By Michael H. Crawford, MD, Editor

SYNOPSIS: A randomized, open, cross-over study of a vegetarian diet (VD) compared to the Mediterranean diet (MD) over three months showed that both reduced body weight and body fat mass. The VD decreased LDL cholesterol, and MD reduced triglyceride levels. The MD reduced inflammatory markers, and the VD reduced vitamin B12 levels.

SOURCES: Sofi F, Dinu M, Pagliai G, et al. Low-calorie vegetarian versus Mediterranean diets for reducing body weight and improving cardiovascular risk profile. CARDIVEG study (Cardiovascular Prevention With Vegetarian Diet). Circulation 2018;137:1103-1113.

Anderson CAM. Dietary patterns to reduce weight and optimize cardiovascular health: Persuasive evidence for promoting multiple, healthful approaches. Circulation 2018;137:1114-1116.

Previous studies of diet and cardiovascular (CV) risk factors have suffered from population bias, lack of randomization, and retrospective analyses. Investigators from Florence, Italy, performed a randomized, open, cross-over study of the lacto-ovo vegetarian diet (VD) to the Mediterranean diet (MD) during two three-month intervals. The authors recruited subjects between the ages of 18-75 years (mean, 51 years; 78% female). Prospective recruits demonstrated a CV risk estimate of < 5% at 10 years, were overweight (body mass index [BMI] > 25 kg/m²), and presented with one or more of the following risk factors: total cholesterol > 190 mg/dL, LDL cholesterol > 115 mg/dL, or glucose > 110 mg/dL, but < 126 mg/dL. The subjects were told not to change their lifestyle or exercise habits. No weight loss goal was given, and both diets were low calorie.

The primary outcome was the difference in BMI and fat mass (bioelectrical impedance technique) changes between the two diets. Secondary endpoints were changes in blood CV risk parameters, such as lipid levels, glucose levels, oxidative stress profiles, and inflammation markers. Of the 118 patients randomized, 107 completed at least one phase of the study and were included in the analysis, and 100 completed the entire study.

Both diets reduced body weight and composition equally. Significant differences in some biochemical parameters were observed (P < 0.01 for all). Vitamin B12 levels fell in the VD group and increased in the MD group. Uric acid fell in the VD group and rose in the MD group. LDL cholesterol fell in the VD (128 to 121 mg/dL) and remained mostly steady in the MD group (124 to 126 mg/dL). Triglycerides increased among VD subjects (109 to 115 mg/dL) and decreased among MD subjects (115 to 108 mg/dL). There were no significant differences in oxidative stress parameters. Only one of 13 inflammatory markers showed a significantly different response to the two diets: Interleukin-17 increased in the VD group (3.7 to 5.1 pg/mL) and decreased in the MD group (5.5 to 3.5 pg/mL). More subjects in the VD group achieved target levels of at least one CV risk factor compared to the MD subjects (46 vs. 35 patients). The authors concluded that VD and MD were equally effective at lowering body weight and fat mass. VD was more effective at lowering LDL cholesterol, and MD was more effective at lowering triglyceride levels.


Much has been written about the benefits of the MD and a few studies have shown that it increases longevity. Less is known about the VD, so this comparison study is of interest. Since both diets were in low calories, it is not surprising that they were both equally effective at lowering weight and fat mass and produced equivalent effects on oxidative stress measures. The only differences detected were in inflammatory markers and lipid levels. The VD lowered LDL, probably because of the absence of meat, and the MD lowered triglycerides, probably because of fewer carbohydrates. Interestingly, the VD increased interleukin-17 levels, perhaps because it lowered vitamin B12 levels, which may have increased homocysteine levels (not measured) and increased inflammation. Disappointingly, neither diet affected HDL levels.

The strengths of this study are its cross-over design and the high adherence rate (85% full adherence). Still, there were several limitations. The number of subjects is small, although relatively large for a diet study. The duration was short, but long enough to show significant differences in some measures. There were no data on sugar intake, salt intake, or blood pressure levels. Only the lacto-ovo VD was tested. Although low in saturated fat, lacto-ovo is not as low in total fat as other similar diets. Since the lacto-ovo VD is low in vitamin B12, the authors recommend testing for B vitamin levels in subjects on this and other vegetarian diets, with supplements provided as needed.

In an editorial that accompanied this study, Anderson summed up the requirements quite well, noting that a healthy diet should be rich in fruits, vegetables, whole grains, legumes, and nuts. Diets also should be low in refined grains, processed food, saturated fat, sugar, and salt. A diet should be sustainable, culturally relevant, and enjoyable. I would add that it should be low enough in calories to keep one’s BMI < 25 kg/m².