Get Involved
About Us
Our Members

Lemon Balm May Improve Cardio-metabolic Outcomes by Reducing Total Cholesterol and Systolic Blood Pressure: Systematic Review and Meta-analysis

Date 10-29-2021
HC# 032142-675
Lemon Balm (Melissa officinalis, Lamiaceae)
Cardio-metabolic Conditions
Systematic Review/Meta-analysis

Heshmati J, Morvaridzadeh M, Sepidarkish M, et al. Effects of Melissa officinalis (Lemon Balm) on cardio-metabolic outcomes: A systematic review and meta-analysis. Phytother Res. December 2020;34(12):3113-3123. doi: 10.1002/ptr.6744.

Cardiometabolic diseases, including type 2 diabetes (T2D), hypertension, and coronary artery disease, are common chronic diseases and major contributors to worldwide mortality. Risk factors include insulin resistance, obesity, oxidative stress, elevated blood pressure, inflammatory markers, total cholesterol (TC), triglycerides (TGs), and low-density lipoprotein (LDL), and decreased high-density lipoprotein (HDL). Many herbs are used as complementary and alternative medicine in cardiometabolic disorders.

Among them, lemon balm (LB; Melissa officinalis, Lamiaceae) has reduced serum lipids and lipid peroxidation, increased antioxidant defenses, and reduced liver enzyme levels comparable to statins in vitro and in vivo. Rich in phenolic acids, triterpenes, and flavonols, LB is also used widely in traditional Asian medicine for gastrointestinal diseases, rheumatoid arthritis, neurological disorders, and other conditions. Rosmarinic acid, LB's main constituent, has demonstrated vasorelaxant endothelial effects in experimental studies, and its citral components are associated with anti-inflammatory effects. LB has also shown hypoglycemic properties; however, the doses used are unsuited for humans. This is the first systematic review (SR) and meta-analysis (MA) of randomized, double-blind, controlled clinical trials (RCTs) of LB's cardiometabolic effects.

The authors searched electronic databases, including those of ongoing clinical trials, from inception through November 2019, as well as reference lists and related SRs, for English language LB RCTs in healthy or diseased individuals aged ≥ 18 years. Of 837 records identified, 30 were duplicates. Screening titles and abstracts eliminated 793, leaving 14 reports to be assessed in full text. Seven were excluded for no relevant variables (n = 4), not a RCT (n = 2), and combination with other treatments (n = 1), leaving seven for the SR and MA.

Table 2 in the article highlights the main characteristics of included studies but shows data from only six. Two RCTs by one research group are either conflated, with one set of data shown, or one has been omitted. The qualitative SR does not acknowledge this discrepancy. Studies selected were published from 2016-2019. Number of participants ranged from 26-72. While Table 2 tallies to 324 participants/patients in all, the discussion section mentions a total of 250.

One study with healthy, active participants was conducted in Japan; the remaining six, with T2D (n = 3 studies), hyperlipidemia (n = 1), or chronic stable angina (n = 2) patients, in Iran. LB doses ranged from 1000-3000 mg/d; duration of intervention, 6-12 weeks. Mean age of participants/patients was 43.4-58.8 years. All RCTs included both genders. Orally administered LB preparations included hydroalcoholic extract (n = 1 study), encapsulated leaf powder (n = 1), and aerial parts (n = 4).*

Standard mean difference (SMD) was used as effect size. Random effects models estimated pooled SMDs. Meta-regression analyses assessed relationships between SMDs and potential variables. A sensitivity analysis (leave-one-out method) rated the effect of each RCT on pooled SMDs. Outcomes linked with cardiometabolic risks included glycemic parameters (n = 4 studies), serum lipids (n = 6), BP (n = 3), and inflammatory markers (n = 2). No statistically significant effects were reported for fasting blood glucose, glycated hemoglobin, or insulin levels. LB was associated with a significant decrease in TC. Subgroup analysis found this effect strongest in RCTs lasting ≥ eight weeks and those using doses ≥ 1500 mg/d. There were no significant effects on TGs, LDL, or HDL. Subgroup analysis suggested stronger effects on LDL in patients vs. healthy participants and at doses ≥ 1500 mg/d, with a trend to increased HDL and reduced TGs. There was a statistically significant effect on systolic but not diastolic BP, with a trend to significant reductions in the latter. Levels of high-sensitivity C-reactive protein levels showed no statistically significant effects, but there was a trend towards significant reduction. No serious adverse effects occurred in any RCT.

Quality of RCTs was assessed using the Cochrane risk of bias tool. Most had moderate-to-serious risks of incomplete outcome data. "Remarkable" heterogeneity in doses used, study duration, sample size, and participant ages, assessed via the I2 index, could not be entirely resolved through random-effects modeling. The limited number of RCTs, small study groups, high risks of bias, and substantial heterogeneity imply that results be interpreted cautiously. The authors note that different LB preparations and modes of administration may produce varying levels of bioactive plant compounds. Several relevant outcomes, including obesity, were not reported by these authors. More high-quality RCTs using well-characterized LB formulations and relevant outcomes are needed.

Mariann Garner-Wizard

*Note that just six studies are represented here due to the unexplained omission mentioned.