Advertisement

Blood Type Is Not Associated with Changes in Cardiometabolic Outcomes in Response to a Plant-Based Dietary Intervention

Open AccessPublished:December 04, 2020DOI:https://doi.org/10.1016/j.jand.2020.08.079

      Abstract

      Background

      Risk of cardiovascular disease is lower in individuals with blood type O and increased in individuals with blood type A, compared with those in other blood groups. However, little evidence is available regarding whether individuals with different blood types benefit from different diet recommendations.

      Objective

      As part of a larger intervention trial using a low-fat vegan diet, this study ascertained whether changes in cardiometabolic outcomes were associated with ABO blood type.

      Design

      A secondary analysis among intervention-group participants in a 16-week randomized clinical trial.

      Participants/setting

      In a larger study of overweight individuals randomly assigned to follow a low-fat vegan diet or to make no diet changes for 16 weeks, ABO blood typing was conducted on 68 intervention-group participants.

      Intervention

      Intervention-group participants were asked to follow a low-fat vegan diet and attend weekly educational classes to aid in diet adherence.

      Main outcome measures

      Body weight, fat mass, visceral fat volume, blood lipid levels, fasting plasma glucose levels, and glycated hemoglobin concentrations.

      Statistical analyses performed

      Student t tests compared participants with blood type A to all other participants, and individuals with blood type O to all other participants.

      Results

      There were no significant differences in any outcome between individuals of blood type A and all other participants, or between individuals of blood type O and all other participants. Mean body weight change was –5.7 kg for blood type A participants and –7.0 kg for all other participants (P = 0.09), and was –7.1 kg for type O participants and –6.2 kg for all other participants (P = 0.33). Mean total cholesterol decreased 17.2 mg/dL in the type A group and 18.3 mg/dL for all other participants (P = 0.90), and decreased 17.4 mg/dL among type O participants and 18.4 mg/dL for all other participants (P = 0.89).

      Conclusions

      Blood type was not associated with the effects of a plant-based diet on body weight, body fat, plasma lipid concentrations, or glycemic control.

      Keywords

      Research Question: Are the effects of a plant-based dietary intervention on body weight, blood lipid concentrations, and glycemic control associated with blood type?
      Key Findings: Participant blood type group was not associated with changes in weight, lipid concentrations, or glycemic control.
      In the 1950s and 1960s, research studies identified a modest reduction in ischemic heart disease risk among individuals with blood type O, compared with those with other blood types.
      • Allan T.M.
      • Dawson A.A.
      ABO blood groups and ischaemic heart disease in men.
      This finding was confirmed in an analysis of participants in the Nurses’ Health Study and Health Professionals Follow-Up Study; individuals with blood types other than O were found to be 11% more likely to develop coronary heart disease, compared with those with blood type O (relative risk 1.11, 95% CI 1.05 to 1.18; P = 0.001).
      • He M.
      • Wolpin B.
      • Rexrode K.
      • et al.
      ABO blood group and risk of coronary heart disease in two prospective cohort studies.
      Similarly, a 2016 meta-analysis of case-control and cohort studies showed a 14% increased coronary artery disease risk for blood group A (odds ratio 1.14, 95% CI 1.03 to 1.26; P = 0.001) and a 15% reduced risk for blood group O (odds ratio 0.85, 95% CI 0.78 to 0.94; P = 0.0008), compared with other blood groups.
      • Chen Z.
      • Yang S.-H.
      • Xu H.
      • Li J.-J.
      ABO blood group system and the coronary artery disease: an updated systematic review and meta-analysis.
      Some evidence also suggests that individuals with blood type A have higher total cholesterol and low-density lipoprotein (LDL) cholesterol concentrations.
      • Etemadi A.
      • Kamangar F.
      • Islami F.
      • et al.
      Mortality and cancer in relation to ABO blood group phenotypes in the Golestan Cohort Study.
      The type O blood group is associated with increased risk of type 2 diabetes, compared with the type B group.
      • Qi L.
      • Cornelis M.C.
      • Kraft P.
      • et al.
      Genetic variants in ABO blood group region, plasma soluble E-selectin levels and risk of type 2 diabetes.
      These studies raise the question as to whether individuals with different blood groups will benefit from different types of diet changes. That possibility was suggested in 1996 by D’Adamo in a popular book recommending diets emphasizing fruits, vegetables, and grains for type A individuals, meat-oriented diets for type O individuals, and other diet variants for the other blood types.
      • D’Adamo P.
      Eat Right For Your Type.
      This hypothesis has not been supported by the limited observational data that are available,
      • Wang J.
      • García-Bailo B.
      • Nielsen D.E.
      • El-Sohemy A.
      ABO genotype, ‘blood-type’ diet and cardiometabolic risk factors.
      but it has not been adequately tested in intervention studies. A single trial testing the effects of diet advice and food samples on intake of fruits, vegetables, and whole grains included, as a secondary analysis, an examination of dietary patterns resembling diets recommended by D’Adamo for specific blood types, finding that the effects of these diet variations did not appear to differ with individual blood types.
      • Wang J.
      • Jamnik J.
      • Garcia-Bailo B.
      • Nielsen D.E.
      • et al.
      ABO genotype does not modify the association between the “blood-type” diet and biomarkers of cardiometabolic disease in overweight adults.
      While the sample size was reasonably large, diet changes were small, as were changes in weight, lipid concentrations, and other clinical measures, leaving open the possibility that greater diet changes would elicit greater effects.
      Therefore, in the course of a 16-week randomized clinical trial assessing the effects of a plant-based dietary intervention on metabolic health outcomes, blood types were ascertained for a secondary analysis in all participants, starting during the second of four study replications (blood type testing was not part of the initial protocol). All intervention-group participants, regardless of blood type, were assigned to a low-fat vegan diet that was similar to that recommended by D’Adamo for individuals with blood type A and recommended against for those with blood type O. The study hypothesis was that blood type would not be associated with the measured outcomes.

      Subjects and Methods

      Study Design

      This blood type study was part of a larger study on the metabolic effects of a low-fat vegan diet conducted between January 2017 and December 2018, using a single-center, randomized, open parallel design.
      • Kahleova H.
      • Petersen K.F.
      • Shulman G.I.
      • et al.
      Effect of a low-fat vegan diet on body weight, insulin sensitivity, postprandial metabolism, and intramyocellular and hepatocellular lipids in overweight adults: A randomized clinical trial.
      Adult men and women with a body mass index (BMI) between 28 and 40 were recruited in the Washington, DC, metropolitan area. Exclusion criteria included a history of diabetes, alcohol or drug abuse, pregnancy, lactation, or current use of a vegan diet. The study protocol was approved by the Chesapeake Institutional Review Board on October 12, 2016, and May 11, 2017. All participants signed a written informed consent. Registration on ClinicalTrials.gov was initiated on October 20, 2016 (identifier: NCT02939638).
      Of 3,115 people screened by telephone, 244 met the participation criteria and were randomly assigned in 4 replications to follow a low-fat vegan diet (vegan group, n = 122) or to make no diet changes (control group, n = 122) for 16 weeks. Starting in the second replication, blood type was determined (Eldoncards HKA 2511-1; Eldon Biologicals) in participants to permit a substudy of the effects of blood type on outcome measures, as a secondary analysis. Participants who had already completed the study before blood typing was added to the protocol were excluded from this subanalysis. In all, 68 intervention-group completers were blood-typed and were deemed to provide an adequate sample as called for by the power analysis described below (see the Figure).
      Figure thumbnail gr1
      FigureFlow diagram of the randomized clinical trial of the effect of blood type on the effectiveness of a plant-based dietary intervention on health outcomes. BMI = body mass index.
      The intervention group was asked to avoid animal products while focusing on consuming fruits, vegetables, grains, and legumes, and to limit added oils, nuts, and seeds. Daily dietary fat intake was limited to 20 to 30 g. No meals were provided. Vitamin B-12 was supplemented (500 μg/day). Participants in the control group were asked to maintain their current diets for the duration of the study, which included animal products. Study participants assigned to the vegan group were instructed to attend weekly classes that provided dietary guidance and encouraged diet adherence. Classes were conducted in person by study physicians, registered dietitians, and research personnel following an established curriculum. Handouts were provided in each class, and individuals who missed classes were contacted by telephone. No compensation was provided for class attendance, but participants who completed all assessments at Weeks 0 and 16 were paid $100.

      Dietary Intake and Physical Activity

      Dietary intake was assessed through a self-reported 3-day diet record that was completed at both baseline and 16 weeks. Participants met with Nutrition Data System for Research certified staff members who collected and analyzed dietary data using Nutrition Data System for Research version 2016.
      ,
      • Schakel S.F.
      Maintaining a nutrient database in a changing marketplace: Keeping pace with changing food products—a research perspective.
      For any food item missing from the program, a data entry was created to match the macronutrient content and the source of the macronutrients. Study participants were instructed to maintain their usual level of physical activity, which was assessed by the International Physical Activity Questionnaire.
      • Hagströmer M.
      • Oja P.
      • Sjöström M.
      The International Physical Activity Questionnaire (IPAQ): A study of concurrent and construct validity.
      In addition, participants continued their usual medications, except as modified by their personal physicians throughout the duration of the study.

      Outcomes

      The primary outcomes were body weight, BMI, fat mass, visceral adipose tissue volume, blood lipid levels, fasting plasma glucose levels, and glycated hemoglobin (HbA1c) concentrations. All variables were measured at baseline and 16 weeks, following a 10 to 12 hour overnight water-only fast, by trained research personnel following specific protocols. Height was measured using a stadiometer. Weight was measured using a calibrated digital scale accurate to 0.1 kg. Body composition was measured using a dual-energy x-ray absorptiometry scan (iDXA; GE Healthcare). Blood type was assessed by fingerstick (Eldoncards HKA 2511-1). Serum glucose level was analyzed using the Hexokinase UV endopoint method (Roche). HbA1c was measured by turbidimetric inhibition immunoassay (Roche). Plasma lipid levels were measured by enzymatic colorimetric methods (Roche). Baseline measurements and statistical analyses were made by staff members who were masked to group assignment.

      Statistical Analysis

      Sample size was estimated based on a recent trial in which 54 people began a similar diet intervention in either the first or second phase of a crossover trial, achieving a mean weight change of –5.8 ± 4.2 kg. With this magnitude of effect and variability, a two-arm trial would need 20 total participants for 80% power and 26 for 90% power. In the same trial, 29 individuals had the dietary intervention is their first treatment, before the crossover, with a mean weight change of –7.4 ± 3.8 kg. With this magnitude of effect and variability, a two-arm trial would need 12 total participants for 80% power and 14 for 90% power.
      To assess whether the dietary intervention had effects that differed according to blood group, participants within the intervention arm were compared in two ways: as blood type A versus non-A (that is, all other blood groups combined) and blood type O versus non-O. Studen t tests and χ2 tests were used to compare blood groups with respect to continuous and categorical baseline characteristics respectively, and two-sample t tests were also conducted to assess differences between these groups in changes in cardiometabolic outcomes in response to the intervention diet. In addition, within-group changes in clinical outcomes were assessed for significance using one-sample t tests. Statistical significance was defined as a P value < 0.05. Statistical analyses were carried out using SAS version 9.4.

      Results

      Baseline mean body weight, BMI, fat mass, visceral fat volume, and HbA1c concentration did not differ significantly between blood type groups (Table 1). Mean total and LDL cholesterol levels were significantly higher in blood type A (226.5 mg/dL [To convert mg/dL cholesterol to mmol/L, multiply mg/dL by 0.0259] and 138.9 mg/dL, respectively) versus non-A (184.2 mg/dL and 107.6 mg/dL, respectively), and significantly lower in blood type O (182.4 mg/dL and 105.0 mg/dL, respectively) versus non-O (208.6 mg/dL and 126.5 mg/dL, respectively).
      Table 1Baseline characteristics of overweight individuals assigned to a plant-based intervention diet in a randomized clinical trial, by blood group category
      CharacteristicBlood Type ABlood Type Non-A (Types B, AB, O)P value
      P values refer to t tests for continuous variables and χ2 for categorical variables.
      N2048
      n (%)
      Sex0.22
      Male2 (10)11 (23)
      Female18 (90)37 (77)
      mean ± standard deviation
      Age (y)53.3 ± 11.353.8 ± 11.70.86
      Body weight (kg)92.0 ± 14.093.2 ± 15.30.78
      BMI
      BMI = body mass index.
      33.3 ± 4.732.1 ± 3.30.32
      Fat mass (g)41.0 ± 10.139.6 ± 9.80.58
      VAT
      VAT = visceral adipose tissue.
      volume (cm3)
      1,621 ± 925.61,554 ± 1,1160.81
      Glycated hemoglobin (%)5.6 ± 0.35.6 ± 0.40.85
      Fasting plasma glucose (mg/dL)
      To convert mg/dL plasma glucose to mmol/L, multiply mg/dL by 0.0555.
      96.5 ± 13.497.5 ± 11.10.75
      Total cholesterol (mg/dL)
      To convert mg/dL cholesterol to mmol/L, multiply mg/dL by 0.0259.
      226.5 ± 37.9184.2 ± 42.2< .001
      LDL
      LDL = low-density lipoprotein.
      cholesterol (mg/dL)
      To convert mg/dL cholesterol to mmol/L, multiply mg/dL by 0.0259.
      138.9 ± 26.5107.6 ± 34.1< .001
      Blood type OBlood type non-O (types A, B, AB)
      N3137
      n (%)
      Sex0.20
      Male8 (26)5 (14)
      Female23 (74)32 (86)
      mean ± standard deviation
      Age (y)53.8 ± 12.753.5 ± 10.60.89
      Body weight (kg)92.0 ± 15.393.5 ± 14.60.68
      BMI
      BMI = body mass index.
      (kg/m2), mean±SD
      32.1 ± 3.532.8 ± 4.00.41
      Fat mass38.7 ± 10.841.1 ± 9.10.32
      VAT
      VAT = visceral adipose tissue.
      volume (cm3)
      1,728 ± 1,0891,443 ± 1,0260.27
      Glycated hemoglobin (%)5.6 ± 0.45.6 ± 0.30.87
      Fasting plasma glucose (mg/dL)
      To convert mg/dL plasma glucose to mmol/L, multiply mg/dL by 0.0555.
      100.1 ± 11.395.0 ± 11.80.08
      Total cholesterol (mg/dL)
      To convert mg/dL cholesterol to mmol/L, multiply mg/dL by 0.0259.
      182.4 ± 43.4208.6 ± 43.60.02
      LDL
      LDL = low-density lipoprotein.
      cholesterol (mg/dL)
      To convert mg/dL cholesterol to mmol/L, multiply mg/dL by 0.0259.
      105.0 ± 34.4126.5 ± 32.70.01
      a P values refer to t tests for continuous variables and χ2 for categorical variables.
      b BMI = body mass index.
      c VAT = visceral adipose tissue.
      d To convert mg/dL plasma glucose to mmol/L, multiply mg/dL by 0.0555.
      e To convert mg/dL cholesterol to mmol/L, multiply mg/dL by 0.0259.
      f LDL = low-density lipoprotein.
      The relationships between blood types and changes in cardiometabolic outcomes in response to the intervention diet are shown in Table 2. Mean body weight change was –5.7 kg for blood type A participants and –7.0 kg for non-A participants (P = 0.09). Similarly, mean body weight change was –7.1 kg for type O participants and –6.2 kg for non-O participants (P = 0.33). Changes in fat mass and VAT paralleled those for body weight and did not differ significantly between blood groups.
      Table 2Clinical changes
      P values for comparisons of between-group (blood group A vs non-A, and blood group O vs non-O) changes (baseline to 16 weeks).
      among overweight individuals assigned to a plant-based intervention diet in a randomized clinical trial, by blood group category
      VariableBlood Group A (n = 20)Blood Group non-A (n = 48)P value between groups
      P values for comparisons of between-group (blood group A vs non-A, and blood group O vs non-O) changes (baseline to 16 weeks).
      BaselineWeek 16ChangeBaselineWeek 16Change
      mean (95% CI)
      Body weight (kg)92.0 (85.5 to 98.6)86.3 (79.5 to 93.1)–5.7 (–6.7 to –4.8)
      P < 0.001 for within-group changes.
      93.2 (88.7 to 97.6)86.2 (82.2 to 90.2)–7.0 (–8.2 to –5.8)
      P < 0.001 for within-group changes.
      0.09
      BMI
      BMI = body mass index (calculated as weight in kilograms/height in meters2).
      33.3 (31.1 to 35.5)31.3 (28.9 to 33.6)–2.1 (–2.4 to –1.7)
      P < 0.001 for within-group changes.
      32.1 (31.2 to 33.1)29.8 (28.9 to 30.7)–2.3 (-2.8 to -1.9)
      P < 0.001 for within-group changes.
      0.36
      Fat mass (g)41.0 (36.3 to 45.8)37.1 (32.1 to 42.2)–3.9 (–4.7 to –3.1)
      P < 0.001 for within-group changes.
      39.6 (36.7 to 42.4)34.8 (32.0 to 37.5)–4.8 (-5.8 to -3.7)
      P < 0.001 for within-group changes.
      0.19
      VAT
      VAT = visceral adipose tissue.
      volume (cm3)
      1621 (1,187 to 2,054)1,385 (957.4 to 1,813)–235.3 (–339.0 to –131.6)
      P < 0.001 for within-group changes.
      1,554 (1,229 to 1,878)1,297 (1,044 to 1,551)–256.2 (–400.8 to –111.7)
      P < 0.001 for within-group changes.
      0.81
      (%)5.6 (5.5 to 5.8)5.5 (5.3 to 5.7)–0.1 (–0.2 to –0.02)
      P < 0.05.
      5.6 (5.5 to 5.7)5.5 (5.4 to 5.6)–0.07 (–0.13 to –0.01)
      P < 0.05.
      0.34
      Fasting plasma glucose (mg/dL)
      To convert mg/dL plasma glucose to mmol/L, multiply mg/dL by 0.0555.
      96.5 (90.2 to 102.8)92.2 (87.8 to 96.6)–4.3 (-10.3 to +1.7)97.5 (94.2 to 100.9)95.9 (92.3 to 99.5)–1.6 (–4.7 to 1.4)0.38
      Total cholesterol (mg/dL)
      To convert mg/dL cholesterol to mmol/L, multiply mg/dL by 0.0259.
      226.5 (208.8 to 244.2)209.3 (195.3 to 223.3)–17.2 (–33.8 to –0.6)
      P < 0.05.
      184.2 (171.8 to 196.6)166.0 (156.2 to 175.7)–18.3 (–25.2 to –11.3)
      P < 0.001 for within-group changes.
      0.90
      LDLf cholesterol (mg/dL)
      To convert mg/dL cholesterol to mmol/L, multiply mg/dL by 0.0259.
      138.9 (126.4 to 151.3)127.9 (115.3 to 140.5)–11.0 (–26.8 to 4.9)107.6 (97.6 to 117.6)93.9 (85.8 to 102.1)–13.6 (–19.8 to –7.5)
      P < 0.001 for within-group changes.
      0.75
      Blood Group O (n = 31)BloodGroup non-O (n = 37)
      Body weight (kg)92.0 (86.4 to 97.6)84.9 (79.8 to 90.0)–7.1 (–8.6 to –5.6)
      P < 0.001 for within-group changes.
      93.5 (88.7 to 98.4)87.3 (82.7 to 92.0)–6.2 (–7.3 to –5.1)
      P < 0.001 for within-group changes.
      0.33
      BMI32.1 (30.8 to 33.4)29.6 (28.4 to 30.8)–2.5 (–3.0 to –2.0)
      P < 0.001 for within-group changes.
      32.8 (31.5 to 34.2)30.8 (29.4 to 32.2)–2.0 (–2.5 to –1.6)
      P < 0.001 for within-group changes.
      0.16
      Fat mass (g)38.7 (34.7 to 42.6)33.7 (29.9 to 37.5)–5.0 (–6.4 to ––3.5)
      P < 0.001 for within-group changes.
      41.1 (38.1 to 44.1)37.0 (33.9 to 40.0)–4.1 (–4.9 to –3.3)
      P < 0.001 for within-group changes.
      0.29
      VAT volume (cm3)1728 (1,329 to 2,128)1376 (1,048 to 1,704)–352.3 (–538.3 to –166.2)
      P < 0.001 for within-group changes.
      1443 (1,101 to 1,786)1,279 (986.0 to 1,572)–164.5 (–279.3 to –49.6)
      P ≤ 0.01.
      0.09
      Glycated hemoglobin (%)5.6 (5.4 to 5.7)5.5 (5.4 to 5.6)–0.09 (–0.18 to –0.002)
      P < 0.05.
      5.6 (5.5 to 5.7)5.5 (5.4 to 5.6)–0.08 (–0.14 to –0.02)
      P ≤ 0.01.


      0.01
      0.82
      Fasting plasma glucose (mg/dL)
      To convert mg/dL plasma glucose to mmol/L, multiply mg/dL by 0.0555.
      100.1 (95.7 to 104.5)97.7 (93.0 to 102.5)–2.4 (-6.6 to 1.9)95.0 (91.1 to 98.9)92.5 (89.1 to 95.9)–2.5 (–6.3 to 1.3)0.97
      Total cholesterol (mg/dL)
      To convert mg/dL cholesterol to mmol/L, multiply mg/dL by 0.0259.
      182.4 (166.2 to 198.6)165.0 (152.7 to 177.3)–17.4 (–26.7 to –8.1)
      P < 0.001 for within-group changes.
      208.6 (194.0 to 223.1)190.2 (177.6 to 202.8)–18.4 (–28.2 to –8.5)
      P < 0.001 for within-group changes.
      0.89
      LDL cholesterol (mg/dL)
      To convert mg/dL cholesterol to mmol/L, multiply mg/dL by 0.0259.
      105.0 (92.2 to 117.9)92.4 (82.2 to 102.6)–12.6 (–21.2 to –4.1)
      P ≤ 0.01.
      126.5 (115.6 to 137.4)113.5 (103.0 to 124.1)–13.0 (–22.1 to –3.9)
      P ≤ 0.01.
      0.96
      a P values for comparisons of between-group (blood group A vs non-A, and blood group O vs non-O) changes (baseline to 16 weeks).
      b BMI = body mass index (calculated as weight in kilograms/height in meters2).
      c VAT = visceral adipose tissue.
      d To convert mg/dL plasma glucose to mmol/L, multiply mg/dL by 0.0555.
      e To convert mg/dL cholesterol to mmol/L, multiply mg/dL by 0.0259.
      P < 0.05.
      ∗∗ P ≤ 0.01.
      ∗∗∗ P < 0.001 for within-group changes.
      Despite differences in baseline lipid level values, changes in lipid levels were similar across groups. Mean total cholesterol decreased 17.2 mg/dL in type A participants and 18.3 mg/dL for non-A participants (P = 0.90). Mean total cholesterol decreased 17.4 mg/dL among type O participants and 18.4 mg/dL for non-O participants. None of these comparisons yielded statistically significant differences. Data for additional outcomes are found in Table 2.

      Discussion

      After 16 weeks on a low-fat vegan diet, no significant differences were found in any of the cardiometabolic changes between blood type groups. Although the intervention diet was similar to that recommended by D’Adamo
      • D’Adamo P.
      Eat Right For Your Type.
      for individuals with blood type A and specifically recommended against for those with type O, there were no associations between these blood types and the outcomes of the dietary intervention.
      The ABO blood types reflect the presence or absence of certain antigens on the surfaces of red blood cells, among other tissues. These antigens may influence cardiovascular risk. Not only do observational studies show differences in cardiovascular risk associated with blood groups, but plasma lipid concentrations may also differ based on blood type. Individuals with blood type A have been shown to have higher total cholesterol and LDL cholesterol concentrations, compared with other blood groups,
      • Etemadi A.
      • Kamangar F.
      • Islami F.
      • et al.
      Mortality and cancer in relation to ABO blood group phenotypes in the Golestan Cohort Study.
      a finding confirmed in the present sample. Compared with individuals with blood type O, non-O individuals also tend to have higher plasma concentrations of factor VIII-von Willebrand factor complex, which plays a role in hemostasis.
      • Wong F.L.
      • Kodama K.
      • Sasaki H.
      • Yamada M.
      • Hamilton H.B.
      Longitudinal study of the association between ABO phenotype and total serum cholesterol level in a Japanese cohort.
      ,
      • Gill J.C.
      • Endres-Brooks J.
      • Bauer P.J.
      • Marks W.J.
      • Montgomery R.R.
      The effect of ABO blood group on the diagnosis of von Willebrand disease.
      Recent evidence also suggests that novel severe acute respiratory syndrome coronavirus 2 infection that causes coronavirus disease 2019 occurs more frequently and with greater severity in individuals with blood type A, and less frequently and with less severity in individuals with blood group O, compared with other blood groups.
      • Ellinghaus D.
      • Degenhardt F.
      • Bujanda L.
      • et al.
      Genomewide association study of severe Covid-19 with respiratory failure.
      ,
      • Wu B.B.
      • Gu D.Z.
      • Yu J.N.
      • et al.
      Association between ABO blood groups and COVID-19 infection, severity and demise: A systematic review and meta-analysis.
      A 2013 systematic review identified no published observational or intervention studies that had investigated the health effects of diets based on blood type.
      • Cusack L.
      • De Buck E.
      • Compernolle V.
      • Vandekerckhove P.
      Blood type diets lack supporting evidence: A systematic review.
      A 2014 report from the Toronto Nutrigenomics and Health study (N = 1,455) described a cross-sectional examination of young adults ranging in age from 20 to 29 years. Using food frequency questionnaires, diet scores were calculated to determine relative adherence to each of the four blood type diets recommended by D’Adamo.
      • D’Adamo P.
      Eat Right For Your Type.
      Regardless of blood type, participants in the highest tertile of the type-A diet score (who consumed more fruit, vegetables, and grains, and less meat) had better cardiometabolic outcomes (ie, lower BMI, waist circumference, serum cholesterol and triglyceride levels, and insulin resistance) compared with individuals not adhering as closely to the type-A diet.
      • Wang J.
      • García-Bailo B.
      • Nielsen D.E.
      • El-Sohemy A.
      ABO genotype, ‘blood-type’ diet and cardiometabolic risk factors.
      Greater adherence to the type AB diet (emphasizing fruits, vegetables, and dairy products) was associated with lower blood pressure, total cholesterol and triglyceride levels, insulin, and insulin resistance, regardless of the participants’ blood type. Greater adherence to the type O diet (emphasizing meat, fruits, and vegetables, and de-emphasizing grains) was associated with lower triglyceride levels, regardless of the participants’ blood type. The authors concluded that, although adherence to certain dietary practices was associated with changes in certain physical variables, these associations were not related to an individual’s blood type.
      The Toronto Healthy Diet Study, a 6-month trial using a dietary intervention that emphasized greater intake of plant-derived foods, included a secondary analysis assessing whether relative adherence to blood type diets influenced the cardiometabolic effects of the intervention.
      • Jenkins D.J.A.
      • Boucher B.A.
      • Ashbury F.D.
      • et al.
      Effect of current dietary recommendations on weight loss and cardiovascular risk factors.
      A food frequency questionaire was completed at baseline and 6 months after the dietary intervention by 576 participants. Although associations between diet adherence and cardiometabolic risk factors were observed, the blood types of the participants exerted no apparent effect.
      • Wang J.
      • Jamnik J.
      • Garcia-Bailo B.
      • Nielsen D.E.
      • et al.
      ABO genotype does not modify the association between the “blood-type” diet and biomarkers of cardiometabolic disease in overweight adults.
      The present study used a more vigorous intervention, excluding all animal products, along with weekly classes to maximize adherence, achieving clinically important and statistically significant effects within each blood group. Even so, there was no significant difference in clinical effect across blood groups. This does not discount the possibility that the effects of diet on health may be dependent on other genetic components.
      • Frazier-Wood Alexis C.
      Dietary patterns, genes, and health: Challenges and obstacles to be overcome.
      These studies, like the present one, suggest that dietary changes, especially increased intake of fruits, vegetables, and grains, are beneficial not only to individuals with blood type A, but also to all individuals regardless of blood type, and that there is no apparent value of limiting these healthful diet changes to a specific set of individuals based on ABO blood group.
      This study has several strengths. All participants in each cohort started simultaneously, allowing the investigators to rule out possible effects of seasonal fluctuations in the diet. The study duration was reasonably long, providing sufficient time for adaptation to the diet. The low attrition rate suggests that the intervention was acceptable and sustainable, providing a reasonable test of the hypothesis. Given that participants were living at home and preparing their own meals or eating at restaurants, these results are applicable to free-living conditions.
      This study also has limitations. Individuals with blood type AB and B were represented in small numbers. Statistical power was improved by combining blood groups (A vs non-A, and O vs non-O). Dietary intake was calculated based on self-reported diet records, which have well-known limitations.
      • Yuan C.
      • Spiegelman D.
      • Rimm E.B.
      • et al.
      Relative validity of nutrient intakes assessed by questionnaire, 24-hour recalls, and diet records as compared with urinary recovery and plasma concentration biomarkers: findings for women.
      However, it is reassuring that the reported changes in nutrient intake were paralleled by weight loss and metabolic changes. The participants were generally health-conscious individuals who were willing to make substantial changes to their diet. In this regard, they may not be representative of the general population, but may be representative of a clinical population seeking help for weight problems.

      Conclusions

      These results indicate that blood type is not associated with the effect of a plant-based diet on body weight, body fat, plasma lipid concentrations, or glycemic control.

      References

        • Allan T.M.
        • Dawson A.A.
        ABO blood groups and ischaemic heart disease in men.
        Br Heart J. 1968; 30: 377-382
        • He M.
        • Wolpin B.
        • Rexrode K.
        • et al.
        ABO blood group and risk of coronary heart disease in two prospective cohort studies.
        Arterioscler Thromb Vasc Biol. 2012; 32: 2314-2320
        • Chen Z.
        • Yang S.-H.
        • Xu H.
        • Li J.-J.
        ABO blood group system and the coronary artery disease: an updated systematic review and meta-analysis.
        Sci Rep. 2016; 6
        • Etemadi A.
        • Kamangar F.
        • Islami F.
        • et al.
        Mortality and cancer in relation to ABO blood group phenotypes in the Golestan Cohort Study.
        BMC Med. 2015; 13: 8
        • Qi L.
        • Cornelis M.C.
        • Kraft P.
        • et al.
        Genetic variants in ABO blood group region, plasma soluble E-selectin levels and risk of type 2 diabetes.
        Hum Mol Genet. 2010; 19: 1856-1862
        • D’Adamo P.
        Eat Right For Your Type.
        G.P. Putnam’s Sons, New York, NY1996
        • Wang J.
        • García-Bailo B.
        • Nielsen D.E.
        • El-Sohemy A.
        ABO genotype, ‘blood-type’ diet and cardiometabolic risk factors.
        PLoS One. 2014; 9e84749
        • Wang J.
        • Jamnik J.
        • Garcia-Bailo B.
        • Nielsen D.E.
        • et al.
        ABO genotype does not modify the association between the “blood-type” diet and biomarkers of cardiometabolic disease in overweight adults.
        J Nutr. 2018; 148: 518-525
        • Kahleova H.
        • Petersen K.F.
        • Shulman G.I.
        • et al.
        Effect of a low-fat vegan diet on body weight, insulin sensitivity, postprandial metabolism, and intramyocellular and hepatocellular lipids in overweight adults: A randomized clinical trial.
        JAMA Network Open. 2020; (in press)
      1. Nutrition Data System for Research [database]. Version 2016. Nutrition Coordinating Center, University of Minnesota, Minneapolis, MN2016
        • Schakel S.F.
        Maintaining a nutrient database in a changing marketplace: Keeping pace with changing food products—a research perspective.
        J Food Comp Anal. 2001; 14: 315-322
        • Hagströmer M.
        • Oja P.
        • Sjöström M.
        The International Physical Activity Questionnaire (IPAQ): A study of concurrent and construct validity.
        Public Health Nutr. 2006; 9: 755-762
      2. SAS [computer program]. Version 9.4. SAS, Cary, NC2013
        • Wong F.L.
        • Kodama K.
        • Sasaki H.
        • Yamada M.
        • Hamilton H.B.
        Longitudinal study of the association between ABO phenotype and total serum cholesterol level in a Japanese cohort.
        Gen Epidemiol. 1992; 9: 405-418
        • Gill J.C.
        • Endres-Brooks J.
        • Bauer P.J.
        • Marks W.J.
        • Montgomery R.R.
        The effect of ABO blood group on the diagnosis of von Willebrand disease.
        Blood. 1987; 69: 1691-1695
        • Ellinghaus D.
        • Degenhardt F.
        • Bujanda L.
        • et al.
        Genomewide association study of severe Covid-19 with respiratory failure.
        N Engl J Med. 2020; 383: 1522-1534
        • Wu B.B.
        • Gu D.Z.
        • Yu J.N.
        • et al.
        Association between ABO blood groups and COVID-19 infection, severity and demise: A systematic review and meta-analysis.
        Infect Genet Evol. 2020; 84: 104485
        • Cusack L.
        • De Buck E.
        • Compernolle V.
        • Vandekerckhove P.
        Blood type diets lack supporting evidence: A systematic review.
        Am J Clin Nutr. 2013; 98: 99-104
        • Wang J.
        • García-Bailo B.
        • Nielsen D.E.
        • El-Sohemy A.
        ABO genotype, ‘blood-type’ diet and cardiometabolic risk factors.
        PLoS One. 2014; 9e84749
        • Jenkins D.J.A.
        • Boucher B.A.
        • Ashbury F.D.
        • et al.
        Effect of current dietary recommendations on weight loss and cardiovascular risk factors.
        J Am Coll Cardiol. 2017; 69: 1103-1112
        • Frazier-Wood Alexis C.
        Dietary patterns, genes, and health: Challenges and obstacles to be overcome.
        Curr Nutr Rep. 2015; 4: 82-87
        • Yuan C.
        • Spiegelman D.
        • Rimm E.B.
        • et al.
        Relative validity of nutrient intakes assessed by questionnaire, 24-hour recalls, and diet records as compared with urinary recovery and plasma concentration biomarkers: findings for women.
        Am J Epidemiol. 2018; 187: 1051-1063

      Biography

      N.D. Barnard is the President of the Physicians Committee for Responsible Medicine, and Adjunct faculty, George Washington University School of Medicine and Health Sciences, Washington, DC.
      E. Rembert is a Clinical Research Assistant, Physicians Committee for Responsible Medicine, Washington, DC.
      A. Freeman and M. Bradshaw are Clinical Research Interns, Physicians Committee for Responsible Medicine, Washington, DC.
      M. Bradshaw is a Clinical Research Intern, Physicians Committee for Responsible Medicine, Washington, DC, and Statistical Clinical Data Management Programmer, Seattle Children’s Hospital, Seattle, WA.
      R. Holubkov is a Professor, University of Utah School of Medicine Department of Pediatrics, Salt Lake City, UT.
      H. Kahleova is the Director of Clinical Research, Physicians Committee for Responsible Medicine, Washington, DC.