ABSTRACT
Background
Curcumin is a commonly used herbal supplement with anti-inflammatory and anti-fibrotic properties. Animal studies and small human trials suggest that curcumin reduces albuminuria in patients with chronic kidney disease (CKD). Micro-particle curcumin is a new, more bioavailable formulation of curcumin.
Methods
To determine whether micro-particle curcumin versus placebo slows the progression of albuminuric CKD we conducted a randomized, double-blind, placebo-controlled trial with 6-month follow-up. We included adults with albuminuria [a random urine albumin-to-creatinine ratio >30 mg/mmol (265 mg/g) or a 24-h urine collection with more than 300 mg of protein] and an estimated glomerular filtration rate (eGFR) between 15 and 60 mL/min/1.73 m2 within the 3 months before randomization. We randomly allocated participants 1:1 to receive micro-particle curcumin capsules (90 mg/day) or matching placebo for 6 months. After randomization, the co-primary outcomes were the changes in albuminuria and the eGFR.
Results
We enrolled 533 participants, but 4/265 participants in the curcumin group and 15/268 in the placebo group withdrew consent or became ineligible. The 6-month change in albuminuria did not differ significantly between the curcumin and placebo groups [geometric mean ratio 0.94, 97.5% confidence interval (CI) 0.82 to 1.08, P = .32]. Similarly, the 6-month change in eGFR did not differ between groups (mean between-group difference –0.22 mL/min/1.73 m2, 97.5% CI –1.38 to 0.95, P = .68).
Conclusions
Ninety milligrams of micro-particle curcumin daily did not slow the progression of albuminuric CKD over 6 months.
Trial registration
ClinicalTrials.gov Identifier: NCT02369549.
Keywords: albuminuria, chronic renal insufficiency, clinical trial, curcumin, GFR
Graphical Abstract
Graphical Abstract.
KEY LEARNING POINTS.
What is already known about this subject?
Curcumin is a popular natural health product. Micro-particle curcumin is a more bioavailable formulation of traditional curcumin.
Based on animal and small-scale human studies, curcumin may reduce albuminuria in the setting of chronic kidney disease.
What this study adds?
Micro-particle curcumin at a dose of 90 mg/day did not reduce albuminuria or affect the estimated glomerular filtration rate over 6 months, compared with placebo.
The findings of Micro-PArticle Curcumin in the treatment of Chronic Kidney Disease (MPAC-CKD) do not justify a larger-scale trial of micro-particle curcumin to examine its effects on clinically meaningful outcomes.
What impact this may have on practice or policy?
Using curcumin to slow the progression of chronic kidney disease is not supported by the results of MPAC-CKD.
It is important to subject promising natural health products to rigorous testing prior to their widespread adoption.
INTRODUCTION
Chronic kidney disease (CKD) is an important global health concern [1]. Although CKD is largely asymptomatic, it is a potent risk factor for cardiovascular disease [2], cognitive decline [3] and premature death [4]. CKD can also lead to kidney failure, which confers great morbidity and mortality, and a life-long dependence on dialysis or transplantation [5, 6]. Interventions that slow the progression of CKD are extremely important but limited in number and even with optimal medical management, a substantial number of patients still progress to kidney failure [7].
Curcumin is a component of the dietary spice turmeric. It commands the fifth largest share of the American herbal supplement market [8]. Curcumin has both anti-inflammatory and anti-fibrotic properties and has been investigated as a treatment for a wide range of conditions. Over 100 in vitro and animal studies show that curcumin ameliorates many causes of kidney damage (Supplementary data, Fig. S1 depicts curcumin's effects relevant to CKD) [9–11], and some small clinical trials suggest curcumin reduces albuminuria in patients with CKD (Supplementary data, Table S1).
Micro-particle curcumin is a new, more bioavailable formulation of curcumin [12]. To determine whether micro-particular curcumin warrants a large-scale trial with clinically meaningful outcomes, we conducted the Micro-PArticle Curcumin in the treatment of Chronic Kidney Disease (MPAC-CKD) randomized clinical trial to investigate the effect of micro-particle curcumin markers of CKD progression.
MATERIALS AND METHODS
Trial design
We conducted a randomized, parallel-group, placebo-controlled trial among patients with albuminuric CKD in which participants were randomly allocated in a 1:1 ratio to micro-particle curcumin or matching placebo using an interactive web-based randomization system. This trial is reported according to the CONSORT (Consolidated Standards of Reporting Trials) guidelines [13]. We previously published a trial protocol with a detailed description of the objectives, interventions, design and analytic plan [14]. Both the trial protocol and statistical analysis plan are available through ClinicalTrials.gov (NCT02369549). Differences between the published protocol, the pre-specified statistical analysis plan, and the trial conduct are listed in Supplementary data, Table S2.
Trial oversight
MPAC-CKD was an investigator-initiated trial. M.A.W. served as the trial sponsor-investigator. The Lilibeth Caberto Kidney Clinical Research at the Lawson Health Research Institute in London, Ontario, Canada managed the trial and was responsible for randomization, data acquisition and distribution of the investigational product. Peer-reviewed funding for MPAC-CKD was provided by the Kidney Foundation of Canada (KFOC130029), and the Canadian Institute of Health Research (CIHR365629). Theravalues® (Tokyo, Japan) and Natural Factors® (Coquitlam, Canada) supplied the micro-particle curcumin and matching placebo but had no role in the design, execution or reporting of the trial. MPAC-CKD was designed by the authors, who warrant the reported data and adherence to the protocol. The data safety monitoring board reviewed unblinded data at 50% enrolment and recommended no alterations to the protocol.
Participants and setting
We conducted MPAC-CKD at three sites at Western University, London, Canada and one site at McMaster University, Hamilton, Canada. Between September 2015 and November 2019, we screened patients with CKD who had both an estimated glomerular filtration rate (eGFR) between 15 and 60 mL/min/1.73 m2 and albuminuria [24-h urine collection with more than 300 mg of protein or a random urine albumin-to-creatinine ratio >30 mg/mmol (265 mg/g)] within the 3 months prior to randomization. The presence of these features indicates a person has a moderate-to-high risk of CKD progression. We required participants to be on a stable dose of an angiotensin-converting enzyme inhibitor or an angiotensin receptor antagonist, or have a documented reason why they were not. This trial was conducted prior to the widespread use of sodium-glucose contransporter-2 inhibitors. We excluded patients with a history of receiving dialysis and those with conditions that could potentially be exacerbated by micro-particle curcumin (active peptic ulcer disease, hepatobiliary disease, history of significant bleeding) [15–17]. A complete list of inclusion and exclusion criteria are provided in Supplementary data, Table S3. The Research Ethics board at Western University approved the trial.
Intervention and randomized allocation
Using a computerized algorithm, we randomly assigned participants in a 1:1 ratio to receive either 90 mg of micro-particle curcumin or matching placebo once daily for 6 months. The algorithm distributed participants in balanced blocks of variable sizes and stratified the allocation by site and baseline presence of diabetes mellitus. Participants, investigators and research staff were unaware of the treatment allocation.
All investigational product was approved by Health Canada. The micro-particle curcumin capsules contained curcumin dispersed in colloidal sub-micron particles. Compared with traditional curcumin, which has very poor bioavailability, this formulation is more readily absorbed and can generate measurable serum curcumin levels at doses used in this trial [12]. With a 27-fold increase in bioavailability compared with traditional curcumin [12], we selected a dose of 90 mg to approximate the doses of traditional curcumin used in previous studies (Supplementary data, Table S1). We imposed no dietary restrictions because dietary sources of turmeric provide extremely low amounts of curcumin [18]. We did require participants to avoid the use of turmeric or curcumin supplements. Our experience measuring serum and urine concentrations of curcumin in this trial is described in Supplementary data, Appendix S1.
Outcomes
Co-primary outcomes
The pre-specified co-primary outcomes were the 6-month change in albuminuria and the 6-month change in eGFR (calculated with the 2009 Chronic Kidney Disease Epidemiology Collaboration creatinine equation). For each participant, we assessed the mean of two consecutive first-morning urine albumin-to-creatinine ratios before randomization and again 6 months after randomization, because we previously demonstrated this improves precision of the measurement [19]. The serum creatinine measurement used in GFR estimation was traceable to isotope dilution mass spectrometry.
Secondary outcomes
Pre-specified secondary outcomes were health-related quality of life, glycemic control and a renal failure composite outcome. We estimated changes in health-related quality of life using the Physical and Mental Component Summary scores of the Short Form-36 (SF-36) Health Survey assessed from randomization to 6 months. The SF-36 Component Summary scores are contributed to by eight domains of health-related quality of life and are standardized to population norms, each providing a score from 0 to 100, with higher scores indicating better quality of life. We measured glycemic control in a subset of patients with diabetes mellitus using the change in the percentage of glycated hemoglobin. We assessed the renal failure composite outcome as death from any cause, a loss of eGFR of 30% or more, or development of an eGFR <15 mL/min/1.73 m2, or the receipt of dialysis or a kidney transplant. We gathered data on minor and major adverse events.
Post hoc exploratory outcomes
We conducted an exploratory subgroup analysis to determine whether the baseline risk of kidney failure modified the effect of micro-particle curcumin on the co-primary outcomes. Using the Kidney Failure Risk Equation [20], we divided patients into those with a 2-year estimated risk of kidney failure ≥10% or <10%.
Statistical methods
We found the albumin-to-creatinine ratio data to be skewed and therefore, we chose to log-transform it for sample size estimation and outcome analysis. This protocol change is described in the Supplementary data, Table S2a.
Sample size
We previously found that patients eligible for MPAC-CKD had a geometric mean urine albumin-to-creatinine ratio of 87 mg/mmol (1134 mg/g) with a standard deviation of the change of 69 mg/mmol (612 mg/g) [19]. We determined that 500 participants would allow for 82% power (alpha 0.025) to exclude a geometric mean ratio of 0.75 [21]. This sample size would also have 90% power to exclude a between-group difference in the mean change in eGFR of 2.3 mL/min/1.73 m2 (using a standard deviation of 4 mL/min/1.73 m2) [22, 23]. Therefore, we secured enough micro-particle curcumin to enroll between 500 and 750 participants. We stopped enrolling soon after 500 participants on a date beyond which the investigation product would expire during the 6-month follow-up.
Analysis of primary and secondary outcomes
We conducted all primary and secondary outcome analyses according to a modified intention-to-treat principle. We excluded 18 patients who underwent random allocation but did not start treatment (4 met exclusion criteria and 14 withdrew consent). The remaining participants were analyzed according to the group to which they were allocated, consistent with the intention-to-treat principle. For the effect of the treatment on albuminuria, we estimated the geometric mean ratio of the 6-month urine albumin-to-creatinine ratio using linear regression, where the outcome was the log-transformed average of two 6-month urine albumin-to-creatinine ratio measurements minus the log-transformed average of two pre-randomization urine albumin-to-creatinine ratio measurements, adjusted for a treatment indicator variable and the log-transformed average of the two pre-randomization urine albumin-to-creatinine ratio measurements. For all other outcomes, we estimated the treatment effect by comparing the between-group differences in the mean absolute change from baseline to 6 months. Because we tested two primary outcomes, we used the Hochberg procedure to test each of the two-sided hypothesis tests (change in albuminuria and change in kidney function) [24]. We first fixed the overall alpha level at 0.05. We then compared the larger of the two P-values with an alpha of 0.05, and because this comparison was non-significant, we then compared the smaller P-value with an alpha of 0.025. We report outcomes tested at the 0.025 significance level with 97.5% confidence intervals (CIs) for both tests. Because our secondary outcomes were exploratory in nature, we report 95% CIs without adjustment for multiple testing. We dealt with missing data for causes other than death using model-based multiple imputation. As described below, we conducted sensitivity analyses to confirm that conclusions were not sensitive to assumptions about the missing-data mechanism. A summary of missing outcome data is shown in Supplementary data, Table S4.
Sensitivity analyses
We conducted the following three sensitivity analyses of the co-primary outcomes and three secondary outcomes (the physical and mental component summary scores of the SF-36 health survey and percentage glycated hemoglobin). First, we adjusted for randomization strata: center and pre-randomization diabetes. Second, we adjusted for randomization strata and the following covariates associated with disease progression: age, sex, tobacco use, mean arterial pressure, baseline use of angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, and baseline use of aldosterone antagonists. Third, we conducted a complete-case analysis. In a fourth sensitivity analysis of the co-primary outcomes, we used different methods to impute missing data due to death on the co-primary outcomes. Finally, we conducted a sensitivity analysis of all continuous outcomes in which we adjusted for the baseline value instead of analyzing the absolute change score.
Pre-specified subgroup analyses
We conducted an exploratory subgroup analysis by including an interaction term in the linear regression models for a higher versus lower 2-year estimated risk of kidney failure [20] and each co-primary outcome. The geometric mean ratio is reported for each Kidney Failure Risk Equation group, and the difference in change in eGFR is reported for each kidney failure risk group, along with 95% CIs.
RESULTS
Participants
Trial enrollment began in September 2015 and the last patient was randomized in November 2019; the last day of follow-up was 15 May 2020. We enrolled 533 participants from two centers in Ontario, Canada. We randomly assigned 265 participants to receive micro-particle curcumin and 268 to placebo. The flow from screening to study conclusion is shown in Fig. 1. Following exclusions, 261 participants remained in the curcumin group and 254 in placebo group. At baseline, participants had a median age 65 years and 75% were male; the median albumin-to-creatinine ratio was 79 mg/mmol (698 mg/g) and the median eGFR was 33 mL/min/1.73 m2. The baseline characteristics of participants in the curcumin and placebo groups were similar (Table 1).
Figure 1:
Screening, randomization and flow of patients in the MPAC-CKD trial.
Table 1:
Baseline characteristics.
Curcumin (n = 261) | Placebo (n = 254) | |
---|---|---|
Demographics | ||
Age, years, median (SD) | 65 (12) | 65 (11) |
Male, n (%) | 201 (77) | 185 (73) |
Self-identified race, n (%) | ||
White/Caucasian | 245 (94) | 228 (90) |
Black/African American/African Canadian | 1 (<1) | 5 (2) |
Hispanic/Latino | 3 (1) | 2 (<1) |
Asian (Far East, Southeast Asia, Indian) | 6 (2) | 5 (2.0) |
Middle East | 3 (1) | 6 (2) |
Aboriginal/Native Persons/American Indian | 2 (<1) | 6 (2) |
Other | 1 (<1) | 2 (<1) |
Physiological measures | ||
Body mass index, median kg/m2 (SD) | 32 (6) | 33 (7) |
Systolic blood pressure, median mmHg (SD) | 142 (19) | 141 (18) |
Diastolic blood pressure, median mmHg (SD) | 77 (11) | 76 (12) |
Kidney function | ||
Serum creatinine, median μmol/L (SD) | 194 (63) | 197(63) |
eGFR median mL/min/1.73m2 (SD)a | 33 (12) | 32 (12) |
eGFR categories, n (%)a | ||
≥60 mL/min/1.73 m2 | 0 (0) | 1 (<1)f |
45–60 mL/min/1.73 m2 | 49 (19) | 47 (19) |
30–44 mL/min/1.73 m2 | 82 (31) | 73 (29) |
15–29 mL/min/1.73 m2 | 128 (49) | 131 (52) |
<15 mL/min/1.73 m2 | 2 (<1)f | 2 (<1)f |
Albumin-to-creatinine ratiob median mg/mmol (IQR) | 80 (41, 155) | 78 (44, 154) |
Albumin-to-creatinine ratiob median g/g (IQR) | 0.71 (0.36, 1.4) | 0.69 (0.39, 1.4) |
Albumin-to-creatinine ratiob categories, n (%) | ||
<30 mg/mmol (<0.27 g/g) | 40 (15)f | 34 (13)f |
30–99 mg/mmol (0.27–0.88 g/g) | 113 (43) | 117 (46) |
100–299 mg/mmol (0.88–2.7 g/g) | 86 (33) | 83 (33) |
≥300 mg/mmol (≥2.7 g/g) | 20 (8) | 19 (7) |
Missing | 2 (<1) | 1 (<1) |
2-year KFRE categoriesc, n (%) | ||
<5% | 88 (34) | 86 (34) |
5–9% | 43 (16) | 31 (12) |
10%–14% | 22 (8) | 21 (8) |
15%–19% | 26 (10) | 25 (10) |
20%–24% | 18 (7) | 15 (6) |
25%–29% | 12 (5) | 14 (6) |
≥30% | 50 (19) | 61 (24) |
Missing | 2 (<1) | 1 (<1) |
Comorbidities | ||
Diabetes, n (% of whole) | 157 (60) | 150 (59) |
Glycated hemoglobin %, median % (SD) | 7.5 (1.4) | 7.7 (1.5) |
Glycated hemoglobin %, n (% of diabetes) | ||
<6.5% | 39 (25) | 27 (18) |
6.5%–6.9% | 27 (17) | 28 (19) |
7.0%–7.4% | 25 (16) | 26 (17) |
7.5%–7.9% | 13 (8) | 20 (13) |
≥8.0% | 53 (34) | 49 (33) |
Insulin, n (% of diabetes) | 48 (31) | 37 (25) |
Oral hypoglycemic, n (% diabetes) | 46 (29) | 48 (32) |
Insulin and oral hypoglycemic, n (% of diabetes) | 54 (34) | 57 (38) |
No hypoglycemic medications, n (%) | 9 (6) | 8 (5) |
History of smoking, n (%) | ||
Current | 30 (11) | 32 (13) |
Remoted | 40 (15) | 43 (17) |
Never | 191 (73) | 179 (70) |
Primary cause of CKD, n (%)e | ||
Diabetic nephropathy | 128 (49) | 126 (50) |
Hypertensive nephrosclerosis | 22 (8) | 19 (7) |
Glomerulonephritis | 44 (17) | 39 (15) |
Polycystic kidney disease | 4 (2) | 2 (<1) |
Other | 63 (24) | 68 (27) |
Medication use, n (%) | ||
ACE inhibitors | 107 (41) | 120 (47) |
Angiotensin II receptor blockers | 111 (43) | 91 (36) |
ACE inhibitors or angiotensin II receptor blockers | 215 (82) | 207 (82) |
Diuretics | 176 (67) | 159 (63) |
Calcium channel blockers | 160 (61) | 156 (61) |
Beta blockers | 133 (51) | 114 (45) |
Aspirin | 141 (54) | 126 (50) |
Antiplatelet drugs (not including aspirin) | 19 (7) | 22 (9) |
Anti-inflammatories (regularly scheduled use) | 7 (3) | 6 (2) |
Oral glucocorticoids | 13 (5) | 15 (6) |
Cholesterol lowering medications | 205 (79) | 192 (76) |
Opioid analgesic | 48 (18) | 39 (15) |
Non-warfarin anticoagulant | 13 (5) | 11 (4) |
Continuous data are summarized as median (25th, 75th percentiles) unless otherwise specified.
CKD Epidemiology Collaboration eGFR equation [25].
Albuminuria (24-hour urine protein ≥ 300 mg, or UACR ≥ 30 mg/mmol) measured in clinic is a criterion for inclusion into MPAC-CKD; however, the post-randomization, pre-medication first-morning urine albumin-to-creatinine ratio measurement (average of two) may not meet this criterion.
Kidney Failure Risk Equation [20].
Participants who smoked within the last 10 years but were currently non-smokers.
Primary cause of CKD as diagnosed by physician in medical notes.
Participants were required to have an eGFR between 15 and 60 mL/min/1.73 m2 and a urine albumin-to-creatinine ratio more than 30 mg/mmol (0.27 g/g) within the 3 months before randomization. Although all participants met eligibility criteria at the time of enrollment, 2% had an eGFR outside the trial's criteria and 15% had urine albumin-to-creatinine ratios <30 mg/mmol (0.27 g/g) at the time of randomization.
ACE: angiotensin-converting enzyme; IQR: interquartile range; KFRE: Kidney Failure Risk Equation; SD: standard deviation.
Co-primary outcomes: change in albuminuria and eGFR
Table 2 shows the changes in albuminuria and eGFR from baseline to 6 months post-randomization. The 6-month change in albuminuria did not differ significantly between the curcumin and placebo groups (geometric mean ratio 0.94, 97.5% CI 0.82 to 1.08, P = .32). Similarly, the 6-month change in eGFR did not differ significantly between the curcumin and placebo groups (mean between-group difference –0.22 mL/min/1.73 m2, 97.5% CI –1.38 to 0.95, P = .68). These results were consistent in four sensitivity analyses of these outcomes (Supplementary data, Tables S5–S8) and in the sensitivity analysis of 6-month eGFR adjusted for baseline eGFR (Supplementary data, Table S9).
Table 2:
Co-primary outcomes: change in albuminuria and eGFR.
Intervention group | ||||
---|---|---|---|---|
Outcomes | Curcumin (n = 261) | Placebo (n = 254) | Between-group difference (97.5% CI) | P-valuec |
Albumin-to-creatinine ratio, mg/mmola | Geometric mean (geometric SD) | Geometric mean ratiob | ||
Baseline | 78.3 (2.5) | 76.8 (2.8) | ||
3 months | 69.5 (2.7) | 70.3 (3.0) | 0.99 (0.88 to 1.12) | .91 |
6 months | 67.0 (2.8) | 67.9 (3.0) | 0.94 (0.82 to 1.08) | .32 |
eGFR, mL/min/1.73 m2 | Mean (SD) | Mean difference | ||
Baseline | 32.5 (12.5) | 31.6 (12.2) | ||
3 months | 30.9 (13.2) | 30.4 (12.7) | ||
6 months | 29.7 (13.1) | 29.4 (12.9) | ||
3-month changed | –1.3 (5.1) | –1.1 (4.8) | –0.21 (–1.21 to 0.79) | .64 |
6-month changed | –2.8 (5.8) | –2.5 (5.7) | –0.22 (–1.38 to 0.95) | .68 |
To convert to mg/g, multiply by 8.84.
The geometric mean ratio was estimated using linear regression, where the outcome is the log-transformed average of two 3-month or two 6-month follow-up visit urinary albumin-to-creatinine ratio measurements minus the log-transformed average of two pre-randomization urinary albumin-to-creatinine ratio measurements, adjusting for a treatment indicator variable and the log-transformed average of two pre-randomization urinary albumin-to-creatinine ratio measurements.
The P-values for the two co-primary outcomes were estimated using the Hochberg procedure. We compared the largest of the two P-values to α = 0.05; because it was non-significant, we compared the smaller P-value with α/2 = 0.025.
Change was calculated as the 3- or 6-month value minus the baseline value.
SD: standard deviation.
Secondary outcomes
Table 3a shows that treatment with curcumin had no significant effect on quality of life as measured by the Physical or Mental Component Summary Scores of the SF-36 Health Survey. There was also no significant effect on the percentage of glycated hemoglobin in a subgroup of participants with diabetes mellitus. These results were consistent in four sensitivity analyses of these outcomes (Supplementary data, Tables S5–S7 and S9). Table 3b shows that treatment with curcumin for 6 months had no significant effect on the risk of the composite outcome of progressive CKD, kidney failure or death.
Table 3a:
Secondary outcomes: health-related quality of life and glycated hemoglobin.
Curcumin (n = 261) | Placebo (n = 254) | Between-group difference in absolute change (95% CI) | |
---|---|---|---|
36-Item Short Form Health Survey | |||
Physical component summary score, mean absolute change (SD) | 0.4 (8.6) | –0.1 (9.8) | 0.6 (–1.1 to 2.3) |
Mental component summary score, mean absolute change (SD) | –1.2 (10.7) | –0.9 (11.1) | –0.2 (–2.2 to 1.7) |
Percentage of glycated hemoglobin, mean absolute change (SD) | –0.03 (1.14) | –0.27 (1.20) | 0.26 (–0.03 to 0.54) |
SD: standard deviation.
Table 3b:
Secondary outcomes: progressive kidney disease.
Curcumin (n = 261) | Placebo (n = 254) | Cause-specific hazard ratio (95% CI) | |
---|---|---|---|
Composite of progressive CKDa, kidney failureb or death | 0.89 (0.60 to 1.33) | ||
Number (%) of events | 47 (18.0) | 49 (19.3) | |
Time to event, median (25th, 75th percentiles) | 110 (75, 176) | 131 (81, 167) | |
Components of the composite outcome | |||
Progressive CKDa | 0.82 (0.54 to 1.25) | ||
Number (%) of events | 42 (16.1) | 47 (18.5) | |
Time to event, median (25th, 75th percentiles) | 144 (82, 181) | 132 (81, 168) | |
Kidney failureb | 0.60 (0.35 to 1.04) | ||
Number (%) of events | 21 (8.1) | 32 (12.6) | |
Time to event, median (25th, 75th percentiles) | 96 (71, 188) | 132 (80, 167) | |
Death | 1.63 (0.39 to 6.80) | ||
Number (%) of events | 5 (1.9) | 3 (1.2) | |
Time to event, median (25th, 75th percentiles) | 100 (62, 145) | 125 (55, 134) |
Progressive CKD: loss of ≥30% of eGFR compared with baseline or new kidney failure.
Kidney failure: decrease of eGFR to <15 mL/min/1.73 m2 or the initiation of renal replacement therapy.
Post hoc exploratory subgroup analyses
Table 4 shows that we found no evidence of statistical interaction between a participant's 2-year risk of kidney failure and the effect of curcumin on either albuminuria or change in eGFR.
Table 4:
Exploratory post hoc subgroup analysis by KFRE categories.a
Albumin-to-creatinine ratio, geometric mean (geometric SD) | Curcumin (n = 261) | Placebo (n = 254) | Geometric mean ratiob (95% CI) | P-value for interaction |
---|---|---|---|---|
KFRE ≥10%, n = 255 | 89.7 (2.5) | 95.0 (2.6) | 0.96 (0.79 to 1.16) | .84 |
KFRE <10%, n = 257 | 52.8 (2.9) | 47.9 (3.0) | 0.93 (0.77 to 1.07) | |
Change in eGFR, absolute change (SD) | Curcumin (n = 261) | Placebo (n = 254) | Between-group difference in absolute change (95% CI) | P-value for interaction |
KFRE ≥10%, n = 255 | –2.6 (3.7) | –2.3 (3.9) | –0.33 (–1.99 to 1.33) | .81 |
KFRE <10%, n = 257 | –2.8 (7.3) | –2.6 (7.2) | –0.07 (–1.72 to 1.57) |
Three participants were not included in this analysis because their KFRE risk scores could not be calculated.
The effect estimate is the geometric mean ratio at 6 months post-randomization. We estimated this using linear regression, where the outcome is the log-transformed average of two 6-month follow-up visit urine albumin-to-creatinine ratio measurements minus the log-transformed average of two pre-treatment urine albumin-to-creatinine ratio measurements, adjusting for a treatment indicator variable and the log-transformed average of two baseline urine albumin-to-creatinine ratio measurements.
KFRE: kidney failure risk equation; SD: standard deviation.
Intervention adherence, study withdrawal and adverse events
Intervention adherence, study withdrawal and adverse events were similar in the curcumin and placebo groups, as shown in Supplementary data, Tables S10, S11 and S12, respectively. The median number of missed doses in each group was of 0 (interquartile ranges 0–6 and 0–4, respectively) and 48% of participants in each group missed at least one dose. Adverse events were rarely experienced, minor and not significantly different between the two groups. The one exception was a yellow discoloration of the urine, which occurred in 9 (4%) of participants receiving placebo and only 1 (<1%) receiving curcumin. This was likely a result of the yellow dye used to match the color of the contents of the placebo capsule to that of curcumin.
DISCUSSION
In patients with albuminuric CKD, 90 mg of micro-particle curcumin daily for 6 months had no significant effect on albuminuria or eGFR. We found no evidence to support a larger-scale trial of micro-particle curcumin for the prevention CKD progression.
Our findings stand in contrast to numerous animal studies (Supplementary data, Fig. S1) and three clinical trials involving patients with CKD (Supplementary data, Table S1). In two separate placebo-controlled trials, Khajehdehi et al. found significant reductions in albuminuria using 500 mg of turmeric (22 mg of curcumin) daily for 3 months, first among 40 patients with diabetic nephropathy and later among 24 patients with lupus nephritis [26, 27]. In a trial more comparable to MPAC-CKD, Vanaie et al. found improvements in albuminuria among 46 patients with albuminuric CKD randomized to curcumin 500 mg once daily or placebo for 16 weeks [28]. However, the curcumin group had a much higher mean baseline level of albuminuria than the placebo group (900 versus 520 mg/day), which may have allowed regression to the mean to account for the trial's findings. Conversely, two clinical trials found no effect of curcumin on kidney outcomes. Jiménez-Osorio et al. randomized 101 patients with albuminuric CKD to curcumin 320 mg/day or placebo [29]. Exposure to curcumin was found to have no significant effect on albuminuria or kidney function. A small pilot study published the following year also showed no effect on albuminuria in 16 patients with CKD [30]. MPAC-CKD substantially extends these findings by excluding differences in albuminuria that are used to define effective therapies.
MPAC-CKD is the largest clinical trial to date to evaluate curcumin's therapeutic potential in patients with albuminuric CKD. However, our study's limitations should also be considered. MPAC-CKD included predominantly Caucasian men with diabetes, which may limit generalizability. However, curcumin is thought to exert its benefit through anti-inflammatory pathways, which are likely common across many causes of kidney disease regardless of sex or ethnicity. We did not exclude patients with advanced CKD, who may have had a limited capacity to respond to treatment. Although we reached our minimum sample size of 500 participants, we may still have lacked sufficient power to detect a biological effect of curcumin. However, the precision of our geometric mean of the urine albumin-to-creatinine ratio was sufficient to exclude 0.7, a ratio that predicted clinically meaningful outcomes in a patient-level data meta-analysis of multiple albuminuria-lowering interventions [31]. Our results were insufficiently precise to exclude a similarly predictive change in eGFR of 0.72 mL/min/1.73 m2/year [32]. While the confidence limits of the difference in eGFR include this value and our treatment period was limited to only 6 months, post hoc conditional power calculations suggest increasing the sample size by even 50% would have provided insufficient power to reliably detect such a difference. We did not collect serum creatinine levels frequently enough to rule out an acute effect on eGFR, as is observed with inhibitors of the renin-angiotensin system. The data available to inform effective curcumin dosing are sparse. We used a more bioavailable formulation of curcumin that has been proven to generate measurable plasma levels [12] and we provided doses equal to, or substantially higher than, those used in prior studies with positive findings [26, 27]; however, it remains possible that the dose we selected was too low to produce an observable benefit. We attempted to measure serum and urine concentrations of curcumin, but had significant difficulties producing reliable results at the detection limit of the mass spectrometer (eAppendix 1). This failure to detect serum curcumin levels using a higher dose of a curcumin formulation proven to generate serum levels suggests that the bioavailability of micro-particle curcumin may be lower in patients with CKD than the healthy volunteers studied by Sasaki et al. [12].
CONCLUSION
The findings of MPAC-CKD do not support the use of micro-particle curcumin to slow the progression of CKD.
Supplementary Material
Contributor Information
Matthew A Weir, Division of Nephrology, Department of Medicine, Western University, London, ON, Canada; London Health Sciences Centre, London, ON, Canada; Department of Epidemiology and Biostatistics, Western University, London, ON, Canada.
Michael Walsh, Departments of Medicine and Health Research Methods, Evidence & Impact, McMaster University, Hamilton, ON, Canada; Population Health Research Institute, Hamilton Health Sciences, McMaster University, Hamilton, ON, Canada.
Meaghan S Cuerden, London Health Sciences Centre, London, ON, Canada.
Jessica M Sontrop, London Health Sciences Centre, London, ON, Canada; Department of Epidemiology and Biostatistics, Western University, London, ON, Canada.
Bradley L Urquhart, Division of Nephrology, Department of Medicine, Western University, London, ON, Canada; Department of Physiology and Pharmacology, Schulich School of Medicine and Dentistry, Western University, London, ON, Canada.
Yong Jin Lim, Department of Physiology and Pharmacology, Schulich School of Medicine and Dentistry, Western University, London, ON, Canada.
Laura C Chambers, London Health Sciences Centre, London, ON, Canada.
Amit X Garg, Division of Nephrology, Department of Medicine, Western University, London, ON, Canada; London Health Sciences Centre, London, ON, Canada; Department of Epidemiology and Biostatistics, Western University, London, ON, Canada.
FUNDING
This work was supported by grants from the Kidney Foundation of Canada (KFOC130029) and the Canadian Institute of Health Research (CIHR365629). Theravalues® (Tokyo, Japan) and Natural Factors Nutritional Products Ltd (Coquitlam, BC, Canada) supplied investigational product and matching placebo as in-kind support. M.W. was supported by the Clive Kearon Mid-career Award from McMaster University Department of Medicine. A.X.G. is supported by a Clinician Investigator Award from the Canadian Institutes of Health Research and the Dr Adam Linton Chair in Kidney Health Analytics.
DATA AVAILABILITY STATEMENT
The data underlying this article will be shared on reasonable request to the corresponding author.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest. The results presented in this paper have not been published previously in whole or part.
REFERENCES
- 1. Hill NR, Fatoba ST, Oke JLet al. Global prevalence of chronic kidney disease - a systematic review and meta-analysis. PLoS One 2016;11:1–18. 10.1371/journal.pone.0158765 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Gansevoort RT, Correa-Rotter R, Hemmelgarn BRet al. Chronic kidney disease and cardiovascular risk: epidemiology, mechanisms, and prevention. Lancet 2013;382:339–52. 10.1016/S0140-6736(13)60595-4 [DOI] [PubMed] [Google Scholar]
- 3. Etgen T, Chonchol M, Förstl Het al. Chronic kidney disease and cognitive impairment: a systematic review and meta-analysis. Am J Nephrol 2012;35:474–82. 10.1159/000338135 [DOI] [PubMed] [Google Scholar]
- 4. Go AS, Chertow GM, Fan Det al. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 2004;351:1296–305. 10.1056/NEJMoa041031 [DOI] [PubMed] [Google Scholar]
- 5. Kurella Tamura M, Covinsky KE, Chertow GMet al. Functional status of elderly adults before and after initiation of dialysis. N Engl J Med 2009;361:1539–47. 10.1056/NEJMoa0904655 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Foote C, Kotwal S, Gallagher Met al. Survival outcomes of supportive care versus dialysis therapies for elderly patients with end-stage kidney disease: a systematic review and meta-analysis. Nephrology (Carlton) 2016;21:241–53. 10.1111/nep.12586 [DOI] [PubMed] [Google Scholar]
- 7. Perkovic V, Jardine MJ, Neal Bet al. Canagliflozin and renal outcomes in type 2 diabetes and nephropathy. N Engl J Med 2019;380:2295–306. 10.1056/NEJMoa1811744 [DOI] [PubMed] [Google Scholar]
- 8. Smith T, Kawa K, Eckl Vet al. Herbal supplement sales in US increase 8.5% in 2017, topping $8 billion. HerbalGram 2018;119:62–71. [Google Scholar]
- 9. Motaharinia J, Panahi Y, Barreto GEet al. Efficacy of curcumin on prevention of drug-induced nephrotoxicity: a review of animal studies. Biofactors 2019;45:690–702. 10.1002/biof.1538 [DOI] [PubMed] [Google Scholar]
- 10. Yaribeygi H, Simental-Mendía LE, Butler AEet al. Protective effects of plant-derived natural products on renal complications. J Cell Physiol 2019;234:12161–72. 10.1002/jcp.27950 [DOI] [PubMed] [Google Scholar]
- 11. Bahrami A, Atkin SL, Majeed Met al. Effects of curcumin on hypoxia-inducible factor as a new therapeutic target. Pharmacol Res 2018;137:159–69. 10.1016/j.phrs.2018.10.009 [DOI] [PubMed] [Google Scholar]
- 12. Sasaki H, Sunagawa Y, Takahashi Ket al. Innovative preparation of curcumin for improved oral bioavailability. Biol Pharm Bull 2011;34:660–5. 10.1248/bpb.34.660 [DOI] [PubMed] [Google Scholar]
- 13. Schulz KF, Altman DG, Moher Det al. CONSORT 2010 changes and testing blindness in RCTs. Lancet 2010;375:1144–6. 10.1016/S0140-6736(10)60413-8 [DOI] [PubMed] [Google Scholar]
- 14. Weir MA, Walsh M, Cuerden MSet al. Micro-particle curcumin for the treatment of chronic kidney disease-1: study protocol for a multicenter clinical trial. Can J Kidney Health Dis 2018;5:1–9. 10.1177/2054358118813088 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Gupta B, Kulshrestha VK, Srivastava RKet al. Mechanisms of curcumin induced gastric ulcer in rats. Indian J Med Res 1980;71:806–14. [PubMed] [Google Scholar]
- 16. Rasyid A, Rahman ARA, Jaalam Ket al. Effect of different curcumin dosages on human gall bladder. Asia Pac J Clin Nutr 2002;11:314–8. 10.1046/j.1440-6047.2002.00296.x [DOI] [PubMed] [Google Scholar]
- 17. Kim D-C, Ku S-K, Bae J-S.. Anticoagulant activities of curcumin and its derivative. BMB Rep 2012;45:221–6. 10.5483/BMBRep.2012.45.4.221 [DOI] [PubMed] [Google Scholar]
- 18. Tayyem RF, Heath DD, Al-Delaimy WKet al. Curcumin content of turmeric and curry powders. Nutr Cancer 2006;55:126–31. 10.1207/s15327914nc5502_2 [DOI] [PubMed] [Google Scholar]
- 19. Sontrop JM, Garg AX, Li Let al. Consecutive first-morning urine samples to measure change in the albumin-to-creatinine ratio: a pilot study of a home urine collection protocol. Can J Kidney Health Dis 2016;3:3. 10.1186/s40697-016-0095-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Tangri N, Stevens LA, Griffith Jet al. A predictive model for progression of chronic kidney disease to kidney failure. JAMA 2011;305:1553–9. 10.1001/jama.2011.451 [DOI] [PubMed] [Google Scholar]
- 21. Wolfe R, Carlin JB. Sample-size calculation for a log-transformed outcome measure. Control Clin Trials 1999;20:547–54. 10.1016/s0197-2456(99)00032-x [DOI] [PubMed] [Google Scholar]
- 22. Halbesma N, Kuiken D-S, Brantsma AHet al. Macroalbuminuria is a better risk marker than low estimated GFR to identify individuals at risk for accelerated GFR loss in population screening. J Am Soc Nephrol 2006;17:2582–90. 10.1681/ASN.2005121352 [DOI] [PubMed] [Google Scholar]
- 23. Hunsicker LG, Adler S, Caggiula Aet al. Predictors of the progression of renal disease in the Modification of Diet in Renal Disease Study. Kidney Int 1997;51:1908–19. 10.1038/ki.1997.260 [DOI] [PubMed] [Google Scholar]
- 24. Hochberg Y. A sharper Bonferroni procedure for multiple tests of significance. Biometrika 1988;75:800–2. 10.1093/biomet/75.4.800 [DOI] [Google Scholar]
- 25. Levey AS, Stevens LA, Schmid CHet al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150:604–12. 10.7326/0003-4819-150-9-200905050-00006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Khajehdehi P, Pakfetrat M, Javidnia Ket al. Oral supplementation of turmeric attenuates proteinuria, transforming growth factor-β and interleukin-8 levels in patients with overt type 2 diabetic nephropathy: a randomized, double-blind and placebo-controlled study. Scand J Urol Nephrol 2011;45:365–70. 10.3109/00365599.2011.585622 [DOI] [PubMed] [Google Scholar]
- 27. Khajehdehi P, Zanjaninejad B, Aflaki Eet al. Oral supplementation of turmeric decreases proteinuria, hematuria, and systolic blood pressure in patients suffering from relapsing or refractory lupus nephritis: a randomized and placebo-controlled study. J Ren Nutr 2012;22:50–7. 10.1053/j.jrn.2011.03.002 [DOI] [PubMed] [Google Scholar]
- 28. Vanaie A, Shahidi S, Iraj Bet al. Curcumin as a major active component of turmeric attenuates proteinuria in patients with overt diabetic nephropathy. J Res Med Sci 2019;24:77. 10.4103/jrms.JRMS_1055_18 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Jiménez-Osorio AS, García-Niño WR, González-Reyes Set al. The effect of dietary supplementation with curcumin on redox status and Nrf2 activation in patients with nondiabetic or diabetic proteinuric chronic kidney disease: a pilot study. J Ren Nutr 2016;26:237–44. 10.1053/j.jrn.2016.01.013 [DOI] [PubMed] [Google Scholar]
- 30. Shelmadine BD, Bowden RG, Moreillon JJet al. A pilot study to examine the effects of an anti-inflammatory supplement on eicosanoid derivatives in patients with chronic kidney disease. J Altern Complement Med 2017;23:632–8. 10.1089/acm.2016.0007 [DOI] [PubMed] [Google Scholar]
- 31. Heerspink HJL, Greene T, Tighiouart Het al. Change in albuminuria as a surrogate endpoint for progression of kidney disease: a meta-analysis of treatment effects in randomised clinical trials. Lancet Diabetes Endocrinol 2019;7:128–39. 10.1016/S2213-8587(18)30314-0 [DOI] [PubMed] [Google Scholar]
- 32. Inker LA, Heerspink HJL, Tighiouart Het al. GFR slope as a surrogate end point for kidney disease progression in clinical trials: a meta-analysis of treatment effects of randomized controlled trials. J Am Soc Nephrol 2019;30:1735–45. 10.1681/ASN.2019010007 [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
Data Availability Statement
The data underlying this article will be shared on reasonable request to the corresponding author.