A randomized crossover trial on the effect of plant-based compared with animal-based meat on trimethylamine-N-oxide and cardiovascular disease risk factors in generally healthy adults: Study With Appetizing Plantfood-Meat Eating Alternative Trial (SWAP-MEAT)

Abstract

Background:
Despite the rising popularity of plant-based alternative meats, there is limited evidence of the health effects of these products.

Objectives:
We aimed to compare the effect of consuming plant-based alternative meat (Plant) as opposed to animal meat (Animal) on health factors. The primary outcome was fasting serum trimethylamine-N-oxide (TMAO). Secondary outcomes included fasting insulin-like growth factor 1, lipids, glucose, insulin, blood pressure, and weight.

Methods:
SWAP-MEAT (The Study With Appetizing Plantfood-Meat Eating Alternatives Trial) was a single-site, randomized crossover trial with no washout period. Participants received Plant and Animal products, dietary counseling, lab assessments, microbiome assessments (16S), and anthropometric measurements. Participants were instructed to consume ≥2 servings/d of Plant compared with Animal for 8 wk each, while keeping all other foods and beverages as similar as possible between the 2 phases.

Results:
The 36 participants who provided complete data for both crossover phases included 67% women, were 69% Caucasian, had a mean ± SD age 50 ± 14 y, and BMI 28 ± 5 kg/m2. Mean ± SD servings per day were not different by intervention sequence: 2.5 ± 0.6 compared with 2.6 ± 0.7 for Plant and Animal, respectively (P = 0.76). Mean ± SEM TMAO concentrations were significantly lower overall for Plant (2.7 ± 0.3) than for Animal (4.7 ± 0.9) (P = 0.012), but a significant order effect was observed (P = 0.023). TMAO concentrations were significantly lower for Plant among the n = 18 who received Plant second (2.9 ± 0.4 compared with 6.4 ± 1.5, Plant compared with Animal, P = 0.007), but not for the n = 18 who received Plant first (2.5 ± 0.4 compared with 3.0 ± 0.6, Plant compared with Animal, P = 0.23). Exploratory analyses of the microbiome failed to reveal possible responder compared with nonresponder factors. Mean ± SEM LDL-cholesterol concentrations (109.9 ± 4.5 compared with 120.7 ± 4.5 mg/dL, P = 0.002) and weight (78.7 ± 3.0 compared with 79.6 ± 3.0 kg, P < 0.001) were lower during the Plant phase.

Conclusions:
Among generally healthy adults, contrasting Plant with Animal intake, while keeping all other dietary components similar, the Plant products improved several cardiovascular disease risk factors, including TMAO; there were no adverse effects on risk factors from the Plant products.This trial was registered at clinicaltrials.gov as NCT03718988.

Full text

See corresponding editorial on page 1151.
A randomized crossover trial on the effect of plant-based compared
with animal-based meat on trimethylamine-N-oxide and cardiovascular
disease risk factors in generally healthy adults: Study With Appetizing
Plantfood—Meat Eating Alternative Trial (SWAP-MEAT)
Anthony Crimarco,1Sparkle Springeld,1Christina Petlura,1Taylor Streaty,1Kristen Cunanan,2Justin Lee,2
Priya Fielding-Singh,1Matthew M Carter,3Madeline A Topf,3Hannah C Wastyk,3,4Erica D Sonnenburg,3,5
Justin L Sonnenburg,3,5,6and Christopher D Gardner1
1Stanford Prevention Research Center, Stanford University School of Medicine, Stanford, CA, USA; 2Quantitative Sciences Unit, Stanford University School of
Medicine, Stanford, CA, USA; 3Microbiology and Immunology, Stanford University School of Medicine, Stanford, CA, USA; 4Department of Bioengineering,
Stanford University School of Medicine, Stanford, CA, USA; 5Center for Human Microbiome Studies, Stanford University, Stanford, CA, USA; and 6Chan
Zuckerberg Biohub, San Francisco, CA, USA
ABSTRACT
Background: Despite the rising popularity of plant-based alternative
meats, there is limited evidence of the health effects of these
products.
Objectives: We aimed to compare the effect of consuming plant-
based alternative meat (Plant) as opposed to animal meat (Ani-
mal) on health factors. The primary outcome was fasting serum
trimethylamine-N-oxide (TMAO). Secondary outcomes included
fasting insulin-like growth factor 1, lipids, glucose, insulin, blood
pressure, and weight.
Methods: SWAP-MEAT (The Study With Appetizing Plantfood—
Meat Eating Alternatives Trial) was a single-site, randomized
crossover trial with no washout period. Participants received
Plant and Animal products, dietary counseling, lab assessments,
microbiome assessments (16S), and anthropometric measurements.
Participants were instructed to consume ≥2 servings/d of Plant
compared with Animal for 8 wk each, while keeping all other foods
and beverages as similar as possible between the 2 phases.
Results: The 36 participants who provided complete data for both
crossover phases included 67% women, were 69% Caucasian, had a
mean ±SD age 50 ±14 y, and BMI 28 ±5 kg/m2.Mean±SD
servings per day were not different by intervention sequence:
2.5 ±0.6 compared with 2.6 ±0.7 for Plant and Animal, respectively
(P=0.76). Mean ±SEM TMAO concentrations were signicantly
lower overall for Plant (2.7 ±0.3) than for Animal (4.7 ±0.9)
(P=0.012), but a signicant order effect was observed (P=0.023).
TMAO concentrations were signicantly lower for Plant among
the n=18 who received Plant second (2.9 ±0.4 compared with
6.4 ±1.5, Plant compared with Animal, P=0.007), but not for the
n=18 who received Plant rst (2.5 ±0.4 compared with 3.0 ±0.6,
Plant compared with Animal, P=0.23). Exploratory analyses of
the microbiome failed to reveal possible responder compared with
nonresponder factors. Mean ±SEM LDL-cholesterol concentrations
(109.9 ±4.5 compared with 120.7 ±4.5 mg/dL, P=0.002) and
weight (78.7 ±3.0 compared with 79.6 ±3.0 kg, P<0.001) were
lower during the Plant phase.
Conclusions: Among generally healthy adults, contrasting Plant
with Animal intake, while keeping all other dietary components
similar, the Plant products improved several cardiovascular disease
risk factors, including TMAO; there were no adverse effects on risk
factors from the Plant products. This trial was registered at clinicaltri-
als.gov as NCT03718988. Am J Clin Nutr 2020;112:1188–1199.
Keywords: diet, meat, plant-based alternative meat, randomized
controlled trial, cardiovascular disease risk factors, trimethylamine-
N-oxide
Introduction
Shifts to more plant-based diets, with fewer animal-based
foods, have been widely recommended for health and environ-
mental benets (1–3). Yet changing dietary behaviors remains
challenging owing to strongly held taste preferences, culinary
traditions, and social and cultural norms (4). Although plant-
based alternative meats (plant-meats)—i.e., vegetarian products
designed to resemble the taste and appearance of traditional
burgers, sausages, or other meats—have been available for years,
their consumer popularity has increased rapidly in recent years
(5). In North America, plant-meat sales grew by 37% from 2017
to 2019 (6). This rapid rise in popularity is due in part to producers
better simulating the taste of animal-meat products, as well
as increased marketing directed toward meat-eating consumers,
rather than just vegetarians (5,6).
1188 Am J Clin Nutr 2020;112:1188–1199. Printed in USA. Copyright ©The Author(s) on behalf of the American Society for Nutrition 2020.
Downloaded from https://academic.oup.com/ajcn/article/112/5/1188/5890315 by guest on 22 November 2020

Effect of plant vs. animal meat on TMAO and CVDRFs 1189
The rapid increase in plant-meat consumption has raised
scrutiny and criticism (7), namely the question of the overall
health effects of plant-meat products compared with animal meat
(8). Most plant-meat products meet the NOVA criteria for “ultra-
processed foods” (9). A recent study found that ultra-processed
food intake led to increased energy intake and weight gain relative
to whole foods (10). Compared with fresh, minimally processed
foods, many plant-meats are relatively high in saturated fat and
sodium, which are metabolically linked to hypercholesterolemia
and hypertension (11,12). Notably, plant-meats also contain
ber, which is absent in animal meats, has been proven to lower
LDL cholesterol (13), and is associated with reduced risk of
cardiovascular disease (CVD) and obesity (14). There is a paucity
of data on the net impact of differing amounts of saturated fat,
sodium, and ber on health factors for direct comparisons of plant
with animal meat.
Two emerging risk factors for CVD and certain cancers include
trimethylamine-N-oxide (TMAO) and insulin-like growth factor
1 (IGF-1). Animal-based foods, particularly red meat (e.g., beef
and pork), have a relatively high content of carnitine and choline,
which are precursors to TMAO (15). Recent trials have reported
that red meat intake raises TMAO blood concentrations (15–
17). In addition, vegans and vegetarians have been reported to
have lower TMAO and IGF-1 than meat eaters (17–22). Some
studies have suggested animal-meat consumption is associated
with greater IGF-1 concentrations and may increase the risk of
prostate and breast cancers (21,22).
The objective of this randomized crossover trial [SWAP-
MEAT (The Study With Appetizing Plantfood—Meat Eating
Alternatives Trial)] was to compare the effects of consuming
plant-based alternative meat (hereafter Plant) with those of con-
suming animal meat (hereafter Animal), primarily beef and pork,
on emerging health risk factors, cardiometabolic risk factors,
and the gut microbiome among generally healthy adults. The
primary study outcome was differences in serum TMAO after
Supported by a research gift from Beyond Meat Inc. (to CDG), National
Heart, Lung, and Blood Institute at the NIH grant T32HL007034 (to CDG),
and Stanford Clinical and Translational Science Award to Spectrum NIH UL1
TR001085 (to CDG).
Funding for this study was provided by Beyond Meat. In an effort to
reduce any inuences on the outcomes of this study, a statistical analysis plan
was submitted to ct.gov. The main analysis was conducted by a third-party
individual who had no involvement with the study design or collection of
data, and was blinded to all study participants.
Supplemental Tables 1–7, Supplemental Figures 1–4, Supplemental
Methods, and Supplemental Results are available from the “Supplementary
data” link in the online posting of the article and from the same link in the
online table of contents at https://academic.oup.com/ajcn/.
Data described in the article, code book, and analytic code will be made
available upon request pending application and approval by the corresponding
author.
Address correspondence to CDG (e-mail: cgardner@stanford.edu).
Abbreviations used: Animal, animal meat; CVD, cardiovascular disease;
CVDRFs, Cardiovascular Disease Risk Factors; IGF-1, insulin-like growth
factor 1; NDS-R, Nutrition Data System for Research; Plant, plant-based al-
ternative meat; SWAP-MEAT, The Study With Appetizing Plantfood—Meat
Eating Alternatives Trial; TMA, trimethylamine; TMAO, trimethylamine-N-
oxide.
Received April 22, 2020. Accepted for publication June 29, 2020.
First published online August 11, 2020; doi: https://doi.org/10.1093/ajcn/
nqaa203.
8 wk of Plant compared with Animal. Secondary outcomes were
differences in fasting plasma IGF-1 concentrations, metabolic
markers (blood lipids, glucose, and insulin), blood pressure,
weight, and microbiota composition (Shannon diversity).
Methods
Procedures for this study were followed in accordance with
the ethical standards from the Helsinki Declaration and were
approved by the Stanford University Human Subjects Committee
(Institutional Review Board). All study participants provided
written informed consent.
Study design
SWAP-MEAT was a single-site, randomized crossover trial
(NCT03718988) among adults assigned to 1 of 2 sequences:
8 wk of Plant followed by 8 wk of Animal, or vice versa. With
the exception of the Plant compared with Animal exchange,
participants were instructed to keep all other dietary habits
similar across the phases. At the end of the rst 8-wk phase,
participants were encouraged to move directly into the second
phase, without a washout period. However, a brief break between
phases was allowed if participant scheduling conicts would
have otherwise prevented second-phase participation. Participant
enrollment began on 5 December, 2018 and continued through
9 July, 2019. The date of nal follow-up data collection was 5
December, 2019.
Participants
The target population was generally healthy omnivorous adults
≥18 y of age who reported typically consuming ≥1 serving of
meat per day and were willing to consume ∼2 servings/d of both
Plant and Animal. They were recruited from available e-mail
lists among the Stanford community. Participants were required
to have a stable dietary history, dened as neither introducing
nor eliminating a major food group in their diet for at least the
previous month.
Exclusion criteria were weighing <110 lbs (50 kg); having
a BMI (in kg/m2)>40; LDL cholesterol >190 mg/dL; systolic
blood pressure >160 mm Hg or diastolic blood pressure >90 mm
Hg; clinically signicant or unstable pulmonary, cardiovascular,
gastrointestinal, hepatic, or renal functional abnormality; or
pregnancy. Individuals were excluded or had their study start
date delayed if within the past 2 mo they had taken systemic an-
tibiotics, antifungals, antivirals or antiparasitics, corticosteroids,
cytokines, methotrexate, or immunosuppressive cytotoxic agents
known to affect the microbiome.
Randomization
Randomization to 1 of the 2 diet sequences (Plant→Animal
or Animal→Plant) was performed in block sizes of 4 by an
independent statistician. Pairs (spousal or parent–child) were
randomly assigned in block sizes of 2. Participants did not learn of
their diet sequence until they had completed all baseline measures
and surveys. Laboratory technicians and study staff conducting
the blood and stool analyses were blinded to the diet sequence.
Downloaded from https://academic.oup.com/ajcn/article/112/5/1188/5890315 by guest on 22 November 2020

1190 Crimarco et al.
TABLE 1 Nutrient proles for Plant and Animal products1
Product Serving size Kcals
Carbohydrates,
g
Protein,
g
Tot al
fat, g
Saturated
fat, g
Fiber,
g
Sodium,
mg
Plant products
Burger 4 oz (113 g) 250 3 20 18 6 2 390
Beef crumbles 1/2 cup (55 g) 90 3 12 3 <1 1 240
Breakfast sausage 1 patty (65 g) 170 2 13 13 5 1 330
Hot Italian sausage 1 cooked link (76 g) 190 5 16 12 5 3 500
Brat sausage 1 cooked link (76 g) 190 5 16 12 5 3 500
Grilled chicken strips 3 oz (85 g) 130 6 22 2 0 3 360
Lightly seasoned chicken strips 3 oz (85 g) 130 5 20 3.5 0 3 340
Animal products
Burger 3.6 oz (100 g) 293 0 16 25 9 0 672
Ground beef 3.6 oz (100 g) 293 0 16 25 9 0 672
Good Morning pork breakfast sausage 1 link (47 g) 110 0 7 9 3 0 320
Hot Italian sausage 1 link (71 g) 170 1 10 14 5 0 480
Pork bratwurst 1 link (57 g) 230 4 8 21 9 0 400
Chicken breast 4 oz (113 g) 140 0 26 3 0.5 0 1402
1Some substitutions (≤10% of products) were necessary in the Animal phase because of limited availability; however, neither the amount, quality, nor the meat
were altered. Animal, animal meat; Plant, plant-based alternative meat.
2For raw meat products (beef and chicken), sodium and other seasonings were added by participants.
Intervention
The study consisted of 2 phases, each lasting 8 wk: Plant and
Animal. Participants were instructed to consume ≥2 servings of
the phase-congruent type of meat product per day and instructed
to match all non-study-provided foods as closely as possible
between the 2 phases. Specic instructions included tracking
items such as the types of burger buns and the garnishes and
condiments used with burger items, and being consistent with
these choices for both the Plant and Animal patties in the 2
phases. Eight weeks on each dietary phase was selected on the
basis of previous dietary interventions that reported signicant
changes within just 4 wk for most of our primary and secondary
outcomes (other than weight) (15,23,24). Participants were
instructed to exclude other sources of plant-meat alternatives
(i.e., tofu and tempeh) in both phases. Fish products were
excluded in the Plant phase. However, to promote recruitment, a
modest compromise was made to allow ≤8 oz (227 g) sh/wk
in the Animal phase with the qualication that no sh was
allowed in the 48 h before a study blood draw (17). All Plant
products were supplied by Beyond Meat and distributed on-site
at the research facility. All Animal products were supplied by
a San Francisco–based organic foods delivery service; the red
meat sources were grass-fed. The cut of ground beef purchased
was “regular” (i.e., 80% lean, 20% fat), which is the type of
ground beef most commonly purchased by US consumers (25).
Notably, the saturated fat content of the animal-meat burgers was
9 g/burger compared with 6 g/burger for the Beyond Meat
product. Participants purchased all other items and ingredients
for their own meals and were encouraged to prepare meals
themselves. They were allowed to eat out occasionally provided
they followed study requirements. Tab l es 1 and 2provide the
study products’ nutrient proles and ingredients. Supplemental
Tabl e 1 provides a weekly order overview.
Measures
Self-reported sociodemographic data on age, gender,
race/ethnicity, marital status, education, and employment
status were collected during the enrollment phase.
Diet data.
Three types of dietary data were collected. During eligi-
bility screening, all potential participants completed a brief
questionnaire about current habitual meat intake. During the
study, participants logged all of their food intake for 3 d (2
weekdays and 1 weekend day) biweekly from week 0 to 16
using Cronometer (Cronometer Pro, Nutrition Tracking Software
for Professionals; https://cronometer.com/pro). In addition, 2
unannounced 24-h multiple-pass diet recall interviews were
administered by a dietitian at baseline and at the end of each
8-wk phase using Nutrition Data System for Research (NDS-R)
(Nutrition Coordinating Center, University of Minnesota) (26).
Adherence to the protocol for consuming ≥2 servings/d was
determined from a composite score with a 50:50 weighting of
Cronometer data and the biweekly survey that asked how many
servings were consumed that week.
Metabolic and anthropometric data.
Fasting blood concentrations for TMAO, IGF-1, lipids,
glucose, and insulin were determined from samples collected
at baseline and weeks 2, 4, 8 (phase 1), 10, 12, and 16
(phase 2). TMAO was measured by LC with tandem MS
(Cleveland HeartLab) (27). IGF-1, glucose, insulin, and blood
lipid concentrations were analyzed by standard methodolo-
gies, all at the Core Laboratory for Clinical Studies (Wash-
ington University, St Louis, MO) (28–33). Height, body
weight, and blood pressure data were collected at Stan-
ford’s Clinical and Translational Research Unit during clinic
visits.
Physical activity.
Participants’ physical activity was assessed at baseline and at
4 and 8 wk of each phase with the International Physical Activity
Questionnaire short form (34).
Microbiome assessment.
A detailed methodology for stool sample collection and 16S
analysis is provided in the Supplemental Methods.
Downloaded from https://academic.oup.com/ajcn/article/112/5/1188/5890315 by guest on 22 November 2020

Effect of plant vs. animal meat on TMAO and CVDRFs 1191
TABLE 2 Ingredients of Plant and Animal products1
Plant products Animal products
Product Ingredients Product Ingredients
Burger Water, pea protein isolate, expeller-pressed canola oil, rened
coconut oil, rice protein, natural avors, cocoa butter, mung
bean protein, methylcellulose, potato starch, apple extract,
pomegranate extract, salt, potassium chloride, vinegar,
lemon juice concentrate, sunower lecithin, beet juice
extract, carrot
Burger Beef (seasonings added by
participants)
Beef crumbles Water, pea protein isolate, canola and sunower oil, rice our,
spice, tomato powder, yeast extract, sugar, potassium
chloride, tapioca maltodextrin, citric acid, salt, acacia gum,
onion extract, natural avor, garlic extract
Ground beef Beef (seasonings added by
participants)
Breakfast sausage Water, pea protein isolate, expeller-pressed canola oil, rened
coconut oil, natural avors, rice protein, methylcellulose,
sunower protein, mung bean protein, nutritional yeast
(dried yeast, niacin, pyridoxine hydrochloride, thiamin
hydrochloride, riboavin, folic acid, cyanocobalamin),
apple extract, salt, vinegar, lemon juice concentrate,
sunower lecithin
Good Morning pork
breakfast sausage
Pork, water, sea salt, organic herbs,
spices
Hot Italian sausage Water, pea protein isolate, rened coconut oil, sunower oil,
natural avor, rice protein, fava bean protein, potato starch,
salt, fruit juice, vegetable juice, apple ber, methylcellulose,
citrus extract, calcium alginate casing
Hot Italian sausage Pork, water, organic spices, organic
chili pepper, sea salt, organic
evaporated cane syrup, organic
garlic, organic paprika
Brat sausage Water, pea protein isolate, rened coconut oil, sunower oil,
natural avor, rice protein, fava bean protein, potato starch,
salt, fruit juice, vegetable juice, apple ber, methylcellulose,
citrus extract, calcium alginate casing
Pork bratwurst Pork shoulder, pork fatback, salt,
milk powder, white and black
pepper, ginger, mustard powder,
nutmeg
Grilled chicken strips Water, soy protein isolate, pea protein isolate, rice our,
canola and sunower oil, soy ber, yeast extract, carrot
ber, maltodextrin, natural avor, spices, distilled vinegar,
titanium dioxide, salt, sugar, molasses powder, potassium
chloride, paprika
Chicken breast Chicken (seasonings added by
participants)
Lightly seasoned
chicken strips
Water, soy protein isolate, pea protein isolate, rice our,
canola and sunower oil, soy ber, yeast extract, carrot
ber, maltodextrin, natural avor, spices, distilled vinegar,
titanium dioxide, salt, sugar, molasses powder, potassium
chloride, paprika
1Animal, animal meat; Plant, plant-based alternative meat.
Process measures and poststudy diet preferences.
A 5-point Likert scale was used to assess food satisfaction
with both Plant and Animal products biweekly throughout the
study. Gastrointestinal symptoms were assessed at baseline and
biweekly using the Gastrointestinal Symptoms Rating Scale (35,
36).
Statistical analysis
Participant baseline characteristics are presented as
means ±SDs or percentages. Nutrient comparisons between
Plant and Animal phases were performed using NDS-R diet
assessment data and paired ttests. Adherence rates for Plant
compared with Animal during the 8-wk phases were contrasted
using combined diet data from Cronometer and the biweekly
self-reported data on daily study product servings consumed
(paired ttests).For our primary outcome, we used a linear mixed-
effects model to investigate if the change in TMAO values from
baseline at the end of each phase was signicantly different
for Plant compared with Animal (“meat type”), adjusting for
the xed effect of diet order (e.g., study arm), phase, and the
random effect of correlated observations for each participant
from the 2 phases. The primary analysis was a complete case
analysis and used participants’ last available laboratory values
in each phase. Participants who did not complete both phases
(i.e., crossover) were excluded from the primary analysis
and accounted for in exploratory analysis. To investigate no
difference in the change of TMAO values between diet types,
a 2-sided likelihood ratio test was used. We set a signicance
level of 0.05 for all analyses. Data were analyzed in R version
3.5.1 (R Foundation for Statistical Computing; 2 July, 2018). The
primary packages used for modeling were “lme4” and “lmerTest”
(37–39).
Similarly, for our secondary outcomes, we used separate
mixed-effects models to evaluate fasting plasma IGF-1, lipids,
insulin, glucose, blood pressure, and weight for Plant compared
with Animal, adjusting for the order of diet and repeated
measures. A 2-sided likelihood ratio test was used to assess no
difference between the diet types.
Sample size determination was based on available resources,
rather than a formal power calculation. We performed an ad hoc
Downloaded from https://academic.oup.com/ajcn/article/112/5/1188/5890315 by guest on 22 November 2020

1192 Crimarco et al.
TABLE 3 Study participants’ baseline sociodemographic, anthropometric, and metabolic characteristics1
Plant–Animal Animal–Plant All
Gender, n
Female 15 9 24
Male 3 9 12
Age, y 49.3 ±11.7 51.1 ±16.0 50.2 ±13.8
Highest level of education achieved, n
Less than high school 0 1 1
High school degree 1 0 1
Some college 0 4 4
College degree 7 7 14
Some postgraduate school 2 1 3
Postgraduate degree 8 5 13
Race/ethnicity, n
Non-Hispanic white 11 14 25
Hispanic/Latinx 3 0 3
Asian 3 2 5
Black/African American 1 0 1
Other 0 2 2
Weight, kg
Women 73.7 ±16.8 71.5 ±14.1 72.8 ±15.6
Men 80.9 ±7.0 93.2 ±18.6 88.7 ±17.1
Both sexes 74.9 ±15.7 82.3 ±19.5 78.0 ±17.6
BMI, kg/m2
Women 27.7 ±5.7 27.0 ±4.3 27.4 ±5.1
Men 26.2 ±2.2 29.7 ±6.0 28.8 ±5.4
Both sexes 27.4 ±5.2 28.3 ±5.3 27.9 ±5.2
Blood pressure, mm Hg
Systolic 112 ±13 116 ±10 114 ±11
Diastolic 67 ±771±969±8
Blood lipids, mg/dL
Total cholesterol 212 ±37 191 ±41 201 ±40
HDL cholesterol 60 ±12 59 ±15 60 ±13
LDL cholesterol 130 ±32 113 ±35 122 ±34
Triglycerides 107 ±44 92 ±32 100 ±39
Fasting glucose, mg/dL 94 ±799±996±8
Fasting insulin, μIU/mL 8.3 ±5.3 9.4 ±5.8 8.8 ±5.5
TMAO concentrations, μM3.5±1.8 3.4 ±2.1 3.4 ±1.4
IGF-1 concentrations, ng/mL 154.6 ±49.5 153.6 ±62.9 154.1 ±55.8
Physical activity, metabolic equivalent minutes per week 2578.1 ±2120.3 4342.7 ±3412.5 3460.4 ±2939.6
1Values are ns or means ±SDs. Animal, animal meat; IGF-1, insulin-like growth factor 1; Plant, plant-based alternative meat; TMAO,
trimethylamine-N-oxide.
power analysis for a paired ttest, which assumed no order effect
for the primary analysis aforementioned. With 38 participants, the
trial had 80% power to detect a −2.5 mean difference in TMAO
values for Plant compared with Animal, assuming an SD of 5.4
at a 5% signicance level (15).
Results
Demographics
Tabl e 3 presents baseline demographic, anthropometric, and
metabolic characteristics. Of the 284 potential participants that
completed an initial online screener for the study, 38 participants
were randomly assigned to the intervention arms, including 2
pairs of same-household participants. Thirty-six participated in
both phases of the intervention (Figure 1). One of the originally
randomly assigned 38 participants dropped out owing to a lack
of satisfaction with study products. Two of the remaining 37
participants dropped out mid-study owing to unrelated health
issues; however, 1 provided complete data through the middle of
the second phase, which was used as the end-phase endpoint for
that 1 participant for the second phase, resulting in n=36 for
complete analyses. The participants were mostly female (67%),
Caucasian (69%), and college educated (83%). BMI ranged from
18 to 39 kg/m2; age ranged from 21 to 75 y.
Diet
For both treatment orders, servings per day over the 8-wk
phases were similar (mean ±SD: 2.5 ±0.6 and 2.6 ±0.7 for
Plant and Animal, respectively, P=0.76).
Figure 2 presents diet data for sodium, saturated fat, ber,
and protein, with a breakdown of plant compared with animal
protein in Supplemental Figure 1. A more extensive prole of
nutrient intake including vitamins, minerals, and carbohydrates
as determined by NDS-R (2 d at the end of each phase) is provided
Downloaded from https://academic.oup.com/ajcn/article/112/5/1188/5890315 by guest on 22 November 2020

Effect of plant vs. animal meat on TMAO and CVDRFs 1193
18 Excluded
7 Did not meet eligibility criteria
8 No longer interested
3 Other
42 Attended study orientation & signed
informed consent
38 Randomized
4 No longer interested
following orientation
19 Randomized to
Plant-Animal sequence
1 Discontinued due to
problem with diet
18 Contributed to analyses
1 was on each phase for 6 wk only,
1 was on the second phase for
4 wk only
18 finished Phase 1 (Plant)
18 finished Phase 1 (Animal)
17 finished Phase 2 (Animal)
1 Discontinued due to
unrelated health reasons
284 Individuals filled out online screener
186 Excluded
164 rejected due to ineligibility
22 could not reach for phone screen
98 Individuals completed phone screen
38 Excluded
30 rejected after phone screen
8 did not show up for in person screen
60 Individuals screened in person for eligibility
1 Discontinued due to
unrelated health reasons after
completing 4 of 8 wk
18 finished Phase 2 (Plant)
19 Randomized to
Animal-Plant sequence
18 Contributed to analyses
1 was on each phase for 6 wk only
FIGURE 1 Participant owchart. Animal, animal meat; Plant, plant-based alternative meat.
in Supplemental Table 2, and as determined by Cronometer (≥3
d every other week) in Supplemental Table 3.
In Figure 2 selected nutrient intakes are presented in 3
categories: 1) overall (total), and then separately for 2) amounts
contributed by study products being provided to participants, and
3) all other nonproducts that participants chose themselves. As
expected from product nutrient proles (Tab le 1) and the similar
servings consumed, the Plant products were higher in ber and
plant protein, similar in sodium and total protein, and lower
in saturated fat compared with the Animal products (Tab le 1,
Figure 2). Energy intake was similar for Plant and Animal overall,
and for products and nonproducts (Supplemental Figure 2).
For nonproducts, across all 4 nutrients, the reported intake was
similar for Plant and Animal. With 1 exception—ber—the total
intake contrasts mirrored those of the product differences: higher
in plant protein; similar in calories, sodium, and total protein;
and lower in saturated fat for Plant compared with Animal.
For ber, despite being signicantly different in intake from
products, the absolute mean difference of >5 g total ber (27.9
compared with 22.3 g) for Plant compared with Animal was not
statistically signicant. As is evident in Figure 2, nutrient intakes
in nonproducts were similar, and thus the differences in total
Downloaded from https://academic.oup.com/ajcn/article/112/5/1188/5890315 by guest on 22 November 2020

1194 Crimarco et al.
Plant Phase
Animal Phase
27.9
5.4
22.5
22.3
0.4
21.9
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
Overall Product Nonproduct
Fiber, g
Mean Daily Fiber Intake
26
8
17
33
13
20
0
5
10
15
20
25
30
35
40
45
Overall Product Nonproduct
Saturated Fat, g
Mean Daily Saturated Fat Intake
2863
939
1924
2788
840
1948
0
500
1000
1500
2000
2500
3000
3500
4000
Overall Product Nonproduct
m, mguidoS
Mean Daily Sodium Intake
85
41 44
86
42 45
0
20
40
60
80
100
120
Overall Product Nonproduct
Protein, g
Mean Daily Protein Intake
P = 0.96
P = 0.23
P = 0.85
P < 0.01
P = 0.02
P = 0.25
P = 0.23
P < 0.001
P = 0.94
P = 0.46
P = 0.72
P = 0.92
AB
CD
FIGURE 2 Nutrient data for means of sodium (A), saturated fat (B), ber (C), and protein (D) consumed daily between the 2 diet phases (n=36). Product
indicates the nutrients from only Plant or Animal, whereas Nonproduct indicates the nutrients from all other sources. Together they add up to the overall
amounts of nutrients. Paired ttests were conducted for energy intake and each nutrient to assess for any differences between the Plant and Animal phases. Data
are based on estimates determined using NDS-R. Animal, animal meat; NDS-R, Nutrition Data System for Research; Plant, plant-based alternative meat.
intake, when present, were due primarily to the differences in the
Plant and Animal products. The nutrient intakes were similar by
order (Supplemental Figure 2).
Physical activity
No signicant differences in physical activity levels were
observed between Plant and Animal phases (Supplemental
Tabl e 4 ).
Primary and secondary outcomes
Overall, the difference in TMAO after 8 wk of Plant compared
with Animal was statistically signicant (P=0.02) (Tabl e 4).
However, as Figure 3 presents, a signicant order effect was
observed. For the n=18 assigned to receive the Plant rst,
mean ±SEM TMAO concentrations were not signicantly
different at weeks 8 and 16 (the end of the 8-wk intervention
phases) (2.5 ±0.4 and 3.0 ±0.6 μM, respectively, P=0.28,
Wilcoxon test). For the n=18 receiving Animal rst, TMAO
concentrations were signicantly lower for the Plant than for
the Animal phase (2.9 ±0.4 and 6.4 ±1.5 μM, respectively,
P=0.007).
LDL cholesterol and weight were signicantly lower for
the Plant than for the Animal phase. Fasting concentrations of
IGF-1, insulin, glucose, HDL cholesterol, and triglycerides,
and blood pressure, were not signicantly different between the
Plant and Animal phases. No order effects were observed among
Downloaded from https://academic.oup.com/ajcn/article/112/5/1188/5890315 by guest on 22 November 2020

Effect of plant vs. animal meat on TMAO and CVDRFs 1195
TABLE 4 Outcome levels at the end of the 8-wk phases1
Outcome
Plant,
mean ±SEM
Animal,
mean ±SEM
Plant–Animal difference,
mean (95% CI) Pvalue2
Primary
TMAO,3μM2.7±0.3 4.7 ±0.9 −2.0 (−3.6, −0.3) 0.012
Secondary
IGF-1, ng/mL 147.6 ±7.5 152.3 ±8.3 −4.7 (−13.9, 4.5) 0.30
Weight, kg 78.7 ±3.0 79.6 ±3.0 −1.0 (−1.5, −0.5) <0.001
Insulin, μIU/mL 9.2 ±1.1 8.8 ±0.9 0.4 (−0.7, 1.5) 0.38
Glucose, mg/dL 94.9 ±1.6 94.5 ±1.4 0.5 (−1.8, 2.8) 0.65
Lipids, mg/dL
LDL-C 109.9 ±4.5 120.7 ±4.5 −10.8 (−17.3, −4.3) 0.002
HDL-C 62.5 ±2.2 61.8 ±2.5 0.7 (−2.4, 3.8) 0.66
Triglycerides 99.7 ±7.3 100.2 ±7.0 −0.6 (−10.5, 9.2) 0.89
Blood pressure, mm Hg
Systolic 114.5 ±2.1 113.1 ±1.9 1.2 (−1.4, 4.1) 0.31
Diastolic 70.0 ±1.4 68.8 ±1.2 1.1 (−0.8, 3.2) 0.20
1n=36. Animal, animal meat; HDL-C, HDL cholesterol; IGF-1, insulin-like growth factor 1; LDL-C, LDL cholesterol; Plant, plant-based alternative
meat; TMAO, trimethylamine-N-oxide.
2Likelihood ratio test from mixed-effects model evaluating change from baseline for each product type (Plant compared with Animal), adjusting for
order and phase.
3Signicant order effect in model (P=0.023).
secondary outcomes. Supplemental Table 5 provides outcome
data by sequence order.
Microbiome
Analysis of stool 16S rRNA proles did not reveal associations
of overall gut microbiota composition with diets, diet order,
or TMAO production. Some specic taxa were associated
with dietary interventions, but none were associated with
TMAO production. Consistent with the gut microbiome being
highly individualized, the vast majority of variance in the 16S
rRNA proles was derived from participants (80.1%). It is
possible that given a larger study cohort or function-focused
analysis of the microbiome, differences between diets, order
effects, or individualized TMAO-production associations could
be identied. The Supplemental Results and Supplemental
Figures 3 and 4present an extended results description and
gures related to gut microbiota proling.
Product satisfaction and gastrointestinal issues
Satisfaction with products was generally high. Most products
scored a mean of ≥3.5 on a 5-point rating scale (5 being highest).
Four of the 6 matched products (e.g., Plant compared with Animal
patties) scored almost identically; 2 of the 6 Plant products scored
<3.0 on average, and lower than their matched Animal products
(Supplemental Table 6).
Participants reported negligible study changes for most of the
13 gastrointestinal symptoms queried for both product types. The
most notable changes, although modest, were increases in ab-
normal abdominal distention in both phases, and a decrease in
daily bowel movements during the Animal phase (Supplemental
Tabl e 7 ).
Discussion
This randomized crossover-design trial compared the effect of
consuming Plant as opposed to Animal products on emerging
health risk factors, cardiometabolic risk factors, and the gut
microbiome in a population of generally healthy adults after
8 wk of consuming ≥2 servings/d of each product type. TMAO,
the primary outcome, improved signicantly in the plant-based
phase relative to the Animal phase. However, an intervention
order effect was identied. Participants who consumed Plant rst
had TMAO concentrations that were not signicantly different
from Animal at the end of that phase; the signicantly higher
TMAO concentrations after the Animal phase were observed only
among the participants who consumed Animal rst and Plant
second. In addition, LDL-cholesterol concentrations and body
weight were lower in the Plant phase.
Consistent with our ndings, several studies have previously
reported that red meat intake raises TMAO concentrations (15–
17). Our nding of an order effect is particularly interesting.
Among the group getting Animal rst and Plant second, the
observed mean increase in TMAO in the Animal phase decreased
back to baseline concentrations within 2 wk during the Plant
phase, and remained stable through 8 wk, suggesting no carryover
effect. In stark contrast, those assigned to the Plant phase rst
had no mean increase in TMAO, and no apparent effect to carry
over, and yet this group was observed to have no mean increase
in TMAO when shifted to the Animal phase. Although some
may take issue with the lack of a washout period in the study
design, in this case it appears the absence of a washout may have
been important in revealing the signicant order effect. Koeth et
al. (40) investigated the effect on TMAO of feeding -carnitine,
abundant in red meat, to vegetarians. They reported no increase
in TMAO, speculating that this may be due to their “vegetarian”
microbiomes. This nding suggests that those in the current study
who were assigned to Plant rst, and were therefore following
a vegetarian diet for 8 wk, may have differentially altered their
microbiomes compared with those who consumed Animal rst,
Downloaded from https://academic.oup.com/ajcn/article/112/5/1188/5890315 by guest on 22 November 2020

1196 Crimarco et al.
A
B
TMAO, μM
TMAO, μM
FIGURE 3 TMAO concentrations at 7 time points (baseline and biweekly during the 16-wk protocol) by randomization order: Plant→Animal (n=18)
and Animal→Plant (n=18). (A) Boxplots with medians, IQRs, 5th and 95th percentiles, and extreme values. (B) Spaghetti plots of each participant’s data
across the full protocol; means are represented by lled squares. Signicant differences in TMAO concentrations after 8 wk of Plant compared with Animal
were observed (P=0.01). However, a signicant order effect was observed (P=0.02). For participants assigned to the Plant group rst, TMAO concentrations
were not signicantly different at weeks 8 and 16 (mean ±SEM: 2.5 ±0.4 and 3.0 ±0.6 μM, respectively, P=0.28, Wilcoxon test). For participants assigned
to the Animal group rst, TMAO concentrations were signicantly lower during the Plant than during the Animal phase (mean ±SEM: 2.9 ±0.4 and 6.4 ±1.5
μM, respectively, P=0.007). Animal, animal meat; Plant, plant-based alternative meat; TMAO, trimethylamine-N-oxide.
in such a way as to prevent the production of TMAO when
the Animal phase came second. Although the 16S microbiome
analysis conducted in the current study was unable to identify
these changes, it is difcult to know if the ndings reect a
true null effect. The data suggest that the plant-rst group may
have suppressed trimethylamine (TMA)-producing taxa leading
to less TMAO in the meat-second phase (TMA is a precursor
to TMAO). However, this is merely a hypothesis that requires
Downloaded from https://academic.oup.com/ajcn/article/112/5/1188/5890315 by guest on 22 November 2020

Effect of plant vs. animal meat on TMAO and CVDRFs 1197
in-depth analysis to investigate fully. Therefore, it is possible
that even if there is an effect on TMA production by the dietary
interventions, it is not detectable by 16S analysis (e.g., if the taxa
that underlie TMA production differ between individuals). It is
possible that such changes are only detectable with additional in-
depth microbiome investigation focused on genes and encoded
functions relevant to TMAO generation. Although we expect that
differences in ber content may inuence the microbiota, how
such changes would interact with TMA production is currently
unclear. We also have observed in another intervention focused
on ber, currently in preparation, that the effect of ber on the
microbiota over a similar duration is individualized and more
modest than anticipated (Carter MM, Topf MA, Wastyk HC,
Sonnenburg ED, Sonnenburg JL and Gardner CD, unpublished
results, 2020). Future studies of potential microbiome changes
that could be determined by metagenomic or other function-
targeted analyses are warranted.
We also observed considerable variability among participants
in TMAO concentration changes; 2 of the 18 participants in
the Animal→Plant group had particularly large excursions of
TMAO during the Animal phase, whereas others had very
little changes from baseline. We assessed baseline red meat
consumption, baseline TMAO concentrations, and adherence as
possible explanations for these response differences, but were
unable to identify meaningful differences between the responders
and the nonresponders.
Although plant-based meat products could support individuals
in adhering to national and global health recommendations to
reduce red meat consumption (3,41–43), concerns have been
raised about their overall health effects (5). Some of these
products fall under the NOVA category of ultra-processed food,
which have been reported to lead to weight gain compared with
whole foods (9,10). However, the nding in the current study was
that weight was modestly but statistically signicantly lower after
8 wk on the Plant than on the Animal phase. Notably, this was
observed despite no differences in reported total energy intake or
physical activity levels between each phase.
The other cardiometabolic benet for the Plant phase was
lower concentrations of LDL cholesterol. This was consistent
with the lower saturated fat and higher ber and plant protein
during the Plant phase, all of which have been established
to lower LDL cholesterol (11,14,44). The observed lack of
differences in blood pressure between the Plant and Animal
phases was consistent with similar sodium intakes during the
2 study phases, which were similar in terms of products,
nonproducts, and total sodium (12). It has been reported that
IGF-1 concentrations are higher in vegetarians than omnivores,
but no IGF-1 differences were observed in the current study (21,
22). In addition, no differences were observed in other secondary
outcomes, although there were no specic a priori hypotheses
as to why glucose, insulin, HDL cholesterol, or triglycerides
might be differentially affected by the Plant and Animal phases—
the balances of carbohydrates and fats were relatively similar
overall.
Beyond its potential benets for human health, others have
advocated for the transition to more plant-based diets because
of the broader environmental, animal, and societal benets of
displacing red meat (2,5,45–48). Plant-based meats may play
an important role in facilitating a global shift toward a more
sustainable food system based on more plant-based and fewer
animal-based foods (45). Although this study did not assess these
impacts,Esheletal.(47) found that by replacing meat with
protein-conserving plant alternatives, Americans could satisfy
key nutritional requirements while eliminating pastureland use
and reducing cropland, nitrogen fertilizer usage, and greenhouse
gas emissions. In addition, Gardner et al. (2) reported signicant
reductions in greenhouse gasses when modeling a 25% shift
from animal to plant protein. Notably, both the Plant and Animal
products in the current study accounted for ∼25% of total calories
and half of total protein.
This study included several strengths. First was the crossover
design that allowed each participant to serve as their own control.
Second were various aspects of internal validity, including high
participant retention and minimal missing data. A third strength
was a high treatment delity supported by providing Plant
and Animal products, and the high and comparable participant
satisfaction with the 2 types of products. A fourth strength
was the collection of multiple sources of dietary intake data:
NDS-R, Cronometer, and food survey data. Finally, a database
lock, predetermined database quality check, and third-party data
analysis reduced bias.
This study’s potential limitations include its crossover de-
sign without a second baseline measurement and the allowance
for some participants to take a break between the 2 phases (12
of the 36 participants took a 1- to 7-wk break). However, a
close examination of the data from those individuals suggested
no apparent pattern of differences from the 24 participants that
transitioned without a break, and the primary testing between
phases remained focused on contrasting the end of each 8-wk
phase. The potential issues with the lack of washout period—
both limitations and opportunities—were described previously.
Another limitation may have been the inclusion and allowance of
chicken and sh in the study, because both of these have different
effects on TMAO than red meat, making the Plant/Animal
contrast less controlled. However, only small amounts of
chicken or sh were consumed by participants, sh was not
allowed within 48 h of a blood sampling which should have
eliminated its effect on serum TMAO concentrations, and the
allowance of these small amounts facilitated recruitment and
retention. Whereas the plant-meat and animal-meat products
were provided, the remainder of the diet was self-selected, thus
limiting the ability to control the intake of other foods and
nutrients. Although this limited the rigor of dietary control, it
increased the generalizability of the ndings. Finally, this study
used 1 of the many different types of plant-meat formulations,
and 1 set of matched meat products; results could have differed
for a different type of plant-based meat and for other cuts of
meat. For example, the “regular” ground beef used in the study
was 80:20 (lean:fat), with 9 g saturated fat/burger compared with
6 g saturated fat/burger for the comparable Beyond Meat
products. Other options for ground beef include lean (85:15) and
extra lean (90:10), both with lower saturated fat content. Of the
different lean:fat types, regular is the most commonly consumed
type in the United States (25).
In light of growing, consistent recommendations to reduce
red meat intake for optimal cardiovascular health, plant-based
meats offer a potentially healthy alternative (5,48). Notably,
their growing popularity among consumers has been coupled
with rising critiques of their ultra-processed composition and
potential adverse health consequences. Until now, there has been
a paucity of data upon which to evaluate these claims. This
study found several benecial effects and no adverse effects
Downloaded from https://academic.oup.com/ajcn/article/112/5/1188/5890315 by guest on 22 November 2020

1198 Crimarco et al.
from the consumption of plant-based meats. The interesting
order effect observed for TMAO reinforces prior ndings about
the microbiome’s potential personalization effects, and warrants
further study.
We acknowledge many individuals that contributed meaningfully to the
study. Jennifer Robinson oversaw the project and staff. Carrie McKinley,
Lindsey Durand, Tania Davila, and Diane Demis conducted the diet
assessments. Alexandra Garrity assisted with data collection and study
intervention materials. Dalia Perelman, Heyjun Park, and Claire Bladier
provided feedback and critical review of the manuscript.
The authors’ responsibilities were as follows—CDG: designed the
research project and contributed to writing both the initial and nal drafts
of the manuscript; AC: wrote the initial and nal drafts of the manuscript;
SS, CP, TS, KC, JL, PF-S, MMC, MAT, HCW, EDS, and JLS: provided
feedback and critical revisions of the manuscript; KC and JL: conducted the
statistical analyses; and all authors: read and approved the nal manuscript.
CDG received funding for the study from Beyond Meat in the form of an
unrestricted research gift made to Stanford University. All other authors report
no conicts of interest.
References
1. Millen BE, Abrams S, Adams-Campbell L, Anderson CA, Brenna JT,
Campbell WW, Clinton S, Hu F, Nelson M, Neuhouser ML, et al.
The 2015 Dietary Guidelines Advisory Committee scientic report:
development and major conclusions. Adv Nutr 2016;7(3):438–44.
2. Gardner CD, Hartle JC, Garrett RD, Offringa LC, Wasserman AS.
Maximizing the intersection of human health and the health of the
environment with regard to the amount and type of protein produced
and consumed in the United States. Nutr Rev 2019;77(4):197–215.
3. US Department of Health Human Services (DHHS) and USDA. 2015–
2020 Dietary Guidelines for Americans [Internet]. 8th ed. Washington
(DC): US DHHS and USDA; 2015 [cited 30 April, 2020]. Available
from: https://health.gov/dietaryguidelines/2015/guidelines/.
4. Sanchez-Sabate R, Sabaté J. Consumer attitudes towards environmental
concerns of meat consumption: a systematic review. Int J Environ Res
Public Health 2019;16(7):1220.
5. Hu FB, Otis BO, McCarthy G. Can plant-based meat alternatives be part
of a healthy and sustainable diet? JAMA 2019;322(16):1547–8.
6. Olayanju JB. Plant-based meat alternatives: perspectives on consumer
demands and future directions [Internet]. Jersey City, NJ: Forbes; 2019
[cited 22 September, 2019]. Available from: https://www.forbes.com/s
ites/juliabolayanju/2019/07/30/plant-based- meat-alternatives-perspec
tives-on-consumer-demands-and- future-directions/#2873b2136daa.
7. Sweeney E. Are beyond meat and impossible burgers better for you?
Nutritionists weigh in [Internet]. New York: HuffPost; 2019 [cited 22
September, 2019]. Available from: https://www.huffpost.com/entry/bey
ond-meat- impossible-burger-healthy_l_5d164ad1e4b07f6ca57cc3ed.
8. Hemler EC, Hu FB. Plant-based diets for cardiovascular disease
prevention: all plant foods are not created equal. Curr Atheroscler Rep
2019;21(5):18.
9. Monteiro CA, Cannon G, Moubarac J-C, Levy RB, Louzada MLC,
Jaime PC. The UN Decade of Nutrition, the NOVA food classication
and the trouble with ultra-processing. Public Health Nutr 2018;21(1):5–
17.
10. Hall KD, Ayuketah A, Brychta R, Cai H, Cassimatis T, Chen KY,
Chung ST, Costa E, Courville A, Darcey V, et al. Ultra-processed diets
cause excess calorie intake and weight gain: an inpatient randomized
controlled trial of ad libitum food intake. Cell Metab 2019;30(1):67–
77.
11. Mozaffarian D, Micha R, Wallace S. Effects on coronary heart disease
of increasing polyunsaturated fat in place of saturated fat: a systematic
review and meta-analysis of randomized controlled trials. PLoS Med
2010;7(3):e1000252.
12. Cogswell ME, Mugavero K, Bowman BA, Frieden TR. Dietary sodium
and cardiovascular disease risk—measurement matters. N Engl J Med
2016;375(6):580–6.
13. Anderson JW, Baird P, Davis RH, Ferreri S, Knudtson M, Koraym
A, Waters V, Williams CL. Health benets of dietary ber. Nutr Rev
2009;67(4):188–205.
14. Veronese N, Solmi M, Caruso MG, Giannelli G, Osella AR, Evangelou
E, Maggi S, Fontana L, Stubbs B, Tzoulaki I. Dietary ber and health
outcomes: an umbrella review of systematic reviews and meta-analyses.
Am J Clin Nutr 2018;107(3):436–44.
15. Wang Z, Bergeron N, Levison BS, Li XS, Chiu S, Jia X, Koeth RA, Li
L, Wu Y, Tang WHW, et al. Impact of chronic dietary red meat, white
meat, or non-meat protein on trimethylamine N-oxide metabolism and
renal excretion in healthy men and women. Eur Heart J 2019;40(7):583–
94.
16. Park JE, Miller M, Rhyne J, Wang Z, Hazen SL. Differential effect
of short-term popular diets on TMAO and other cardio-metabolic risk
markers. Nutr Metab Cardiovasc Dis 2019;29(5):513–17.
17. Cho CE, Taesuwan S, Malysheva OV, Bender E, Tulchinsky NF, Yan J,
Sutter JL, Caudill MA. Trimethylamine-N-oxide (TMAO) response to
animal source foods varies among healthy young men and is inuenced
by their gut microbiota composition: a randomized controlled trial. Mol
Nutr Food Res 2017;61(1):1600324.
18. Djuric Z, Mitchell CM, Davy KP, Neilson AP. A Mediterranean
diet does not alter plasma trimethylamine N-oxide concentrations
in healthy adults at risk for colon cancer. Food Funct 2019;10(4):
2138–47.
19. Randrianarisoa E, Lehn-Stefan A, Wang X, Hoene M, Peter A,
Heinzmann SS, Zhao X, Königsrainer I, Königsrainer A, Balletshofer
B, et al. Relationship of serum trimethylamine N-oxide (TMAO) levels
with early atherosclerosis in humans. Sci Rep 2016;6:26745.
20. Pollak M. Insulin and insulin-like growth factor signalling in neoplasia.
Nat Rev Cancer 2008;8(12):915–28.
21. Allen NE, Appleby PN, Davey GK, Kaaks R, Rinaldi S, Key TJ. The
associations of diet with serum insulin-like growth factor I and its main
binding proteins in 292 women meat-eaters, vegetarians, and vegans.
Cancer Epidemiol Biomarkers Prev 2002;11(11):1441–8.
22. Allen N, Appleby P, Davey G, Key T. Hormones and diet: low insulin-
like growth factor-I but normal bioavailable androgens in vegan men.
Br J Cancer 2000;83(1):95–7.
23. Gardner CD, Coulston A, Chatterjee L, Rigby A, Spiller G,
Farquhar JW. The effect of a plant-based diet on plasma lipids in
hypercholesterolemic adults: a randomized trial. Ann Intern Med
2005;142(9):725–33.
24. Sacks FM, Svetkey LP, Vollmer WM, Appel LJ, Bray GA, Harsha
D, Obarzanek E, Conlin PR, Miller ER, Simons-Morton DG, et al.
Effects on blood pressure of reduced dietary sodium and the Dietary
Approaches to Stop Hypertension (DASH) diet. N Engl J Med
2001;344(1):3–10.
25. Krebs A. Popular and versatile – ground beef reigns! [Internet].
Boca Raton, FL: PerishableNews.com; 2019 [cited 20 June, 2020].
Available from: https://www.perishablenews.com/meatpoultry/popular
-and- versatile-ground-beef-reigns/.
26. Feskanich D, Sielaff BH, Chong K, Buzzard IM. Computerized
collection and analysis of dietary intake information. Comput Methods
Programs Biomed 1989;30(1):47–57.
27. Tang WHW, Wang Z, Levison BS, Koeth RA, Britt EB, Fu X, Wu Y,
Hazen SL. Intestinal microbial metabolism of phosphatidylcholine and
cardiovascular risk. N Engl J Med 2013;368(17):1575–84.
28. Bystrom C, Sheng S, Zhang K, Cauleld M, Clarke NJ, Reitz R. Clinical
utility of insulin-like growth factor 1 and 2; determination by high
resolution mass spectrometry. PLoS One 2012;7(9):e43457.
29. Peterson JI, Young DS. Evaluation of the hexokinase/glucose-6-
phosphate dehydrogenase method of determination of glucose in urine.
Anal Biochem 1968;23(2):301–16.
30. Morgan CR, Lazarow A. Immunoassay of insulin: two antibody system:
plasma insulin levels of normal, subdiabetic and diabetic rats. Diabetes
1963;12(2):115–26.
31. Rambaldi DC, Reschiglian P, Zattoni A, Johann C. Enzymatic
determination of cholesterol and triglycerides in serum lipoprotein
proles by asymmetrical ow eld-ow fractionation with on-line, dual
detection. Anal Chim Acta 2009;654(1):64–70.
32. Friedewald WT, Levy RI, Fredrickson DS. Estimation of the
concentration of low-density lipoprotein cholesterol in plasma,
without use of the preparative ultracentrifuge. Clin Chem 1972;18(6):
499–502.
33. Miller WG, Myers GL, Sakurabayashi I, Bachmann LM, Caudill SP,
Dziekonski A, Edwards S, Kimberly MM, Korzun WJ, Leary ET,
et al. Seven direct methods for measuring HDL and LDL cholesterol
compared with ultracentrifugation reference measurement procedures.
Clin Chem 2010;56(6):977–86.
Downloaded from https://academic.oup.com/ajcn/article/112/5/1188/5890315 by guest on 22 November 2020

Effect of plant vs. animal meat on TMAO and CVDRFs 1199
34. Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML,
Ainsworth BE, Pratt M, Ekelund U, Yngve A, Sallis JF, et al.
International physical activity questionnaire: 12-country reliability and
validity. Med Sci Sports Exerc 2003;35(8):1381–95.
35. Svedlund J, Sjödin I, Dotevall G. GSRS—a clinical rating scale for
gastrointestinal symptoms in patients with irritable bowel syndrome and
peptic ulcer disease. Digest Dis Sci 1988;33(2):129–34.
36. Winham DM, Hutchins AM. Perceptions of atulence from bean
consumption among adults in 3 feeding studies. Nutr J 2011;10(1):128.
37. R Development Core Team. R: a language and environment for
statistical computing [Internet]. Vienna, Austria: R Foundation for
Statistical Computing; 2020 [cited 18 May, 2020]. Available from:
https://www.R-project.org/.
38. Bates D, Mächler M, Bolker B, Walker S. Fitting linear mixed-effects
models using lme4. J Stat Soft 2015;67(1):1–48.
39. Kuznetsova A, Brockhoff PB, Christensen RH. lmerTest package: tests
in linear mixed effects models. J Stat Soft 2017;82(13):1–26.
40. Koeth RA, Lam-Galvez BR, Kirsop J, Wang Z, Levison BS, Gu X,
Copeland MF, Bartlett D, Cody DB, Dai HJ, et al. -Carnitine in
omnivorous diets induces an atherogenic gut microbial pathway in
humans. J Clin Invest 2018;129(1):373–87.
41. Eckel RH, Jakicic JM, Ard JD, de Jesus JM, Miller NH, Hubbard
VS, Lee I-M, Lichtenstein AH, Loria CM, Millen BE, et al. 2013
AHA/ACC guideline on lifestyle management to reduce cardiovascular
risk: a report of the American College of Cardiology/American Heart
Association Task Force on Practice Guidelines. J Am Coll Cardiol
2014;63(25 Part B):2960–84.
42. World Health Organization. Q&A on the carcinogenicity of the
consumption of red meat and processed meat [Internet]. Geneva: WHO;
2015 [cited 8 April, 2020]. Available from: https://www.who.int/featur
es/qa/cancer-red- meat/en/.
43. Bouvard V, Loomis D, Guyton KZ, Grosse Y, Ghissassi FE,
Benbrahim-Tallaa L, Guha N, Mattock H, Straif K; International
Agency for Research on Cancer Monograph Working Group.
Carcinogenicity of consumption of red and processed meat. Lancet
Oncol 2015;16(16):1599–600.
44. Jenkins DJ, Kendall CW, Mehling CC, Parker T, Rao AV, Agarwal S,
Novokmet R, Jones PJ, Raeini M, Story JA, et al. Combined effect of
vegetable protein (soy) and soluble ber added to a standard cholesterol-
lowering diet. Metabolism 1999;48(6):809–16.
45. Willett W, Rockström J, Loken B, Springmann M, Lang T,
Vermeulen S, Garnett T, Tilman D, DeClerck F, Wood A, et al.
Food in the Anthropocene: the EAT–Lancet Commission on healthy
diets from sustainable food systems. Lancet 2019;393(10170):
447–92.
46. Poore J, Nemecek T. Reducing food’s environmental impacts through
producers and consumers. Science 2018;360(6392):987–92.
47. Eshel G, Stainier P, Shepon A, Swaminathan A. Environmentally
optimal, nutritionally sound, protein and energy conserving plant based
alternatives to U.S. meat. Sci Rep 2019;9:10345.
48. Godfray HCJ, Aveyard P, Garnett T, Hall JW, Key TJ, Lorimer
J, Pierrehumbert RT, Scarborough P, Springmann M, Jebb
SA. Meat consumption, health, and the environment. Science
2018;361(6399):eaam5324.
Downloaded from https://academic.oup.com/ajcn/article/112/5/1188/5890315 by guest on 22 November 2020