Solà R, Valls RM, Godàs G, et al. Cocoa, hazelnuts, sterols and soluble fiber cream reduces lipids and inflammation biomarkers in hypertensive patients: a randomized controlled trial. PLoS One. 2012;7(2):e31103. doi:10.1371/journal.pone.0031103.

Markers that help to determine cardiovascular disease (CVD), including low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, and blood pressure (BP), are known to be influenced by food and components therein, such as plant sterols, fiber, flavanols, nuts, and cocoa (Theobroma cacao). As the effect of the combination of these foods and compounds on CVD risk is unknown, this randomized, double-blind, controlled, parallel, multicenter trial investigated the impact of cocoa alone, alongside cocoa mixtures containing hazelnuts (Corylus avellana), phytosterols, and fiber, on metabolic parameters associated with CVD risk.

This study took place at the Hospital Universitari Sant Joan de Reus in Reus, Spain and primary care centers in Alcover, Vic, and Centelles, Spain. Included subjects were over 20 years of age and had either prehypertension as determined by a systolic BP of 120-139 mmHg or a diastolic BP of 80-89 mmHg, or hypertension as characterized by a systolic BP of 140-159 mmHg or a diastolic BP of 90-99 mmHg. Other inclusion criteria were LDL cholesterol concentrations of 130-189 mg/dL and 1 of the following CVD risk factors: age (men >45 years or women >55 years); smoking; low HDL cholesterol in men (<40 mg/dL) or women (<46 mg/dL); or early onset of CVD in family history. Subjects were excluded if they suffered from diabetes or other chronic diseases, were on medication for hypolipidemia, had fasting triglyceride levels >350 mg/dL, or a body mass index (BMI) of >35 kg/m².

All included subjects in the study acclimated to the cocoa treatment (control product A) for a 2-week period prior to the study using a special diet; during the study, subjects were placed on an isocaloric diet consisting of 35% fat (<7% saturated fatty acids [SFAs]), 50% carbohydrates, 15% protein, and <200 mg/day cholesterol. To discourage SFA intake, subjects were instructed not to consume certain meats and dairy products and to avoid chocolate, nuts, and soy (Glycine max) products.

Subjects were randomized into groups consuming 6 portions per day of either control product A (1 g of cocoa solids), product B (cocoa and 5 g of hazelnuts), product C (cocoa, 30 g/day of hazelnuts, and 2 g/day of phytosterols), or product D (cocoa, 30 g/day of hazelnuts, 2 g/day of phytosterols, and 20 g/day of fiber). Product D is a patented product called "LMN" in the study. The dosage of hazelnuts and fiber were based on previous studies of their positive effects on CVD. All products were manufactured for the study by La Morella Nuts S.A.; Castellvell del Camp, Spain. Uneaten products, as well as empty wrappers, were monitored for compliance; non-compliance was considered to be <80%.

Primary outcomes were BP, LDL cholesterol, apolipoprotein B-100 (ApoB), the ratio of ApoB to apolipoprotein A (ApoA), oxidized LDL (oxLDL), and high-sensitivity C-reactive protein (hsCRP). Baseline measurements were taken after the 2-week acclimation period, and endpoint assessments were done after the 4-week treatment. Weight was also taken every 2 weeks, and diets were adjusted to ensure uniform weight throughout the study. In addition, waist circumference and endothelial function were assessed.

At baseline, with the exception of certain characteristics such as age, instance of high BP, and family history of CVD, no differences between the groups were observed. Of a total of 113 subjects randomized to treatments, non-compliance resulted in 11 excluded subjects; this left a per-protocol group of n=25 consuming product A, n=26 for product B, n=26 for product C, and n=25 for product D. In terms of compliance, there was 98% adherence to the diet and 94% adherence to the treatments.

There were no changes in the body weight or waist circumference of the subjects at the end of the study. Those taking product A had both a decreased systolic BP (-7.89 mmHg) and diastolic BP (-5.54 mmHg) as compared to baseline measurements. It is not mentioned whether this is significant. Also, the diastolic BP of subjects taking product B was significantly different from those consuming product A (P=0.0357). At the end of the study, subjects taking product C had lowered LDL cholesterol concentrations by -11.2%, significantly different from those taking product A (P=0.0002). Also at the end of the study, the ApoB/ApoA ratio of the product C group was decreased by -7.8% and was significantly different from those taking product A (P=0.0085). Those consuming product D also reduced LDL cholesterol by -9.2% and ApoB/ApoA ratio by -10.5%, both significantly different from subjects consuming product A (P=0.0018 and P=0.0005, respectively).

Subjects taking product D showed reduced hsCRP by 0.96 mg/L and oxLDL by 4.0 U/L; these were significantly different from control product A (P<0.0083 and P=0.0252, respectively). In n=14 of subjects in the control product A group, endothelial function was assessed. From baseline to 2 weeks into the study, vasodilator response increased from 1.800 to 2.173 AU (95% confidence interval [CI]: -0.044 to 0.702; P=0.029). Adverse side effects of a bloated feeling were mentioned by 2 subjects and low appetite by 1 subject.

Because the LDL cholesterol-lowering effects seen in this study were comparable to those seen in studies using just sterols, it is suggested that the effects are due mainly to the sterols in the combination treatment. The effects on oxLDL were comparable with those seen in other studies for fiber; this may be due to its antioxidant effect. In regards to the lowering of BP in the product A group, this was an unexpected result and, as this treatment was used as a "control," BP was not compared between the groups. It is stated that the use of product A as a control to compare the bioactivity and responses of other groups is an inappropriate study design as this product had bioactivity. Lastly, the improvement of cocoa on the vascular endothelial function is an intriguing result that deserved more attention in this study. This potential mechanism of action in alleviating CVD risk will ideally be further explored in future studies.