined in the primary analysis was applied to the subgroup analysis. The relevant ICD-10, RxNorm (National Library of Medicine), and CPT codes used in the study are provided in the eTable in Supplement 1.
Statistical Analysis
Baseline covariates included age, sex, race, comorbidities, and medication/supplement use were balanced using propensity score matching (PSM). Patient race data were extracted from the TriNetX database. Including race as a baseline covariate aligns with our aim to control for potential disparities in treatment outcomes and adverse events associated with AMD and patients with diverse racial backgrounds. PSM was performed using the TriNetx built-in analysis platform (1:1 matching by nearest neighbor greedy matching algorithm with a caliper of 0.25 SDs).19 Standardized mean difference (SMD) for the continuous and categorical variables stratified by use of CBNS or not was calculated, and SMD less than or equal to 0.1 was considered balanced.19 Relative risk (RR) ratio, the difference between the RR ratio in each cohort (or RR difference), and 95% CIs were calculated for univariate comparison of cohorts and development of primary and secondary outcomes after matching. RR was a ratio of probability of either the primary or secondary outcome occurring in the CBNS cohort vs the probability of the event occurring in the control cohort. Furthermore, risk difference was calculated as the difference in risk to develop the primary or secondary outcome in the CBNS cohort and the control cohort. For continuous data, analyses were conducted using independent t tests. For categorical data analyses were conducted using χ2 tests. 95% CI of RR was calculated assuming the natural log follows a normal distribution. All P values were 2-sided but not adjusted for multiple analyses, and a P value <.05 was considered significant.
Results
A total of 66 804 patients (mean [SD] age, 64.9 [10.1] years; 44 124 female [66.1%]; 22 406 male [33.5%]) without AMD who were taking CBNS and 1 809 440 patients (mean [SD] age, 67.0 [9.5] years; 999 534 female [55.2%]; 726 491 male [40.2%]) without AMD who were not taking CBNS were identified before PSM. In the CBNS cohort, patients had the following races and ethnicities: 1015 Asian (1.5%), 4495 Black or African American (6.7%), 1463 Hispanic or Latino (2.2%), 54 745 White (82.0%), and 4716 unknown (7.1%). In the control cohort, patients had the following races and ethnicities: 65 140 Asian (3.6%), 226 723 Black or African American (12.5%), 110 557 Hispanic or Latino (6.1%), 1 048 389 White (57.9%), and 374 735 unknown (20.7%).
After PSM, criteria for age, sex, race, medical comorbidities including hypertension, hyperlipidemia, severe nonproliferative diabetic retinopathy, and smoking status were applied, a total of 66 799 patients were analyzed in each cohort. Baseline characteristics are shown before and after PSM in each cohort in Table 1. SMD was used to assess balance between both cohorts before and after PSM.
Table 1. Baseline Characteristics of Patients 50 Years or Older Without Age-Related Macular Degeneration (AMD) Who Initiated Curcuma-Based Nutritional Supplements (CBNS) and Those Who Did Not (Control) Before and After Propensity Score Matching (PSM).
| Characteristic | Before PSM | After PSM | ||||
|---|---|---|---|---|---|---|
| CBNS cohort (n = 66 804) | Control cohort (n = 1 809 440) | SMD | CBNS cohort (n = 66 799) | Control cohort (n = 66 799) | SMD | |
| Age, mean (SD), y | 64.9 (10.1) | 67.0 (9.52) | 0.22 | 64.9 (10.1) | 64.7 (9.91) | 0.02 |
| Sex, No. (%) | ||||||
| Female | 44 124 (66.05) | 999 534 (55.24) | 0.22 | 44 119 (66.05) | 43 766 (65.52) | 0.01 |
| Male | 22 406 (33.54) | 726 497 (40.15) | 0.14 | 22 402 (33.54) | 22 819 (34.16) | 0.01 |
| Unknown | 274 (0.42) | 83 415 (4.61) | 0.27 | 278 (0.42) | 214 (0.32) | 0.02 |
| Race, No. (%) | ||||||
| Asian | 1015 (1.52) | 65 140 (3.60) | 0.13 | 1015 (1.52) | 935 (1.40) | 0.01 |
| Black or African American | 4495 (6.73) | 226 723 (12.53) | 0.20 | 4495 (6.73) | 4723 (7.07) | 0.01 |
| Hispanic or Latino | 1463 (2.19) | 110 557 (6.11) | 0.20 | 1463 (2.19) | 1343 (2.01) | 0.01 |
| White | 54 745 (81.95) | 1 048 389 (57.94) | 0.54 | 54 741 (81.95) | 55 016 (82.36) | 0.01 |
| Unknown race | 4716 (7.06) | 374 735 (20.71) | 0.40 | 4716 (7.06) | 4275 (6.40) | 0.03 |
| Systemic associations, No. (%) | ||||||
| Hypertension | 25 873 (38.73) | 831 076 (45.93) | 0.15 | 25 871 (38.73) | 25 778 (38.59) | 0.02 |
| Hyperlipidemia | 19 861 (29.73) | 590 782 (32.65) | 0.06 | 19 859 (29.73) | 20 594 (30.83) | 0.02 |
| Nicotine dependence | 3694 (5.53) | 159 774 (8.83) | 0.13 | 3694 (5.53) | 4035 (6.04) | 0.02 |
| Severe NPDR (type 1) | 1 (0.001) | 36 (0.002) | 0.006 | 0 | 13 (0.02) | 0.02 |
| Severe NPDR (type 2) | 13 (0.02) | 904 (0.05) | 0.02 | 13 (0.02) | 19 (0.03) | 0.01 |
| Medications/supplements, No. (%) | ||||||
| β-Blocker | 17 683 (26.47) | 395 182 (21.84) | 0.11 | 17 682 (26.47) | 16 326 (24.44) | 0.05 |
| Melatonin | 3941 (5.90) | 44 874 (2.48) | 0.17 | 3941 (5.90) | 3968 (5.94) | 0.002 |
| Atorvastatin | 12 519 (18.74) | 307 424 (16.99) | 0.05 | 12 518 (18.74) | 12 331 (18.46) | 0.01 |
| Simvastatin | 6260 (9.37) | 168 459 (9.31) | 0.002 | 6259 (9.37) | 6079 (9.10) | 0.001 |
| Rosuvastatin | 5445 (8.15) | 85 768 (4.74) | 0.14 | 5444 (8.15) | 3841 (5.75) | 0.09 |
| Zeaxanthin | 160 (0.24) | 1086 (0.06) | 0.05 | 160 (0.24) | 134 (0.20) | 0.01 |
| Lutein | 1744 (2.61) | 9228 (0.51) | 0.17 | 1743 (2.61) | 1503 (2.25) | 0.02 |
| Vitamin C | 11 270 (16.87) | 87 577 (4.84) | 0.39 | 11 269 (16.87) | 11 242 (16.83) | <0.001 |
| Vitamin E | 5144 (7.70) | 46 503 (2.57) | 0.23 | 5143 (7.70) | 4810 (7.20) | 0.02 |
| Zinc | 1617 (2.42) | 12 123 (0.67) | 0.14 | 1617 (2.42) | 1396 (2.09) | 0.02 |
| Copper | 327 (0.49) | 2352 (0.13) | 0.06 | 327 (0.49) | 281 (0.42) | 0.01 |
Abbreviations: NPDR, nonproliferative diabetic retinopathy; SMD, standardized mean difference.
Table 2 demonstrates the incidence of developing ocular complications, including nonexudative AMD, advanced nonexudative AMD, exudative AMD, and blindness among AMD-naive patients in the CBNS cohort compared with the control cohort with subsets aged 50, 60 and 70 years or older. Among patients 50 years or older without a history of AMD, those with instances of CBNS prescription codes had a reduced risk of developing nonexudative AMD (485 of 66 799 [0.73%] vs 2079 of 66 799 [3.11%]; RR, 0.23; 95% CI, 0.21-0.26; risk difference [RD], −2.39%; 95% CI, −2.53% to −2.24%; P < .001), advanced nonexudative AMD (19 of 66 799 [0.03%] vs 176 of 66 799 [0.26%]; RR, 0.11; 95% CI, 0.07-0.17; RD, −0.24%; 95% CI, −0.28% to −0.19%; P < .001), exudative AMD (206 of 66 799 [0.31%] vs 748 of 66 799 [1.12%]; RR, 0.28; 95% CI, 0.24-0.32; RD, −0.81%; 95% CI, −0.90% to −0.72%; P < .001), and blindness (90 of 66 799 [0.14%] vs 196 of 66 799 [0.29%]; RR, 0.46; 95% CI, 0.36-0.59; RD, −0.16%; 95% CI, −0.21% to −0.11%; P < .001) when compared with patients not taking CBNS. The mean (SD) follow-up time was 1501 (876) days and 1540 (931) days in the CBNS and control cohort, respectively.
Table 2. Incidence of Developing Dry Age-Related Macular Degeneration (AMD), Advanced Dry AMD, Exudative (Wet) AMD, and Legal Blindness in Patients Without AMD Taking Curcuma-Based Nutritional Supplements (CBNS) vs Those Not Taking CBNS (Control).
| Age, y | CBNS cohort | Control cohort | Risk ratio (95% CI) | Risk difference, % (95% CI) | P value | ||
|---|---|---|---|---|---|---|---|
| Total | Events, No. (%) | Total | Events, No. (%) | ||||
| Dry AMD | |||||||
| ≥50 | 66 799 | 485 (0.73) | 66 799 | 2079 (3.11) | 0.23 (0.21 to 0.26) | −2.39 (−2.53 to −2.24) | <.001 |
| ≥60 | 55 718 | 482 (0.87) | 55 718 | 2157 (3.87) | 0.22 (0.20 to 0.25) | −3.00 (−3.18 to −2.83) | <.001 |
| ≥70 | 35 078 | 426 (1.21) | 35 078 | 1869 (5.33) | 0.23 (0.21 to 0.25) | −4.11 (−4.38 to −3.85) | <.001 |
| Advanced dry AMD | |||||||
| ≥50 | 66 799 | 19 (0.03) | 66 799 | 176 (0.26) | 0.11 (0.07 to 0.17) | −0.24 (−0.28 to −0.19) | <.001 |
| ≥60 | 55 718 | 19 (0.03) | 55 718 | 185 (0.33) | 0.10 (0.06 to 0.17) | −0.30 (−0.35 to −0.25) | <.001 |
| ≥70 | 35 078 | 19 (0.05) | 35 078 | 178 (0.51) | 0.11 (0.07 to 0.17) | −0.45 (−0.53 to −0.38) | <.001 |
| Wet AMD | |||||||
| ≥50 | 66 799 | 206 (0.31) | 66 799 | 748 (1.12) | 0.28 (0.24 to 0.32) | −0.81 (−0.90 to −0.72) | <.001 |
| ≥60 | 55 718 | 205 (0.37) | 55 718 | 712 (1.28) | 0.29 (0.25 to 0.33) | −0.91 (−1.02 to −0.80) | <.001 |
| ≥70 | 35 078 | 185 (0.53) | 35 078 | 628 (1.79) | 0.30 (0.25 to 0.35) | −1.26 (−1.42 to −1.11) | <.001 |
| Legal blindness | |||||||
| ≥50 | 66 799 | 90 (0.14) | 66 799 | 196 (0.29) | 0.46 (0.36 to 0.59) | −0.16 (−0.21 to −0.11) | <.001 |
| ≥60 | 55 718 | 80 (0.14) | 55 718 | 153 (0.28) | 0.52 (0.40 to 0.69) | −0.13 (−0.19 to −0.08) | <.001 |
| ≥70 | 35 078 | 54 (0.15) | 35 078 | 83 (0.24) | 0.65 (0.46 to 0.92) | −0.08 (−0.15 to −0.02) | .01 |
Among the older subset of patients without a history of AMD, those with instances of CBNS prescription codes had a reduced risk of developing nonexudative AMD (aged ≥60 years: 482 of 55 718 [0.87%] vs 2157 of 55 718 [3.87]%; RR, 0.22; 95% CI, 0.20-0.25; RD, −3.00%; 95% CI, −3.18% to −2.83%; P < .001; aged ≥70 years: 426 of 35 078 [1.21%] vs 1869 of 35 078 [5.33%]; RR, 0.23; 95% CI, 0.21-0.25; RD, −4.11%; 95% CI, −4.38% to −3.85%; P < .001), advanced nonexudative AMD (aged ≥60 years: 19 of 55 718 [0.03%] vs 185 of 55 718 [0.33%]; RR, 0.10; 95% CI, 0.06-0.17; RD, −0.30%; 95% CI, −0.35% to −0.25%; P < .001; aged ≥70 years: 19 of 35 078 [0.05%] vs 178 of 35 078 [0.51%]; RR, 0.11; 95% CI, 0.07-0.17; RD, −0.45%; 95% CI, −0.53% to −0.38%; P < .001), exudative AMD (aged ≥60 years: 205 of 55 718 [0.37%] vs 712 of 55 718 [1.28%]; RR, 0.29; 95% CI, 0.25-0.33; RD, −0.91%; 95% CI, −1.02% to −0.80%; P < .001; aged ≥70 years: 185 of 35 078 [0.53%] vs 628 of 35 078 [1.79%]; RR, 0.30; 95% CI, 0.25-0.35; RD, −1.26%; 95% CI, −1.42% to −1.11%; P < .001), and blindness (aged ≥60 years: 80 of 55 718 [0.14%] vs 153 of 55 718 [0.28%]; RR, 0.52; 95% CI, 0.40-0.69; RD, −0.13%; 95% CI, −0.19% to −0.08%; P < .001; aged ≥70 years: 54 of 35 078 [0.15%] vs 83 of 35 078 [0.24%]; RR, 0.65; 95% CI, 0.46-0.92; RD, −0.08%; 95% CI, −0.15% to −0.02%; P = .01) when compared with patients not taking CBNS. The mean (SD) follow-up time in patients 60 years or older was 1513 (882) days and 1570 (943) days in the CBNS and control cohort, respectively. Furthermore, mean (SD) follow-up time in patients 70 years or older was 1538 (890) days and 1610 (961) days in the CBNS and control cohort, respectively (Table 2).
The RR of requiring subsequent ocular interventions, including intravitreal anti-VEGF therapy, was also evaluated among AMD-naive patient subsets 50, 60, and 70 years or older stratified by instances of CBNS prescription codes or not. Among patients without a history of AMD, those with instances of CBNS prescription codes had a reduced risk of requiring subsequent anti-VEGF therapy (aged ≥50 years: 172 of 66 799 [0.26%] vs 1164 of 66 799 [1.74%]; RR, 0.15; 95% CI, 0.13-0.17; RD, −1.49%; 95% CI, −1.59% to −1.38%; P < .001; aged ≥60 years: 161 of 55 718 [0.29%] vs 931 of 55 718 [1.67%]; RR, 0.17; 95% CI, 0.15-0.20; RD, −1.38%; 95% CI, −1.50% to −1.27%; P < .001; aged ≥70 years: 137 of 35 078 [0.39%] vs 638 of 35 078 [1.82]%; RR, 0.22; 95% CI, 0.18-0.26; RD, −1.43%; 95% CI, −1.58% to −1.27%; P < .001) when compared with patients not using CBNS at every age subset (Table 3). The Figure summarizes RRs of developing nonexudative AMD, advanced nonexudative AMD, exudative AMD, legal blindness, or requiring an intravitreal anti-VEGF injection among patients with nonexudative AMD taking CBNS compared with those not taking CBNS at subsets of patients aged 50, 60, and 70 years or older.
Table 3. Incidence of Requiring Intravitreal Anti–Vascular Endothelial Growth Factor (VEGF) Therapy in Patients Without Age-Related Macular Degeneration (AMD) Taking Curcuma-Based Nutritional Supplements (CBNS) vs Those Not Taking CBNS (Control).
| Age, y | CBNS cohort | Control cohort | Risk ratio (95% CI) | Risk difference, % (95% CI) | P value | ||
|---|---|---|---|---|---|---|---|
| Total | Events, No. (%) | Total | Events, No. (%) | ||||
| ≥50 | 66 799 | 172 (0.26) | 66 799 | 1164 (1.74) | 0.15 (0.13 to 0.17) | −1.49 (−1.59 to −1.38) | <.001 |
| ≥60 | 55 718 | 161 (0.29) | 55 718 | 931 (1.67) | 0.17 (0.15 to 0.20) | −1.38 (−1.50 to −1.27) | <.001 |
| ≥70 | 35 078 | 137 (0.39) | 35 078 | 638 (1.82) | 0.22 (0.18 to 0.26) | −1.43 (−1.58 to −1.27) | <.001 |
Figure. Relative Risks of Age-Related Macular Degeneration (AMD) Development.
Relative risks of developing nonexudative (dry) AMD, advanced dry AMD, exudative (wet) AMD, legal blindness, and requiring an intravitreal anti–vascular endothelial growth factor (VEGF) injection among patients with dry AMD using curcuma-based nutritional supplements compared with those not at subsets of patients aged 50, 60, and 70 years or older.
A subanalysis of patients with early nonexudative AMD and a CBNS prescription record compared with patients not taking CBNS was performed (Table 4). After exclusions and PSM for the same covariates in the initial analysis of the AMD-naive group, a total of 1111 patients were subsequently analyzed in each cohort. Use of CBNS in patients with early nonexudative AMD was associated with lower rates of developing advanced nonexudative AMD (48 [4.32%] vs 84 [7.50%]; RR, 0.58; 95% CI, 0.41-0.81; RD, −3.16%; 95% CI, −5.11% to −1.20%; P < .001) when compared with matched patients not taking CBNS with a mean (SD) follow-up time of 1653 (989) days and 1654 (1046) days, respectively. There was no difference in subsequent development of exudative AMD in patients with early nonexudative AMD who were taking CBNS when compared with matched patients not taking CBNS (Table 4).
Table 4. Incidence of Developing Advanced Dry Age-Related Macular Degeneration (AMD) and Exudative (Wet) AMD in Patients With Early AMD Taking Curcuma-Based Nutritional Supplements (CBNS) vs Those Not Taking CBNS (Control) After Propensity Score Matching.
| AMD type | Events, No. (%) | Risk ratio (95% CI) | Risk difference, % (95% CI) | P value | |
|---|---|---|---|---|---|
| CBNS cohort (n = 1111) | Control cohort (n = 1111) | ||||
| Advanced dry AMD | 48 (4.32) | 84 (7.50) | 0.58 (0.41 to 0.81) | −3.16 (−5.11 to −1.20) | <.001 |
| Exudative AMD | 136 (12.24) | 160 (14.40) | 0.85 (0.69 to 1.05) | −2.16 (−4.99 to 0.66) | .13 |
Discussion
Use of CBNS was associated with a decreased risk of developing subsequent nonexudative AMD, advanced nonexudative AMD, exudative AMD, or blindness in patients without a history of AMD when compared with a matched cohort of patients not taking CBNS. Patients without a history of AMD and instances of CBNS prescription codes had less associated risk of requiring intravitreal anti-VEGF injections when compared with a matched control group. Among patients with early nonexudative AMD, use of CBNS use was associated with a reduced risk of developing subsequent advanced nonexudative AMD but not exudative AMD. These findings were observed even after adjustment for relevant covariates and across different age groups, suggesting that CBNS may protect from the development and progression of AMD, although these associations could be due to confounding factors rather than a cause-and-effect mechanism.
A small prospective, single-center, nonblinded, randomized clinical trial (RCT) by Allegrini et al20 suggested that patients with exudative AMD who started CBNS treatment required less intravitreal anti-VEGF therapy (median intravitreal injections, 4 vs 7) and had improved median best-corrected visual acuity when compared with matched patients not taking CBNS. In a separate prospective, single-center, nonblinded, RCT presented as an abstract but not published further, Knepper et al21 used oral curcumin for 2 years in 18 patients with intermediate AMD and found that the total mean drusen volume reduced from 0.0253 mm3 to 0.0220 mm3 with no cases of choroidal neovascularization or GA after 2 years. In the phase 2 trial presented as an abstract, Knepper et al22 expanded on his previous findings and showed that curcumin in combination with resveratrol and quercetin reduced mean drusen volume by 10% after 1 year among patients with advanced dry AMD. Although these findings support the therapeutic potential of CBNS against AMD, this study did not have a control group and may not be generalizable to racially and culturally diverse populations and was too small to provide adequate safety data.21,23,24,25 In the present study, when compared with AMD-naive patients 50 years or older without a record of CBNS use, those with a CBNS prescription record had an associated risk reduction of 77% in the development of nonexudative AMD (RR, 0.23; P < .001), 89% in the development of advanced nonexudative AMD (RR, 0.11; P < .001), 72% in the development of exudative AMD (RR, 0.28; P < .001), and 54% in the development of blindness (RR, 0.46; P < .001) using a large heterogenous database of matched patients. Although a significant reduction in the risk of AMD complications associated with use of CBNS was observed in this study compared with smaller effects of other vitamins and supplements previously described, confounding factors could account for the associations observed and future prospective trials are needed to validate the magnitude of the benefit conferred with use of CBNS and determine its safety.
Limitations
Limitations within this study are inherent in the analysis of large sets of deidentified aggregated medical health records data. This study relied on accurate ICD-10 diagnosis and coding, and as such, there may be inaccuracy of true coding for nonneovascular as well as neovascular forms of AMD. In addition, for rates of developing exudative AMD in patients with early nonexudative AMD taking CBNS when compared with matched patients not taking CBNS, no difference was found after applying strict inclusion/exclusion criteria and PSM matching for covariates. Multiplicity adjustments were not applied to the outcomes of the analyses, and the results of the present study are associations that generate hypotheses but do not determine cause and effect. Furthermore, despite best efforts to control for variations between both cohorts with PSM, we were unable to fully control for all possible risk factors in the development of AMD or in the decision to pursue intravitreal anti-VEGF injection therapy when an ocular intervention was undertaken. Likewise, adherence to CBNS may be a marker for a healthy user bias that may result in false association of CBNS use with better outcomes. To select patients with sustained CBNS use when constructing cohorts, a requirement for 3 instances of CBNS record over the course of 3 years after the index event was made. Despite this, differences in duration, dose, frequency, and bioavailability of CBNS could not be standardized as CBNS is only regulated as a supplement by the US Food and Drug Administration.26,27 Further, although 3 instances of CBNS prescription records were required for enrollment into the CBNS cohort, certainty in the prescribed use of this natural supplement cannot be ascertained given the nature of this study. Balance was achieved between the number of days each cohort was followed, yet differences in loss of follow-up rates could confound the results of the study. Also, other confounding factors not controlled for may contribute to the results noted, as some of the RR values were in the lower bounds of the 95% CI, and sensitivity analysis was unable to be performed.
Conclusions
In conclusion, this cohort study provides evidence for a decreased associated risk of developing ocular complications, including nonexudative AMD, advanced nonexudative AMD, exudative AMD, or blindness, as well as requiring intervention with intravitreal anti-VEGF pharmacotherapy, in patients with and without AMD who are taking CBNS when compared with matched patients not taking CBNS in a large real-world live patient dataset. Despite its limitations, this study complements previous findings that supplements with biochemical properties like CBNS may affect ocular outcomes of patients with and without preexisting AMD and provides hypotheses for potential future RCTs to evaluate the safety and potential efficacy of CBNS use in patients with and without AMD.
eTable. Comprehensive Overview of Medications, Diagnoses, and Procedures With Coding Data
Data Sharing Statement.
References
- 1.World Health Organization . Blindness and vision impairment. Accessed June 15, 2024. https://www.who.int/news-room/fact-sheets/detail/blindness-and-visual-impairment
- 2.Ambati J, Fowler BJ. Mechanisms of age-related macular degeneration. Neuron. 2012;75(1):26-39. doi: 10.1016/j.neuron.2012.06.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Parravano M, Costanzo E, Scondotto G, Trifirò G, Virgili G. Anti-VEGF and other novel therapies for neovascular age-related macular degeneration: an update. BioDrugs. 2021;35(6):673-692. doi: 10.1007/s40259-021-00499-2 [DOI] [PubMed] [Google Scholar]
- 4.Khanani AM, Patel SS, Staurenghi G, et al. ; GATHER2 trial investigators . Efficacy and safety of avacincaptad pegol in patients with geographic atrophy (GATHER2): 12-month results from a randomised, double-masked, phase 3 trial. Lancet. 2023;402(10411):1449-1458. doi: 10.1016/S0140-6736(23)01583-0 [DOI] [PubMed] [Google Scholar]
- 5.Heier JS, Lad EM, Holz FG, et al. ; OAKS and DERBY study investigators . Pegcetacoplan for the treatment of geographic atrophy secondary to age-related macular degeneration (OAKS and DERBY): 2 multicenter, randomized, double-masked, sham-controlled, phase 3 trials. Lancet. 2023;402(10411):1434-1448. doi: 10.1016/S0140-6736(23)01520-9 [DOI] [PubMed] [Google Scholar]
- 6.Age-Related Eye Disease Study 2 Research Group . Lutein + zeaxanthin and omega-3 fatty acids for age-related macular degeneration: the Age-Related Eye Disease Study 2 (AREDS2) randomized clinical trial. JAMA. 2013;309(19):2005-2015. doi: 10.1001/jama.2013.4997 [DOI] [PubMed] [Google Scholar]
- 7.Age-Related Eye Disease Study Research Group . The Age-Related Eye Disease Study (AREDS): design implications. AREDS report no. 1. Control Clin Trials. 1999;20(6):573-600. doi: 10.1016/S0197-2456(99)00031-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Eigner D, Scholz D. Ferula asa-foetida and Curcuma longa in traditional medical treatment and diet in Nepal. J Ethnopharmacol. 1999;67(1):1-6. doi: 10.1016/S0378-8741(98)00234-7 [DOI] [PubMed] [Google Scholar]
- 9.Aggarwal BB, Kumar A, Bharti AC. Anticancer potential of curcumin: preclinical and clinical studies. Anticancer Res. 2003;23(1A):363-398. [PubMed] [Google Scholar]
- 10.Woo JM, Shin DY, Lee SJ, et al. Curcumin protects retinal pigment epithelial cells against oxidative stress via induction of heme oxygenase-1 expression and reduction of reactive oxygen. Mol Vis. 2012;18:901-908. [PMC free article] [PubMed] [Google Scholar]
- 11.Tyagi AK, Prasad S, Yuan W, Li S, Aggarwal BB. Identification of a novel compound (β-sesquiphellandrene) from turmeric (Curcuma longa) with anticancer potential: comparison with curcumin. Invest New Drugs. 2015;33(6):1175-1186. doi: 10.1007/s10637-015-0296-5 [DOI] [PubMed] [Google Scholar]
- 12.Partridge L, Fuentealba M, Kennedy BK. The quest to slow ageing through drug discovery | Nature Reviews Drug Discovery. Accessed June 15, 2024. https://www.nature.com/articles/s41573-020-0067-7 [DOI] [PubMed]
- 13.Vaiserman A, Koliada A, Zayachkivska A, Lushchak O. Curcumin: a therapeutic potential in ageing-related disorders. PharmaNutrition. 2020;14:100226. doi: 10.1016/j.phanu.2020.100226 [DOI] [Google Scholar]
- 14.Tayyem RF, Heath DD, Al-Delaimy WK, Rock CL. Curcumin content of turmeric and curry powders. Nutr Cancer. 2006;55(2):126-131. doi: 10.1207/s15327914nc5502_2 [DOI] [PubMed] [Google Scholar]
- 15.Xu XY, Meng X, Li S, Gan RY, Li Y, Li HB. Bioactivity, health benefits, and related molecular mechanisms of curcumin: current progress, challenges, and perspectives. Nutrients. 2018;10(10):1553. doi: 10.3390/nu10101553 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Chandrasekaran PR, Madanagopalan VG. Role of curcumin in retinal diseases—a review. Graefes Arch Clin Exp Ophthalmol. 2022;260(5):1457-1473. doi: 10.1007/s00417-021-05542-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Vandenbroucke JP, von Elm E, Altman DG, et al. ; STROBE Initiative . Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS Med. 2007;4(10):e297. doi: 10.1371/journal.pmed.0040297 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Social Security Administration . Meaning of blindness as defined in the law. Accessed June 16, 2024. https://www.ssa.gov/OP_Home/cfr20/404/404-1581.htm
- 19.Austin PC. An Introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res. 2011;46(3):399-424. doi: 10.1080/00273171.2011.568786 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Allegrini D, Raimondi R, Angi M, et al. Curcuma-based nutritional supplement in patients with neovascular age-related macular degeneration. J Med Food. 2021;24(11):1191-1196. doi: 10.1089/jmf.2020.0199 [DOI] [PubMed] [Google Scholar]
- 21.Knepper PA, Pfahler N, Bielskus I, Volpe NJ, Aman S, Zaparackas ZA. Two-year, phase 1 study to evaluate the efficacy of oral curcumin on drusen volume. Invest Ophthalmol Vis Sci. 2022;63(7):369-F0200. [Google Scholar]
- 22.Knepper PA, Pfahler N, Klovaite G, Aman S, Zaparackas Z. Macular drusen size in a phase 2 clinical trial evaluating resveratrol, quercetin, and curcumin (RQC). Invest Ophthalmol Vis Sci. 2023;64(8):2774. [Google Scholar]
- 23.Bielskus I, Pfahler N, Kakouri A, Aman S, Zaparackas Z, Knepper PA. Can drusen growth in age-related macular degeneration be slowed by oral curcumin? Invest Ophthalmol Vis Sci. 2021;62(8):330. [Google Scholar]
- 24.Bielskus I, Kakouri A, Pfahler NM, et al. Curcumin acts to regress macular drusen volume in dry AMD. Invest Ophthalmol Vis Sci. 2020;61(7):1036. [Google Scholar]
- 25.Pfahler N, Bielskus I, Volpe NJ, Aman S, Zaparackas Z, Knepper PA. Response of drusen volume to curcumin is correlated with risk alleles for age-related macular degeneration. Invest Ophthalmol Vis Sci. 2022;63(7):372-F0203. [Google Scholar]
- 26.National Cancer Institute . Curcumin (curcuma, turmeric) and cancer (PDQ)—health professional version. Accessed June 17, 2024. https://www.cancer.gov/about-cancer/treatment/cam/hp/curcumin-pdq [PubMed]
- 27.Jantawong C, Priprem A, Intuyod K, et al. Curcumin-loaded nanocomplexes: acute and chronic toxicity studies in mice and hamsters. Toxicol Rep. 2021;8:1346-1357. doi: 10.1016/j.toxrep.2021.06.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable. Comprehensive Overview of Medications, Diagnoses, and Procedures With Coding Data
Data Sharing Statement.
