Critique 225: Moderate alcohol consumption may lower the risk of chronic kidney disease — 8 April 2019

Hu EA, Lazo M, Rosenberg SD, Grams ME, Steffen LM, Coresh J, Rebholz CM.  Alcohol Consumption and Incident Kidney Disease: Results From the Atherosclerosis Risk in Communities Study.  J of Renal Nutrition, 2019, in press.  DOI:

Authors’ Abstract

Objective(s):  Moderate alcohol consumption has been found to be associated with lower risk of coronary heart disease and myocardial infarction, which share similar risk factors and pathophysiology with chronic kidney disease (CKD).  However, there is inconsistent evidence on the association between alcohol consumption and CKD.

Design and Methods:  We conducted a prospective analysis of 12,692 participants aged 45-64 years from the Atherosclerosis Risk in Communities (ARIC) study.  We categorized participants into 6 alcohol consumption categories: never drinkers, former drinkers, ≤1 drink per week, 2 to 7 drinks per week, 8 to 14 drinks per week, and ≥15 drinks per week based on food frequency questionnaire responses at visit 1 (1987-1989).  Incident CKD was defined as estimated glomerular filtration rate <60 mL/minute/1.73 m2 accompanied by ≥25% estimated glomerular filtration rate decline, a kidney disease-related hospitalization or death or end-stage renal disease.

Results:  During a median follow-up of 24 years, there were 3,664 cases of incident CKD. Current drinkers were more likely to be men, whites, and to have a higher income level and education level.  After adjusting for total energy intake, age, sex, race-center, income, education level, health insurance, smoking, and physical activity, there was no significant association between being a former drinker and risk of incident CKD.  Participants who drank ≤1 drink per week, 2 to 7 drinks per week, 8 to 14 drinks per week, and ≥15 drinks per week had, respectively, a 12% (hazard ratio [HR]: 0.88, 95% confidence interval [CI]: 0.79-0.97), 20% (HR: 0.80, 95% CI: 0.72-0.89), 29% (HR: 0.71, 95% CI: 0.62-0.83), and 23% (HR: 0.77, 95% CI: 0.65-0.91) lower risk of CKD compared with never drinkers.

Conclusion(s):  Consuming a low or moderate amount of alcohol may lower the risk of developing CKD. Therefore, moderate consumption of alcohol may not likely be harmful to the kidneys.

Forum Comments

While the majority of cohort studies have found that moderate consumers of alcohol tend to have lower risk of developing chronic kidney disease (CKD), potential mechanisms for such an effect are unclear.  The present large prospective, population-based, multi-cultural study provides important data on the effects of alcohol intake on the initial diagnosis of CKD among middle-aged adults over an extended follow-up period.

Specific comments by Forum members:  Forum members viewed the present analyses favourably, as it appears to be very well done, based on a very large number of incident cases of CKD (n=3,364) and with well-defined and appropriate diagnostic criteria for the outcome.  It includes data on both Caucasian and African-American subjects, with an adequate percentage (25%) of never drinkers to serve as a control group.  Reviewer Ellison thought it key that the main outcome was an initial diagnosis of CKD, which was based not only on a lower glomerular filtration rate (GFR) but also on a decrease in GFR from previous measurements over a follow-up period extending to 24 years.  He also noted: “There was good assessments of potential confounders (which, surprisingly, showed little effect).  For example, when the authors compared results from their Model 1 (adjusted only for total energy intake) with those from Model 2 (also adjusted for race, age, sex, and most other factors) and those from Model 3 (when they added diabetes, hypertension, and other potential intermediary factors), the estimate for the 8-14 drinks/week group changed only from 0.74 to 0.71, to 0.72.  This suggests that the effect of alcohol on the risk of CKD is not primarily through its effects on glucose metabolism or blood pressure.”

Forum member Puddey noted: “The authors of the current paper rightly identify the inconsistency of previous reports of the effects of moderate alcohol consumption on incidence of chronic kidney disease.  This relatively large study provides increased support, however, for the contention that mild to moderate consumption of alcohol may be conferring a reno-protective effect which they speculate is mediated through decreased renal arteriosclerosis.  They also raise the possibility that an alcohol-related decrease in incidence of diabetes mellitus could be responsible, although at least with heavier consumption of alcohol, an increased rather than a decreased incidence of type II diabetes mellitus has previously been reported in the ARIC study in men (Kao et al).

“The results of the present study concur with results of other recent large prospective population-based studies (Koning et al), as well as a recent meta-analysis of the effects of higher alcohol intake on the risk of renal damage (Cheungpasitporn et al). However, they remain in some ways surprising given that at similar levels of alcohol intake an increase in blood pressure is usually seen (Roerecke et al).  We can conjecture that effects of alcohol to beneficially modify other cardiovascular risk factors might therefore be counteracting any potential negative impact of alcohol-related hypertension on the kidney, with a decrease in intrarenal atherosclerotic vascular changes and improved overall renal outcomes in mild to moderate drinkers.  The association of alcohol with decreased incidence of chronic kidney disease was J-shaped in this study and with heavier drinking (>20 drinks/week) was no longer significant.”

Reviewer Ellison also noted: “It appears from the data presented that there were few subjects reporting above 20 – 25 drinks/week, and there were even fewer who developed CKD.  With so little data to support a clear increase in risk for the few subjects reporting more than a couple of drinks/day, it suggests to me that the relation of moderate alcohol intake to CKD may be more of a U-shaped than a J-shaped relation, although studies of heavier drinkers favour an increase in risk.”  He added: “It would be beneficial if all investigators could separate subjects with alcohol use disorders (AUD), especially those with overt alcoholism, from otherwise moderate drinkers who might report slightly more than the recommended levels on occasion.  There are many differences between characteristics of usual moderate drinkers and those with AUDs, which make attribution of cause of differences in risk more difficult for epidemiologists to explore.  Weaknesses of this study include their use of only baseline drinking as the exposure variable, no data on type of beverage or pattern of drinking, and not being able to identify “probable under-reporters” of intake or those with AUD.”

Forum member Skovenborg wrote: “I agree that the study is well planned and performed with a study population including 25% never drinkers and the participants were extensively characterized, making a thorough adjustment for confounders possible.  A weakness is the assessment of alcohol consumption at baseline only.  The drinking pattern was not established; a Japanese study has shown that among Japanese men those who drank a middle-range quantity of alcohol, specifically those who drank 4-7 days/week, had lower risk of CKD than non-drinkers (Sato et al).  The association between alcohol intake and CKD is probably J-shaped; however, a recent meta-analysis found an inverse association between high alcohol consumption and risk for developing CKD in males (Cheungpasitporn et al).

Potential mechanisms for a possible protective effect of moderate alcohol on CKD:  Reviewer Puddey added: “It’s worth recalling that effects of alcohol on the kidney have long been a subject of debate and contention.  Reports from the nineteenth century cited chronic or heavy consumption of alcohol as a cause of “those types of Bright’s disease which are associated with the granular contracting kidney” as well as “gouty kidney” (Roberts), findings which would not surprise given the links between heavy alcohol intake and both hypertension and gout.  Others, however, found no evidence from autopsies of increased kidney disease in intemperate drinkers (Dickinson).  A report from a case control study from Maryland (on which I was a co-author) (Perneger et al) found an increased risk of end stage renal disease with alcohol intakes >2 drinks per day but this was independent of any association with alcohol-related hypertension.  Several other reports (Klein et al, Vriz et al) have suggested that heavy alcohol use can increase the risk of microalbuminuria, an effect which, at least in the HARVEST study, appeared to be mediated by alcohol-related hypertension as measured by 24-hr ambulatory blood pressure (Vriz et al).”

Additional mechanisms to explore:  Reviewer Skovenborg also noted: “Few studies seem to have included medication as a confounding/mediating factor.  A recent analysis of 2003-2004 NHANES data found statistically significant increased odds of renal dysfunction among respondents who self-reported therapeutic doses of acetaminophen and light-moderate amount of alcohol [OR=1.64(1.28–2.10) even after adjusting for hypertension, diabetes and obesity; adjusted OR (95% CI)=1.78 (1.22–2.58)] (Ndetan et al). The toxic effects of acetaminophen and alcohol on the kidney were hypothesized but no plausible biological pathway was proposed and the threshold doses at which these effects begin to occur are unknown. The findings of this study suggest that even therapeutic doses of acetaminophen and light-moderate amount of alcohol could be health problematic if consumed concomitantly.”

References from Forum critique

Cheungpasitporn W, Thongprayoon C, Kittanamongkolchai W, Brabec BA, O’Corragain OA, Edmonds PJ et al.  High alcohol consumption and the risk of renal damage: a systematic review and meta-analysis.  QJM 2015;108:539-548.

Dickinson WH. Alcohol as a cause of renal disease. Br Med J 1872, pp 573-575.

Kao WH, Puddey IB, Boland LL, Watson RL, Brancati FL.  Alcohol consumption and the risk of type 2 diabetes mellitus: Atherosclerosis Risk in Communities Study.  Am J Epidemiol 2001;154:748-757.

Klein R, Klein BE, Moss SE.  Prevalence of microalbuminuria in older-onset diabetes. Diabetes Care 1993;16:1325-1330.

Koning SH, Gansevoort RT, Mukamal KJ, Rimm EB, Bakker SJ, Joosten MM.  Alcohol consumption is inversely associated with the risk of developing chronic kidney disease. Kidney Int 2015;87:1009-1016.

Ndetan H, Evans Jr MW, Singal AK, Brunnerd LJ, Calhoun K, Singh KP.  Light to moderate drinking and therapeutic doses of acetaminophen: An assessment of risks for renal dysfunction.  Preventive Medicine Reports 2018;12:253-258.

Perneger TV, Whelton PK, Puddey IB, Klag MJ.  Risk of end-stage renal disease associated with alcohol consumption.  Am J Epidemiol 1999;150:1275-1281.

Roberts W.  Alcohol as a cause of Bright’s Disease; and Dr. Dickinson’s statistics.  Br Med J 1872, pp 653-654.

Roerecke M, Kaczorowski J, Tobe SW, Gmel G, Hasan OSM, Rehm J.  The effect of a reduction in alcohol consumption on blood pressure: a systematic review and meta-analysis. Lancet Public Health 2017;2:e108-e120.

Sato KK, Hayashi T, Uehara S, et al.  Drinking pattern and risk of chronic kidney disease: the Kansai Healthcare Study.  Am J Nephrol 2014;40:516-522.

Vriz O, Piccolo D, Cozzutti E, Milani L, Gelisio R, Pegoraro F, et al.  The effects of alcohol consumption on ambulatory blood pressure and target organs in subjects with borderline to mild hypertension.  Am J Hypertens 1998;11:230-234.

Forum Summary

While numerous studies have found that moderate consumers of alcohol tend to be at a lower risk of developing chronic kidney disease (CKD), potential mechanisms for such an effect are unclear.  The present large prospective, population-based, multi-cultural study provides important additional data on the effects of alcohol intake on the initial diagnosis of CKD over many years.  It is based on data from the Atherosclerosis Risk in Communities Study and reports the relation of varying levels of alcohol consumption, versus non-drinking, to the incidence of CKD over a 24 year follow-up period.

Among the strengths of the study are the very large number of subjects (n=3,664) who developed CKD during follow up, using advanced methods for diagnosing the disease.  Only baseline alcohol intake was used as the exposure and type of beverage or pattern of drinking could not be assessed; however, the authors considered multiple appropriate potentially confounding variables in the analyses, including total energy intake, age, sex, race-center, income, education level, health insurance, smoking, and physical activity.  In comparison with the 25% of subjects in their analyses who were life-time abstainers, for subjects reporting all levels of alcohol consumption (≤1 drink per week, 2 to 7 drinks per week, 8 to 14 drinks per week, and ≥15 drinks per week), there was a statistically significant decrease in risk of incident CKD (from a 12% decrease in risk for the lowest category of alcohol intake to a 29% decrease for subjects reporting that they consumed 8-14 drinks/week).

The authors conclude:  “Consuming a low or moderate amount of alcohol may lower the risk of developing CKD.  Therefore, moderate consumption of alcohol may not likely be harmful to the kidneys.”  Forum members agree that this study supports much earlier research showing a protective effect of moderate drinking on the risk of kidney disease.  While mechanisms are not clear, effects on renal vessels that are similar to those described for coronary and cerebrovascular arteries may play a role.

*                      *                      *                      *                      *                      *

Contributions to this critique by the International Scientific Forum on Alcohol Research were provided by the following members:

Ian Puddey, MD, Emeritus Professor, Faculty of Health & Medical Sciences, University of Western Australia, Nedlands, Australia

Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark

R. Curtis Ellison, MD, Professor of Medicine, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA

Creina Stockley, PhD, MSc Clinical Pharmacology, MBA; Adjunct Senior Lecturer at the University of Adelaide, Australia (Formerly, Health and Regulatory Information Manager, Australian Wine Research Institute, Glen Osmond, South Australia)

Harvey Finkel, MD, Hematology/Oncology, Retired (Formerly, Clinical Professor of Medicine, Boston University School of Medicine, Boston, MA, USA)

Linda McEvoy, PhD, Department of Radiology, University of California at San Diego (UCSD), La Jolla, CA, USAMedicine, Boston University Medical Center, Boston, MA, USA)

Giovanni de Gaetano, MD, PhD, Department of Epidemiology and Prevention, IRCCS Istituto Neurologico Mediterraneo NEUROMED, Pozzilli, Italy