Alcohol Consumption, Specific Alcoholic Beverages, and Abdominal Aortic Aneurysm Otto Stackelberg, Martin Björck, Susanna C. Larsson, Nicola Orsini and Alicja Wolk Circulation. published online June 25, 2014; Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231 Copyright © 2014 American Heart Association, Inc. All rights reserved. Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circ.ahajournals.org/content/early/2014/06/25/CIRCULATIONAHA.113.008279
Data Supplement (unedited) at: http://circ.ahajournals.org/content/suppl/2014/06/25/CIRCULATIONAHA.113.008279.DC1.html
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation is online at: http://circ.ahajournals.org//subscriptions/
Downloaded from http://circ.ahajournals.org/ at Harvard University on July 1, 2014
DOI: 10.1161/CIRCULATIONAHA.113.008279
Alcohol Consumption, Specific Alcoholic Beverages, and Abdominal Aortic Aneurysm
Running title: Stackelberg et al.; Alcohol consumption and AAA Otto Stackelberg, MD1; Martin Björck, MD, PhD2; Susanna C Larsson, PhD1; Nicola Orsini, PhD1; Alicja Wolk, DMSc1
1
Unit of Nutritional Epidemiology, Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Stoc St ockh oc khol kh o m, S ol Sweden; weden; 2Dept of Surgical Sciences, Sciien encces, Section off Va V Vascular sccul ulaar Surgery, Uppsala Univ Univ i ersiity y, Uppsala, Upppsa psala, a,, Sweden Sweeden den University,
Add Ad Address dress for forr Correspondence: Corr Co rres espo pond nden ence c : O Otto S Stackelberg, k lb MD Unit of Nutritional Epidemiology Institute of Environmental Medicine, Karolinska Institutet Box 210 SE-171 77 Stockholm, Sweden Tel: +46 (0)8 524 867 95 Fax: +46 (0)8 304 571 E-mail: [email protected]
Journal Subject Codes: Etiology:[8] Epidemiology, Cardiovascular (CV) surgery:[35] CV surgery: aortic and vascular disease, Atherosclerosis:[135] Risk factors
1 Downloaded from http://circ.ahajournals.org/ at Harvard University on July 1, 2014
DOI: 10.1161/CIRCULATIONAHA.113.008279
Abstract
Background—Studies investigating the role of alcohol consumption in development of abdominal aortic aneurysm (AAA) are scarce. We aimed to examine associations between total alcohol consumption, and specific alcoholic beverages, with hazard of AAA. Methods and Results—44 715 men from the Cohort of Swedish Men and 35 569 women from the Swedish Mammography Cohort, aged 46 to 84 years at baseline 1998, constituted study population. Cox proportional hazards models were used to estimate hazard ratios (HR) with 95% confidence intervals (CI) for the associations between alcohol consumption, assessed through a Swedish food frequency questionnaire, and AAA, identified by means of linkage to the Swe wedi dish di sh IInpatient np pat atie ient until Register and the Swedish Vascular Registry (Swedvasc). Over the 14-year follow-up follow w-u up un nti till December 2011 (1 ( 019 954 person-years), AAA occurred in 1020 men and 194 women. Compared with alcohol/week AAA Co ompared mp d w ithh a co cconsumption nsum ns umpt um ptio pt ionn of 1 gglass lass la ss ooff al alco cohhol hol/weeek ((12 1 ggrams 12 ram ra ms ooff et ms eethanol), haanool) l),, th thee HR R ooff AA A A among Corresponding am mong on men who w o consumed wh conssum med d 100 glasses/week glassse gl s s/week k was was 0.80 0.80 (95% (95 95% % CI, CI, 0.68–0.94). 0..68–0 0.9 94) 4). Corres C o spon pondi nding HR among Among amon ongg women woome m n who who consumed consum co umed ed 5 glasses/week gla l sses/ s/we week ek was was 0.57 0.5 57 (95% (95% CI,, 0.40–0.82). 0.4 40– 0–0.82 82). ) A m ng mo participants total alcohol consumption part pa rtic icip ipan ants ts ffree reee fr re from om ccardiovascular ardi ar diov ovas asccul ular ar ddisease, isea is ease se to tota tall al alco coho holl co cons nsum umpt ptio ionn di didd no nott se seem em tto o be associated with hazard of the disease. The most commonly consumed alcoholic beverages – beer among men and wine among women – were inversely associated, whereas no association was observed for liquor. Conclusions—Moderate alcohol consumption, specifically wine and beer, was associated with a lower hazard of abdominal aortic aneurysm. The associations between higher doses of alcohol and risk of the disease remain unknown.
Key words: abdominal aortic aneurysm, epidemiology, risk factor, alcohol
2 Downloaded from http://circ.ahajournals.org/ at Harvard University on July 1, 2014
DOI: 10.1161/CIRCULATIONAHA.113.008279
Introduction While the effects of moderate alcohol consumption on cardiovascular diseases (CVD) are wellknown,1-4abdominal aortic aneurysm (AAA) remains an understudied vascular entity that ought not to be mislabeled for an atherosclerotic disease given the differences in pathogenesis and risk factors.5 To our best knowledge, there are only two previous prospective studies that have studied the relation between alcohol consumption and AAA risk. Both studies were conducted in selected male populations (restricted to men free from CVDs before and during the study period, and to smokers only) and demonstrated inconsistent results.6, 7 According to contemporary population-based data from northern Europe, AAAs are found in approximately 2.2% of 65 year old men,8-10 and in 0.4% of 70 year oldd wo women. wome menn. me n.11 T Those hose ho diagnosed with the condition and under surveillance with ultrasound can be offered surgical repair epa pair ir tto o pr prev prevent eventt th ev thee expanding dilatation of the aabdominal bddominal aorta ffrom r m ru ro rup rupturing. pturing.12, 13 The disease ac accounts cco oun u ts for 111 1 00 000 yyearly earrly ly ddeaths eath ea thss on only ly y iin n th the he U U.S .S S144 and ndd rremains em mains ains ann im impo important port po rtan an nt pu publ public blic bl icc hhealth e lt ea lthh problem prob pr oble ob lem le m in nm many anyy coun an ccountries, ounntrrie iess, ddespite espi es pite pi te iinternational ntterrna nati tion ti nal a differences difffe fere renc ncees nc es in in repair repa repa pair i incidence. ir inccid iden ence en ce.15 IIts ce ts eetiology tioolog olog gy plaain ned d bby y an a iinteraction n er nt e acctio i n off hhereditary ered er edit ed i ar it a y and and environmental envi en viro vi ro onm nmen e ta en tall risk risk factors fac a to tors rs of of which w ich male wh could be exp explained sex, advanced age, and smoking are considered to be the most significant.8, 16 However, many potential modifiable risk factors remain to be investigated.5, 16 We sought to investigate the associations between alcohol consumption, as well as specific alcoholic beverages, and hazard of AAA in two large population-based prospective cohorts of men and women.
Methods Study population Two population-based prospective cohorts constituted study populations for this study; the
3 Downloaded from http://circ.ahajournals.org/ at Harvard University on July 1, 2014
DOI: 10.1161/CIRCULATIONAHA.113.008279
Cohort of Swedish Men (COSM) and the Swedish Mammography Cohort (SMC). In brief, the SMC was established between 1987 and 1990, when all female residents in two counties in central Sweden (Västmanland and Uppsala), born between 1914 and 1948 (n = 90 303), were invited to complete a questionnaire on diet, alcohol consumption, anthropometrics, and education – a questionnaire of which 74% completed. In late autumn 1997, all women still alive and living in the study area (n = 56 030) received an expanded questionnaire including ~350 lifestyle related items. Except for some sex-specific questions, an identical questionnaire was sent simultaneously (1997) to all male residents in two counties in central Sweden (Västmanland and Örebro), born between 1918 and 1952 (n = 100 303), constituting the COSM. Information on smoking, moking, an inevitable confounder in the possible associations between alcohol consumption cons n um ns mpt ptio ionn and io a d an AAA hazard, was first included in the 1997 questionnaires which had a response rate of 70% for the SMC, questionnaires available online he SM SMC C, aand nd 49% 9% ffor or the COSM. All questionn naires are availabl blle on nli linne ne at http://ki.se/en/imm/unit-of-nutritional-epidemiology. ht p:/ http ://ki.se/en/ n/im im mm/un un nit it--of -of-n -nut utri ut riti tion onal on a -ep -epideemioloogy ogy. This Karolinska Th is study stu tuddy dy was wass approved app p ro ove v d by the thee regional reg egiiona iona nall Ethical Ethiccal Eth cal Review Revie evie iew w Board Booar a d at at K aro oli lins nskka ns ka IInstitutet, nstit stituutet ut t, wass co considered Stockholm, Sweden. Swe wede deen. n A completed com mpl pletted questionnaire quessti qu t on onna nair na iree wa ir w ons nsid ider id ered er ed to iimply mply mp ly iinformed n or nf orme medd co me cconsent. nsent. Ascertainment of cases Cases were defined as incident diagnosis, repair, or death due to AAA, identified by linkage of the cohorts to three national Swedish Registers. Thus, cases were clinical events and not based on routine screening of all study participants. Diagnosis was identified through the Swedish Inpatient Register, and death through the Swedish National Cause of Death Register, by use of the International Classification of Diseases (eight, ninth and tenth revision [ICD-8, 9, and 10]). Repair was identified through the Swedish Inpatient Register by use of the NOMESCO Classification of Surgical Procedures, and the Swedish National Registry for Vascular Surgery
4 Downloaded from http://circ.ahajournals.org/ at Harvard University on July 1, 2014
DOI: 10.1161/CIRCULATIONAHA.113.008279
(Swedvasc) by use of the integrated AAA module in that registry. Although no specific validity assessment for AAA has been performed in the Swedish Inpatient Register, it has a high validity in general17 and a nearly complete hospitalization coverage of the Swedish population since 1987.18 Surgical procedures have been reported as incorrect in 2%, and missing in 5.3%, of the records.17 Since the NOMESCO classification only classifies type of procedure performed, the Swedvasc further enabled information on rupture status and aneurysmal localization of repairs. The register, founded in 1987, covers all hospitals with a vascular service in Sweden since 1994 as well as 93.1% of all AAA repairs.19 All individual discharges, repairs, and deaths, due to intact (ICD-8: 4412; ICD-9: 4414; ICD-10: CD-10: I71.4) and ruptured (ICD-8: 4412; ICD-9: 4413; ICD-10: I71.3) aneury aneurysm ysm iin n th thee in infr infrafraa or suprarenal abdominal aorta and in the common or internal iliac arteries, were identified up to 31st 311st December Decem ecem embe b r 2011, be 2011 20 11, and were considered as cases cas ases as es for final analyses. analy yse s s.. Asssessment se of alcohol alcoho hol consumption ho cons co nsum ns umpt um pttio on Assessment of In n tthe he 11997 9977 fo 99 food od ffrequency requuen requ ncy qquestionnaire, uest ue stio st ionnna io nnaire airee, part pparticipants arttic i ip ipaantss rreported epoorte ep orteed aalcohol lco coho co hol cons ho cconsumption onssum mpt ptio ionn stat io sstatus tattus (never, never, former, form merr, current curr cu rren rr entt drinker), en d in dr nke ker) r),, frequency freq fr eq queenc ncyy of consumption c ns co nsum umpt um pttio ionn of sspecific p ciifi pe f c be beve beverages, vera ve rage ges, ge s, aand nd amounts consumed at a single occasion. Frequency during the past year was reported by use of eight predefined categories, ranging from never to 3 times per day, while amount of wine, beer, and liquor was reported in open-ended questions. Among current drinkers, missing values of frequency or amounts were assumed to mean the least possible consumption, i.e. 0–1 times/month for frequency and one standard glass for the amount. If both frequency and amount was missing for a beverage, zero consumption was assumed.20 After average weekly frequency of beverages had been calculated, frequency and amount was multiplied to obtain average weekly standard glass consumption. One standard glass (12 grams of ethanol), hereby referred to
5 Downloaded from http://circ.ahajournals.org/ at Harvard University on July 1, 2014
DOI: 10.1161/CIRCULATIONAHA.113.008279
as glasses, was calculated as 15 cl of wine, 8 cl of strong wine, 66 cl of class I beer (
The online version of this article, along with updated information and services, is located on the World Wide Web at: http://circ.ahajournals.org/content/early/2014/06/25/CIRCULATIONAHA.113.008279
Data Supplement (unedited) at: http://circ.ahajournals.org/content/suppl/2014/06/25/CIRCULATIONAHA.113.008279.DC1.html
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: http://www.lww.com/reprints Subscriptions: Information about subscribing to Circulation is online at: http://circ.ahajournals.org//subscriptions/
Downloaded from http://circ.ahajournals.org/ at Harvard University on July 1, 2014
DOI: 10.1161/CIRCULATIONAHA.113.008279
Alcohol Consumption, Specific Alcoholic Beverages, and Abdominal Aortic Aneurysm
Running title: Stackelberg et al.; Alcohol consumption and AAA Otto Stackelberg, MD1; Martin Björck, MD, PhD2; Susanna C Larsson, PhD1; Nicola Orsini, PhD1; Alicja Wolk, DMSc1
1
Unit of Nutritional Epidemiology, Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Stoc St ockh oc khol kh o m, S ol Sweden; weden; 2Dept of Surgical Sciences, Sciien encces, Section off Va V Vascular sccul ulaar Surgery, Uppsala Univ Univ i ersiity y, Uppsala, Upppsa psala, a,, Sweden Sweeden den University,
Add Ad Address dress for forr Correspondence: Corr Co rres espo pond nden ence c : O Otto S Stackelberg, k lb MD Unit of Nutritional Epidemiology Institute of Environmental Medicine, Karolinska Institutet Box 210 SE-171 77 Stockholm, Sweden Tel: +46 (0)8 524 867 95 Fax: +46 (0)8 304 571 E-mail: [email protected]
Journal Subject Codes: Etiology:[8] Epidemiology, Cardiovascular (CV) surgery:[35] CV surgery: aortic and vascular disease, Atherosclerosis:[135] Risk factors
1 Downloaded from http://circ.ahajournals.org/ at Harvard University on July 1, 2014
DOI: 10.1161/CIRCULATIONAHA.113.008279
Abstract
Background—Studies investigating the role of alcohol consumption in development of abdominal aortic aneurysm (AAA) are scarce. We aimed to examine associations between total alcohol consumption, and specific alcoholic beverages, with hazard of AAA. Methods and Results—44 715 men from the Cohort of Swedish Men and 35 569 women from the Swedish Mammography Cohort, aged 46 to 84 years at baseline 1998, constituted study population. Cox proportional hazards models were used to estimate hazard ratios (HR) with 95% confidence intervals (CI) for the associations between alcohol consumption, assessed through a Swedish food frequency questionnaire, and AAA, identified by means of linkage to the Swe wedi dish di sh IInpatient np pat atie ient until Register and the Swedish Vascular Registry (Swedvasc). Over the 14-year follow-up follow w-u up un nti till December 2011 (1 ( 019 954 person-years), AAA occurred in 1020 men and 194 women. Compared with alcohol/week AAA Co ompared mp d w ithh a co cconsumption nsum ns umpt um ptio pt ionn of 1 gglass lass la ss ooff al alco cohhol hol/weeek ((12 1 ggrams 12 ram ra ms ooff et ms eethanol), haanool) l),, th thee HR R ooff AA A A among Corresponding am mong on men who w o consumed wh conssum med d 100 glasses/week glassse gl s s/week k was was 0.80 0.80 (95% (95 95% % CI, CI, 0.68–0.94). 0..68–0 0.9 94) 4). Corres C o spon pondi nding HR among Among amon ongg women woome m n who who consumed consum co umed ed 5 glasses/week gla l sses/ s/we week ek was was 0.57 0.5 57 (95% (95% CI,, 0.40–0.82). 0.4 40– 0–0.82 82). ) A m ng mo participants total alcohol consumption part pa rtic icip ipan ants ts ffree reee fr re from om ccardiovascular ardi ar diov ovas asccul ular ar ddisease, isea is ease se to tota tall al alco coho holl co cons nsum umpt ptio ionn di didd no nott se seem em tto o be associated with hazard of the disease. The most commonly consumed alcoholic beverages – beer among men and wine among women – were inversely associated, whereas no association was observed for liquor. Conclusions—Moderate alcohol consumption, specifically wine and beer, was associated with a lower hazard of abdominal aortic aneurysm. The associations between higher doses of alcohol and risk of the disease remain unknown.
Key words: abdominal aortic aneurysm, epidemiology, risk factor, alcohol
2 Downloaded from http://circ.ahajournals.org/ at Harvard University on July 1, 2014
DOI: 10.1161/CIRCULATIONAHA.113.008279
Introduction While the effects of moderate alcohol consumption on cardiovascular diseases (CVD) are wellknown,1-4abdominal aortic aneurysm (AAA) remains an understudied vascular entity that ought not to be mislabeled for an atherosclerotic disease given the differences in pathogenesis and risk factors.5 To our best knowledge, there are only two previous prospective studies that have studied the relation between alcohol consumption and AAA risk. Both studies were conducted in selected male populations (restricted to men free from CVDs before and during the study period, and to smokers only) and demonstrated inconsistent results.6, 7 According to contemporary population-based data from northern Europe, AAAs are found in approximately 2.2% of 65 year old men,8-10 and in 0.4% of 70 year oldd wo women. wome menn. me n.11 T Those hose ho diagnosed with the condition and under surveillance with ultrasound can be offered surgical repair epa pair ir tto o pr prev prevent eventt th ev thee expanding dilatation of the aabdominal bddominal aorta ffrom r m ru ro rup rupturing. pturing.12, 13 The disease ac accounts cco oun u ts for 111 1 00 000 yyearly earrly ly ddeaths eath ea thss on only ly y iin n th the he U U.S .S S144 and ndd rremains em mains ains ann im impo important port po rtan an nt pu publ public blic bl icc hhealth e lt ea lthh problem prob pr oble ob lem le m in nm many anyy coun an ccountries, ounntrrie iess, ddespite espi es pite pi te iinternational ntterrna nati tion ti nal a differences difffe fere renc ncees nc es in in repair repa repa pair i incidence. ir inccid iden ence en ce.15 IIts ce ts eetiology tioolog olog gy plaain ned d bby y an a iinteraction n er nt e acctio i n off hhereditary ered er edit ed i ar it a y and and environmental envi en viro vi ro onm nmen e ta en tall risk risk factors fac a to tors rs of of which w ich male wh could be exp explained sex, advanced age, and smoking are considered to be the most significant.8, 16 However, many potential modifiable risk factors remain to be investigated.5, 16 We sought to investigate the associations between alcohol consumption, as well as specific alcoholic beverages, and hazard of AAA in two large population-based prospective cohorts of men and women.
Methods Study population Two population-based prospective cohorts constituted study populations for this study; the
3 Downloaded from http://circ.ahajournals.org/ at Harvard University on July 1, 2014
DOI: 10.1161/CIRCULATIONAHA.113.008279
Cohort of Swedish Men (COSM) and the Swedish Mammography Cohort (SMC). In brief, the SMC was established between 1987 and 1990, when all female residents in two counties in central Sweden (Västmanland and Uppsala), born between 1914 and 1948 (n = 90 303), were invited to complete a questionnaire on diet, alcohol consumption, anthropometrics, and education – a questionnaire of which 74% completed. In late autumn 1997, all women still alive and living in the study area (n = 56 030) received an expanded questionnaire including ~350 lifestyle related items. Except for some sex-specific questions, an identical questionnaire was sent simultaneously (1997) to all male residents in two counties in central Sweden (Västmanland and Örebro), born between 1918 and 1952 (n = 100 303), constituting the COSM. Information on smoking, moking, an inevitable confounder in the possible associations between alcohol consumption cons n um ns mpt ptio ionn and io a d an AAA hazard, was first included in the 1997 questionnaires which had a response rate of 70% for the SMC, questionnaires available online he SM SMC C, aand nd 49% 9% ffor or the COSM. All questionn naires are availabl blle on nli linne ne at http://ki.se/en/imm/unit-of-nutritional-epidemiology. ht p:/ http ://ki.se/en/ n/im im mm/un un nit it--of -of-n -nut utri ut riti tion onal on a -ep -epideemioloogy ogy. This Karolinska Th is study stu tuddy dy was wass approved app p ro ove v d by the thee regional reg egiiona iona nall Ethical Ethiccal Eth cal Review Revie evie iew w Board Booar a d at at K aro oli lins nskka ns ka IInstitutet, nstit stituutet ut t, wass co considered Stockholm, Sweden. Swe wede deen. n A completed com mpl pletted questionnaire quessti qu t on onna nair na iree wa ir w ons nsid ider id ered er ed to iimply mply mp ly iinformed n or nf orme medd co me cconsent. nsent. Ascertainment of cases Cases were defined as incident diagnosis, repair, or death due to AAA, identified by linkage of the cohorts to three national Swedish Registers. Thus, cases were clinical events and not based on routine screening of all study participants. Diagnosis was identified through the Swedish Inpatient Register, and death through the Swedish National Cause of Death Register, by use of the International Classification of Diseases (eight, ninth and tenth revision [ICD-8, 9, and 10]). Repair was identified through the Swedish Inpatient Register by use of the NOMESCO Classification of Surgical Procedures, and the Swedish National Registry for Vascular Surgery
4 Downloaded from http://circ.ahajournals.org/ at Harvard University on July 1, 2014
DOI: 10.1161/CIRCULATIONAHA.113.008279
(Swedvasc) by use of the integrated AAA module in that registry. Although no specific validity assessment for AAA has been performed in the Swedish Inpatient Register, it has a high validity in general17 and a nearly complete hospitalization coverage of the Swedish population since 1987.18 Surgical procedures have been reported as incorrect in 2%, and missing in 5.3%, of the records.17 Since the NOMESCO classification only classifies type of procedure performed, the Swedvasc further enabled information on rupture status and aneurysmal localization of repairs. The register, founded in 1987, covers all hospitals with a vascular service in Sweden since 1994 as well as 93.1% of all AAA repairs.19 All individual discharges, repairs, and deaths, due to intact (ICD-8: 4412; ICD-9: 4414; ICD-10: CD-10: I71.4) and ruptured (ICD-8: 4412; ICD-9: 4413; ICD-10: I71.3) aneury aneurysm ysm iin n th thee in infr infrafraa or suprarenal abdominal aorta and in the common or internal iliac arteries, were identified up to 31st 311st December Decem ecem embe b r 2011, be 2011 20 11, and were considered as cases cas ases as es for final analyses. analy yse s s.. Asssessment se of alcohol alcoho hol consumption ho cons co nsum ns umpt um pttio on Assessment of In n tthe he 11997 9977 fo 99 food od ffrequency requuen requ ncy qquestionnaire, uest ue stio st ionnna io nnaire airee, part pparticipants arttic i ip ipaantss rreported epoorte ep orteed aalcohol lco coho co hol cons ho cconsumption onssum mpt ptio ionn stat io sstatus tattus (never, never, former, form merr, current curr cu rren rr entt drinker), en d in dr nke ker) r),, frequency freq fr eq queenc ncyy of consumption c ns co nsum umpt um pttio ionn of sspecific p ciifi pe f c be beve beverages, vera ve rage ges, ge s, aand nd amounts consumed at a single occasion. Frequency during the past year was reported by use of eight predefined categories, ranging from never to 3 times per day, while amount of wine, beer, and liquor was reported in open-ended questions. Among current drinkers, missing values of frequency or amounts were assumed to mean the least possible consumption, i.e. 0–1 times/month for frequency and one standard glass for the amount. If both frequency and amount was missing for a beverage, zero consumption was assumed.20 After average weekly frequency of beverages had been calculated, frequency and amount was multiplied to obtain average weekly standard glass consumption. One standard glass (12 grams of ethanol), hereby referred to
5 Downloaded from http://circ.ahajournals.org/ at Harvard University on July 1, 2014
DOI: 10.1161/CIRCULATIONAHA.113.008279
as glasses, was calculated as 15 cl of wine, 8 cl of strong wine, 66 cl of class I beer (
