Table 1:
Trials of change in alcohol consumption and corresponding change in blood pressure meeting inclusion criteria
| Sex, age, country | Design, number of participants, and length of alcohol intervention | Inclusion and exclusion criteria | Alcohol intervention | Hypertension status at baseline | Blood pressure measurement | Quality assessment | ||
|---|---|---|---|---|---|---|---|---|
| Abe et al, 199448 | M, not reported, Japan | Crossover, n=14, 1 week | Essential hypertension with alcohol intake 30–120 mL per day; exclusion: secondary hypertension and cardiovascular, renal, hepatic, metabolic, and endocrine disorders | Hospital-based alcohol administration (1 mL/kg bodyweight, at dinner) vs non-alcoholic drinks (same calories, at dinner) | Hypertensive | 24 h | Low | |
| Aguilera et al, 199949 | M, 24–53 years, Spain |
Crossover, n=42, 4·5 weeks | Alcohol intake 100–380 g per day, admitted to the Alcohol Unit for voluntary alcohol detoxification | Hospital-based alcohol administration (total dose 2 g/kg) vs 1 month of abstinence (verified by interviews of relatives and GGT levels) | Normotensive | 24 h | Low | |
| Baros et al, 200822 | M/W, 44 years (mean), USA | Randomised parallel group, n=120, 12 weeks | Dependence on alcohol but not on other substances (except nicotine), no other major psychiatric diagnoses, medically stable, seeking outpatient treatment for alcoholism; liver enzymes (ALT, AST) less than 2·5 times the upper limit of normal | Naltrexone combined with either cognitive behavioural or motivational enhancement therapy for alcohol dependence; outcome: continued drinking vs abstinence | Mixed | Sitting | Low | |
| Chiva-Blanch et al, 201235 | M, 55–75 years, Spain | Randomised crossover, n=67, 4 weeks | Men at high cardiovascular risk (diabetes mellitus or ≥3 cardiovascular disease risk factors) | Common background diet plus red wine or gin (30 g alcohol per day) vs de-alcoholised red wine | Mixed, hypertensive and Normotensive subsamples |
Sitting | Low | |
| Cordain et al, 200050 | W, 30–50 years, USA | Randomised crossover, n=20, 10 weeks | Sedentary and overweight premenopausal women (BMI 27–33 kg/m2), alcohol intake two drinks per month to two drinks per week, willingness to consume two standard servings of red wine per day 5 days per week, for a total of 10 consecutive weeks; exclusion: health problems that can affect normal food intake and normal physical activity; use of any medications (including oral contraceptives) that can affect metabolism, appetite, or plasma lipids, glucose, and insulin; history of alcohol abuse or misuse; current alcohol intake greater than two standard servings per week; total avoidance of alcoholic beverages; use of supplemental omega-3 fatty acids; and participation in formal exercise more than two times per week | Red wine vs (near) abstinence | Normotensive | Sitting | Low | |
| Cox et al, 199323 | M, 20–45 years, Australia | Randomised parallel group, n=72, 4 weeks | Healthy, moderately drinking (≥210 mL per week), sedentary (<30 min vigorous-intensity exercise [energy expenditure >31·4 kJ/min] each week for 6 months before screening), BMI <30 kg/m2, systolic blood pressure 125–160 mm Hg, diastolic blood pressure <110 mm Hg | Usual alcohol intake vs reduced alcohol intake (substituted low alcohol beverages) | Mixed | Supine | High | |
| Cushman et al, 199824 | M/W, 21–79 years, USA | Randomised parallel group, n=549, 104 weeks | PATHS trial; outpatient veterans (average alcohol intake ≥three drinks per day), diastolic blood pressure 80–99 mm Hg; exclusion: alcohol or psychoactive substance dependence, alcohol-attributed medical complications, major psychiatric diagnoses, cardiovascular end-organ damage, severe or secondary hypertension, malignancies, seizure disorders, coagulopathies, or current pregnancy | Cognitive-behavioural alcohol reduction intervention programme or control observation; both groups reduced their alcohol consumption | Mixed, hypertensive subsample | Sitting | High | |
| Droste et al, 201330 | M/W, 63 years (mean), Luxembourg | Randomised parallel group, n=100, 20 weeks | Outpatients of the Department of Neurology and had undergone carotid and intra-cranial bitemporal colour-coded duplex sonography; inclusion criteria were age >30 years and the presence of plaques or stenosis without haemodynamic compromise (ie, <70%) in at least one common carotid artery, the carotid bifurcation or the internal carotid artery. Exclusion: history of ocular or cerebral ischaemia within the past 3 months, atrial fibrillation, a repeatedly measured systolic blood pressure >160 mm Hg | Common diet and exercise with red wine vs abstinence | Normotensive | 24 h | High | |
| Estruch et al, 201136 | M, 30–50 years, Spain | Randomised crossover, n=40, 4 weeks | Healthy men (alcohol intake 10–40 g per day) and no cardiovascular risk factors or receiving any medication or multivitamin or vitamin E supplements | Common diet and 30 g/day (red wine or gin) with dinner vs washout period (abstinence) | Normotensive | Office, details not reported | Low | |
| Flanagan et al, 200237 | M/W, 21–41 years, UK | Randomised crossover, n=21, 1 week | Healthy participants; exclusion: current diabetes or other current illness | Three units of alcohol daily for 1 week vs abstinence | Normotensive | Office, details not reported | Low | |
| Gepner et al, 20159 | M/W, 59 years (mean), Israel | Randomised parallel group, n=224, 104 weeks | Diagnosis of type 2 diabetes; exclusion: >one drink per week, personal or family history of addiction, smoking, stroke, or myocardial infarction; major surgery within the past 3 months; >two insulin injections per day or an insulin pump; triglyceride concentration >4·52 mmol/L (400 mg/dL), HbA1c level <6·4% or ≥10%; women with first-degree relatives with breast cancer; or pregnant women | White or red wine vs mineral water with dinner; beverages were provided | Mixed | Office, details not reported | High | |
| Gepner et al, 201610 | M/W, 57 years (mean), Israel | Randomised parallel group, n=54, 26 weeks | Age between 40 and 75 years, diagnosis of type 2 diabetes, alcohol abstainers (≤one drink per week), non-smokers, clinically stable, willingness to drink wine if so assigned by randomization, as part of a Mediterranean diet intervention | Common diet with dry red or white wine vs mineral water | Mixed | 24 h | High | |
| Hansen et al, 200525 | M/W, 38–75 years, Denmark | Randomised parallel group, n=69, 4 weeks | Healthy participant; exclusion: regular use of lipid lowering drugs, antihypertensives, and antioxidant supplements, uncommon dietary habits (eg, vegetarianism), and alcoholism. Major weight changes (43 kg) during intervention, elevated plasma concentrations (410 mg/L) of C-reactive protein | Red wine (men: 383 g alcohol/day, women: 255 g alcohol/day) vs water and grape extract tablets (wine-equivalent dose or half dose) or water and placebo tablets | Normotensive | Supine | Low | |
| Howes et al, 198638 | M, 18–35 years, UK | Randomised crossover, n=10, 1 week | Drinkers with less than 40 g per day usually | 0·8 g alcohol/kg bodyweight per day (taken between 1700 and midnight) vs abstinence | Normotensive | Supine | Low | |
| Hsieh et al, 199531 | M, 49 years (mean), Japan | Parallel, n=17, 4 weeks | Regular drinkers >40 g per day with untreated mild hypertension (sitting diastolic blood pressure 90–104 mm Hg ≥2 readings); exclusion: abnormal renal function, diabetes, serious liver dysfunction, known secondary causes of hypertension | Usual alcohol intake vs counselling to reduce alcohol intake as much as possible | Hypertensive | Supine | Low | |
| Kawano et al, 199839 | M, 36–76 years, Japan | Randomised crossover, n=34, 4 weeks | Habitually drinking (≥30 mL daily alcohol consumption) patients attending the Hypertension Clinic with essential hypertension | Usual drinking vs abstinence or reduced alcohol intake | Hypertensive | 24 h | Low | |
| Kawano et al, 199651 | M, 35–69 years, Japan | Crossover, n=16, 1 week | Mild-to-moderate hypertension; exclusion: serious cardiac, renal, or neurological disorders | Hospital-based alcohol administration (1 mL/kg bodyweight, at dinner) vs non-alcoholic drinks (same calories, at dinner) | Hypertensive | 24 h | Low | |
| Kim et al, 200934 | M/W, 30–65 years, USA | Crossover, n=20, 8 weeks | Insulin resistant, non-diabetic, not taking any medications known to affect carbohydrate metabolism; haematocrit >32%, ALT<2 times the upper limit of normal, and triglyceride concentration <4·5 mmol/L | 30 g/day (vodka or red wine) with dinner or before bedtime vs no alcohol itntake; beverages were provided | Normotensive | Office, details not reported | Low | |
| Lang et al, 199526 | M/W, 43 years (mean), France | Randomised parallel group, n=106, 104 weeks | Hypertensive (>140/90 mm Hg) and excessive drinkers (GGT >1·5 times normal); exclusion: planned departure or retirement in the next 2 years; diagnosis of secondary hypertension; severe liver disease (cirrhosis, alcoholic hepatitis, or alcohol related haemorrhage); high GGT not related to alcohol | Counselling to reduce alcohol intake (by trained physicians) vs continuing care (by physicians not trained); both groups reduced their alcohol consumption | Hypertensive | Sitting | High | |
| Maheswaran et al, 199227 | M, 44 years (mean), UK | Randomised parallel group, n=41, 8 weeks | Patients from hypertension clinic who regularly consumed more than 20 units of alcohol per week; exclusion: diastolic blood pressure exceeding 105 mm Hg at the time of recruitment, diabetes, known or suspected secondary causes of hypertension, diagnosed with alcoholism (problem with alcohol requiring referral to an alcohol addiction unit for admission and detoxification), having received advice previously and had reported reducing their alcohol consumption | Counselling to reduce alcohol intake vs no counselling | Hypertensive | Standing | Low | |
| Maiorano et al, 199552 | M, 46 years (mean), Italy | Crossover, n=15, 1 week | Normotensive men with history of heavy alcohol intake; exclusion: none | Hospital-based usual alcohol intake vs abstinence | Normotensive | 24 h | Low | |
| Mori et al, 201654 | M/W, 40–70 years, Australia | Randomised crossover, n=24, 4 weeks | Regular drinkers (men and postmenopausal women) with type 2 diabetes; women usually consumed 2–3 standard drinks per day (20–30 g per day) and men 3–4 standard drinks/day (30–40 g per day); exclusion criteria included type 1 diabetes, recent (<3 months) symptomatic heart disease, angina pectoris, history of myocardial infarction or stroke, peripheral vascular disease, major surgery 3 months or less, blood pressure >170/100 mm Hg, liver or renal disease (plasma creatinine >120 mmol/L), HbA1c >8·5% (>69 mmol/L), and current smokers or ex-smokers less than 2 years | Red wine vs equivalent volumes of dealcoholised red wine or water | Mixed | 24 h | Low | |
| Mori et al, 201553 | W, 24–45 years, Australia | Randomised crossover, n=24, 4 weeks | Regular, healthy, premenopausal, non-smoking drinkers. BMI <30 kg/m2, no history of hypertension, dyslipidaemia, diabetes mellitus, liver disease, or coronary, cerebrovascular or peripheral vascular disease, no clinical evidence of vascular disease, no medications (including aspirin, non-steroidal anti-inflammatory drugs, or the oral contraceptive pill) | Higher volume red wine (lower level drinkers, 146 g alcohol per week; higher level drinkers, 218 g alcohol per week) vs equal amounts of de-alcoholised red wine. Lower volume red wine (lower level drinkers, 42 g alcohol per week; higher level drinkers, 73 g alcohol per week) vs equal amounts of dealcoholised red wine | Normotensive | 24 h | Low | |
| Naissides et al, 200628 | W, 50–70 years, Australia | Randomised parallel group, n=45, 6 weeks | Moderately hypercholesterolaemic postmenopausal women; exclusion: hormone replacement therapy, lipid lowering medication, use of steroids and other agents that might influence lipid metabolism, use of warfarin, smoking, hyperthyroidism or hypothyroidism, diabetes mellitus, cardiovascular events within past 6 months, psychological unsuitability, major systemic diseases, gastrointestinal problems, proteinuria, liver and renal failure, apolipoprotein genotype (E2/E2 exclusion) | Common diet with red wine vs water or de-alcoholised red wine | Normotensive | Central | Low | |
| Parker et al, 199032 | M, 20–70 years, Australia | Randomised parallel group, n=59, 4 weeks | Stable, treated hypertension (systolic blood pressure ≥125–180, diastolic blood pressure <115 mm Hg), regular drinkers; regular treatment with antihypertensive drugs for at least the preceding 6 months, a minimum alcohol intake of 210 mL per week (about three standard drinks per day), no history of renal or hepatic disease or diabetes mellitus, not on current treatment with nonsteroidal anti-inflammatory drugs, and no history of a myocardial infarction, stroke, or coronary artery bypass surgery within the previous 12 months | Usual alcohol intake vs alcohol reduction (substitution with low alcohol content); beverages were provided | Hypertensive | Supine | High | |
| Puddey et al, 198540 | M, 25–55 years, Australia | Randomised crossover, n=46, 6 weeks | Healthy, normotensive regular drinkers (average alcohol intake ≥210 mL per week); exclusion: less than 210 mL alcohol per week, taking beta-blockers, chronic disease | Usual alcohol intake vs alcohol reduction (substitution with low alcohol content); beverages were provided | Normotensive | Supine | High | |
| Puddey et al, 198741 | M, 25–65 years, Australia | Randomised crossover, n=44, 6 weeks | Regular treatment with antihypertensive drugs for at least the preceding 6 months, minimum alcohol intake of 210 ml per week, no underlying renal disease | Usual alcohol intake vs alcohol reduction (substitution with low alcohol content); beverages were provided | Hypertensive | Supine | High | |
| Puddey et al, 199229 | M, 25–70 years, Australia | Randomised parallel group, n=86, 16 weeks | Overweight and moderately drinking men with minimum alcohol intake of 210 mL per week (about three standard drinks per day); body mass index of >25 kg/m2 or current weight greater than 120% of ideal weight for age; no current use of antihypertensive or nonsteroidal anti-inflammatory drugs; and no history of renal or hepatic disease, diabetes mellitus, myocardial infarction or coronary artery surgery, stroke, or substantial weight loss (>10 kg) in the preceding 12 months; blood pressure entry criteria (systolic blood pressure >130 mm Hg and <160 mm Hg, diastolic blood pressure >80 mm Hg and <105 mm Hg) | Usual alcohol intake vs alcohol reduction (substitution with low alcohol content); beverages were provided | Mixed | Supine | High | |
| Queipo-Ortuno et al, 201242 | M, 45–50 years, Spain | Randomised crossover, n=10, 3 weeks | Healthy, not receiving treatment for diabetes, hypertension, or dyslipidaemia, any acute or chronic inflammatory diseases, infectious diseases, viral infections, cancer, or a previous cardiovascular event at study entry, antibiotic therapy, prebiotics, probiotics, symbiotics, or vitamin supplements or any other medical treatment influencing intestinal microbiota during the 3 months before the study | Red wine or gin (30 g/day) vs de-alcoholised red wine or abstinence (initial washout period) | Mixed | Details not reported | Low | |
| Rakic et al, 199843 | M, 21–65 years, Australia | Randomised crossover, n=55, 4 weeks | Drinkers (210–500 mL alcohol per week [with >60% of total intake as beer]), no history of hypertension and no use of any drug affecting blood pressure, no liver, renal, and cardiovascular disorders and no hospitalisation for any medical or surgical illness during the preceding 3 months | Usual alcohol intake vs alcohol reduction (substitution with low alcohol content); beverages were provided | Normotensive | 24 h, supine | Low | |
| Shai et al, 200711 | M/W, 41–74 years, Israel | Randomised parallel group, n=91, 12 weeks | Type 2 diabetes, alcohol abstainers (≤one drink per week), non-smokers, clinically stable, and willingness to drink wine as part of a Mediterranean diet intervention. Exclusion: HbA1c <6.4% or >10%, insulin >2 injections/day or use of an insulin pump, fasting serum triglyceride ≥400 mg/dL, serum creatinine >2 mg/dL, liver dysfunction (≥3-fold increase in ALT and/or AST), evidence of severe diabetic complications (such as proliferative retinopathy or diabetic nephropathy), evidence of autonomic neuropathy manifesting as postural hypotension and/or hypoglycaemia unawareness, use of medications that might interact with moderate alcohol consumption, presence of active cancer and/or chemotherapy treatment in the past 3 years, presence of a major illness that might require hospitalisation, clinically assessed as having high potential of addictive behaviour or personal or family history of addiction or alcohol abuse, women with first degree relatives with breast cancer, pregnant or lactating women; and participation in another interventional trial | Common diet with initiation of alcohol intake (red or white wine with dinner) vs non-alcoholic diet malt beer with dinner; beverages were provided | Mixed | Sitting | High | |
| Ueshima et al, 199344 | M, 30–59 years, Japan | Randomised crossover, n=54, 3 weeks | Civil servants with systolic blood pressure >140 or diastolic blood pressure >90 mm Hg, more than 28 mL alcohol at least 4 times per week; exclusion: systolic blood pressure >179 or diastolic blood pressure >109 mm Hg, taking antihypertensive medication | Usual alcohol intake vs alcohol reduction (abstention or reduction as much as possible) | Hypertensive | Sitting | Low | |
| Ueshima et al, 198745 | M, 30–59 years, Japan | Randomised crossover, n=49, 2 weeks | Civil servants with blood pressure 140/90 mm Hg to 180/110 mm Hg; exclusion: diabetes with medication or less than 3 times a week alcohol consumption | Usual alcohol intake vs alcohol reduction (abstention or reduction as much as possible) | Hypertensive | Details not reported | Low | |
| Wallace et al, 198833 | M/W, 17–69 years, UK | Randomised parallel group, n=641, 52 weeks | Patients with excessive alcohol consumption (defined as at least 35 units/week in men and 21 units/week for women) | Common brief advice on general health (smoking, exercise, and diet) with counselling to reduce alcohol vs no counselling to reduce alcohol; both groups reduced their alcohol consumption | Mixed | Office, details not reported | High | |
| Zilkens et al, 200347 | M, 20–65 years, Australia | Randomised crossover, n=16, 4 weeks | Drinkers with 40–110 g/day, with more than 60% derived from beer; exclusion: smoking within the last 6 months, BMI >30, CVD (by clinical history, physical examination or electrocardiogram), diabetes mellitus, blood pressure >160/90 mm Hg or treatment with antihypertensive agents, total cholesterol >7.5 mmol/L or use of lipid-decreasing agents, aspirin or non-steroidal anti-inflammatory drugs | Usual alcohol intake vs alcohol reduction (substitution with low alcohol content); beverages were provided | Normotensive | Supine | Low | |
| Zilkens et al, 200546 | M, 39–65 years, Australia | Randomised crossover, n=24, 4 weeks | Regular drinkers with 30–60 g/day; exclusion: smoking within the last 6 months, BMI >30, cardiovascular disease (by clinical history, physical examination or electrocardiogram), diabetes, blood pressure >160/90 mm Hg or antihypertensive medication, total cholesterol >7.5 mmol/L or use of lipid-decreasing agents, aspirin or non-steroidal anti-inflammatory drugs | Red wine or beer vs abstinence or de-alcoholised red wine | Normotensive | 24 h | Low | |
M=men. W=women. M/W=men and women combined. M,W=men and women stratified. ALT=alanine aminotransaminase. AST=aspartate aminotransaminase. GGT=γ-glutamyl transferase. BMI=body-mass index. CVD=cardiovascular disease.