EXPLORING SOYMILK AND CARDIOMETABOLIC HEALTH: A SYSTEMATIC REVIEW AND META- ANALYSIS

Amid growing concerns about sustainability and health, dietary guidelines consistently advocate for a transition towards plant-based diets.1-8 The market for dairy, meat, and egg alternatives has expanded rapidly in response, with plant-based foods projected to reach almost 10% of the global protein market by 2030.9 Though these foods may aid in the transition to and maintenance of plant-based diets, systems like the World Health Organization-endorsed Nova food classification system labels these alternatives as ultra-processed and suggest limiting consumption.10

This contradiction is exemplified by fortified soymilk, which is recognized by the Dietary Guidelines for Americans (2020-2025) as the sole nutritionally comparable plant-based alternative to cow’s milk, and yet is classified as an ultra-processed food.1 Furthermore, the common practice of adding sugars to soymilk, intended to match the natural sweetness of cow's milk (Nova-classified as minimally processed), would disqualify it from meeting the Food and Drug Administration’s proposed definition of a “healthy” food.11 To address the contradictory messaging and explore the health effects of soymilk, a recent analysis evaluated the cardiometabolic impacts of soymilk compared to cow’s milk.

Effects of Soymilk and Cow’s Milk in Cardiometabolic Health

A University of Toronto research team has undertaken a comprehensive systematic review and meta- analysis of randomized controlled trials that compare the effects of soymilk with cow's milk on cardiometabolic health in adults. This evidence synthesis aims to explore the specific contributions of sweetened and unsweetened soymilk to health outcomes, including lipids, glycemic control, blood pressure, inflammation, adiposity, and renal and hepatic function in adults of any health status over a duration of ≥3 weeks. The protocol was registered at ClinicalTrials.gov (identifier, NCT05637866).

Preliminary data restricted to the effect of substituting sweetened soymilk for cow’s milk was presented at the 2023 American Society for Nutrition Conference (NUTRITION 2023, Boston, MA, USA). The search identified 10 trials with a total of 277 adult participants. Available outcome data included LDL cholesterol (6 trials), HDL cholesterol (4 trials), non-HDL cholesterol (3 trials), triglycerides (4 trials), HbA1c (1 trial), fasting plasma glucose (3 trials), 2-hour plasma glucose (1 trial), fasting insulin (2 trials), systolic blood pressure (3 trials), diastolic blood pressure (3 trials), C-reactive protein (1 trial), body weight (3 trials), body mass index (BMI) (3 trials), waist circumference (2 trials), alanine aminotransferase (ALT) (1 trial), and aspartate aminotransferase (AST) (1 trial).

Participants consumed a median of 602mL of sweetened soymilk daily, which contained 24.5g protein and 19g added sugars (7g/250 mL). In comparison, the median daily consumption of cow’s milk was 538 mL, with 24.5g protein and 26g lactose (12g/250 mL). In these trials, replacing cow's milk with sweetened soymilk led to moderate reductions in LDL cholesterol (MD, -0.22 mmol/L [95% CI, -0.34 to - 0.09 mmol/L]) and large reductions in systolic (-7.51 mmHg [-10.26 to -4.76 mmHg]) and diastolic (-4.08 mmHg [-6.59 to -1.58 mmHg]) blood pressure. No differences in HDL-C, non-HDL-C, triglycerides, HbA1c, fasting plasma glucose, 2-hour plasma glucose, fasting insulin, C-reactive protein, body weight, BMI, waist circumference, AST, or ALT were seen. The 8% reduction in LDL-cholesterol and 7.5 mmHg reduction in systolic blood pressure is comparable to the reductions seen with drugs used as add-on therapy to statins for cholesterol lowering (e.g., bile acid sequestrants) and first line drugs used for blood pressure lowering at submaximal doses (e.g., ACE inhibitors [angiotensin-converting enzyme inhibitors]; ARBs [angiotensin receptor blockers], and CCBs [calcium channel blockers]), respectively.12,13

The certainty of the evidence was rated as moderate for the reduction in LDL-cholesterol, moderate and low for the reductions in systolic and diastolic blood pressure, respectively, and generally moderate for the effects across all other outcomes using the established Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach.14,15 There were no trials that examined apolipoprotein B, plasma glucose area under the curve, body fat, creatinine, estimated glomerular filtration rate, creatinine clearance, albuminuria, albumin-to-creatinine ratio, uric acid, intrahepatocellular lipid, or fatty liver index, highlighting an area for future research.