November 30, 2023

expert reaction to study looking at the cardiometabolic effects of omnivorous vs vegan diets

A new study published in JAMA Network Open looks at the effects of omnivorous vs vegan diets in identical twins. 

Prof Margaret Rayman, Co-Director Nutritional Medicine, University of Surrey, said:

“The Abstract says: “The twins randomized to the vegan diet experienced significant mean (SD) decreases in low-density lipoprotein cholesterol concentration (−13.9 [5.8] mg/dL; 95%CI, −25.3 to −2.4mg/dL), fasting insulin level (−2.9 [1.3] μIU/mL; 95%CI, −5.3 to −0.4μIU/mL), and body weight (−1.9 [0.7] kg; 95%CI, −3.3 to −0.6 kg).”

“This is totally predictable, and I am happy that the study showed that.

“Disadvantages of the study are that the sample size was small but that was acknowledged under the Limitations section.

“In the Limitations section, no mention was made of:

• Data from Tong et al. BMC Medicine (2020) 18:353 (and similar data) that showed that compared with meat eaters and after adjustment for socio-economic factors, lifestyle confounders & BMI, the risks of hip fracture were higher in fish eaters, vegetarians and vegans, equivalent to rate differences of 2.9 (0.6–5.7), 2.9 (0.9–5.2), and 14.9 (7.9–24.5) more cases for every 1000 people > 10 y.
  The vegans also had higher risks of total (1.43; 1.20–1.70), leg (2.05; 1.23–3.41), and other main site fractures (1.59; 1.02–2.50) than meat eaters.
• Further data from Tong TYN et al. BMJ 2019; 366:l4897 (| doi: 10.1136/bmj.l4897) that showed that vegetarians/vegans had 43% higher risk of haemorrhagic stroke and 20% higher risk of total stroke than meat eaters.
• No mention was made of data from Appleby PN et al. (Am J Clin Nutr 2016; 103(1): 218-30) that showed that all-cause mortality was no different in regular meat-eaters than in vegetarians [HR 1.00 (95% CI 0.93-1.08)] or in vegans [(HR 1.14 (0.97-1.35), P_{heterogeneity}= 0.056]

“These are important points that need to be stressed in the Limitations section as cardiovascular risk is not the only risk that people have that affects morbidity or mortality.  The data mentioned above are from the EPIC Trial where a considerably larger sample size was recruited.”

Prof Tom Sanders, Professor emeritus of Nutrition and Dietetics, King’s College London (KCL), said:

“This 8 week study compares the effect on risk factors for cardiometabolic disease of a vegan diet compared with a mixed diet (omnivore) in 22 pairs of twins. The main result is a modest reduction (0.36 mmol/L) in low density lipoprotein cholesterol LDL-C). The lower intake of saturated fat and dietary cholesterol would explain their findings. The observed reduction in LDL-C is much smaller than the differences (1 mmol/l or greater) reported in observational studies and with trials of the Portfolio diet. The authors failed to find significantly lower levels of TMAO levels which the authors expected to be lower because the changes in the gut microbiome. However, any effect may have been masked by the large variability in those on the omnivore diet.

“Vegan diets are defined by what they exclude rather than what is consumed. There are good and bad vegan diets. For example, a bad vegan diet would be lacking vitamin B12, and high in fat, salt and sugar. The increased popularity of vegan diets has led to a large increase “vegan junk food” high in fat, salt and sugar. A healthy vegan diet would consist of whole grain cereals, pulses, nuts and fresh fruit and vegetables with vegetable oils low in saturated fat (e.g. olive oil or rapeseed oil) and supplemented with vitamin B12. It is important to recognise that it usually takes several years to deplete body stores of vitamin B12 so it is not surprising that no differences in blood levels of serum vitamin B12 were observed in this study.

“Although the use of twins removed the impact of genetics, it fails to measure the interaction between specific genes and diet. For example, carriage of the apolipoprotein E epsilon 4 allele results in larger increase in LDL-C in response dietary saturated and cholesterol compared with carrier of epsilon 3 allele.”

Dr Duane Mellor, Registered Dietitian and Senior Lecturer, Aston Medical School, Aston University, said:

“This is an interesting trial which unfortunately in the paper itself provides very little information about the diets eaten or the nutritional intake of participants. This is in part because participants were allowed to select their own meals which were delivered and then could add additional snacks to this. It is also unclear from the paper how the number of participants was selected to be 22 pairs of twins and then how these twins were randomised, especially given the nature of this type of study where participants would know what they are eating, and cannot be blinded (as can be done with pharmaceutical studies),  is particularly important to avoid the risks of introducing sources of bias or statistical error. When delving into the supplementary information it was apparent that those being provided the vegan/plant-based diet consumed on average about 200kcal a day less than those consuming the mixed/ omnivorous diet. This could explain the non-significant reduction in weight and perhaps at least partly explain the reduction in LDL cholesterol.

“It is also difficult to fully interpret the data about cholesterol as total cholesterol changes are not reported. This is important as LDL cholesterol levels were estimated in this study, in the same way it is done in a routine health test and uses the values of total cholesterol, HDL cholesterol and a type of fat in the blood called triglycerides to calculate the amount of LDL cholesterol. So it is possible that there could be confounding caused by the differences in total cholesterol levels in this study.

“It is also interesting that the levels of dietary satisfaction (details also only found in the supplementary information) were lower in the vegan/plant-based group. This supports the idea when individuals wish to and are encouraged to follow a healthier diet, it should be based around their preferences and not based on one particular dietary approach or another favoured by their health professional or online influencer. Having said that a vegan or plant-based diet can be healthy as long as it includes a range of foods including pulses, nuts, seeds, vegetables and fruit, with care being taken to include sources of iodine, vitamin B12, iron, calcium and vitamin D.

“Overall this is an interesting study undertaken in twins, but it really needs the details in the extra supplementary information to make sense, so it can be seen that the vegan/plant-based group ate fewer calories and it is important to appreciate that the people in this study were provided with all their food, which can be quite different to where people are making their own food choices.”

‘Cardiometabolic Effects of Omnivorous vs Vegan Diets in Identical Twins’ by Matthew J. Landry et al. was published in JAMA Network Open at 16:00 UK Time Thursday 30 November.

DOI: 10.1001/jamanetworkopen.2023.44457

Declared interests

Dr Duane Mellor: I have no conflicts of interest in relation to this study

Prof Tom Sanders: Honorary Nutritional Director HEART UK, Member of Scientific Committee British Nutrition Foundation.

Prof Margaret Rayman: I do not have any interests which might be regarded by a reasonable and objective third party as giving rise to a conflict with my role as an expert in this story.