The effect of an apple polyphenol extract rich in epicatechin and flavan-3-ol oligomers on brachial artery flow-mediated vasodilatory function in volunteers with elevated blood pressure

The study recruited 60 otherwise healthy male and female volunteer subjects, aged 40-65 years, with borderline hypertension or mild hypertension. A study flow diagram is shown in Fig. 1.

The inclusion criteria of the study called for the presence of borderline hypertension (BP 130-139/85-89 mmHg) [14] or mild hypertension (BP 140-165/90-95 mmHg) [15] at the screening visit (mean BP value of two measurements), otherwise good general health as ascertained by medical history, clinical laboratory assessments and physical examination, and willingness to comply with the study procedures. Written informed consent was obtained from all the subjects prior to study participation. Exclusion criteria consisted of body mass index (BMI) >32 kg/m², total serum cholesterol ≥8 mmol/l, any abnormal safety laboratory parameter or abnormal finding in the electrocardiogram (ECG) evaluated to be clinically significant, coronary artery disease, pregnancy or lactation, alcohol abuse as evaluated by medical history, laboratory determinations and an alcohol breath test, regular smoking, diabetes mellitus, apple allergy, use of lipid-lowering medication, regular use of any medication known or believed to affect endothelial function or blood vessel tone, high consumption of products containing flavonoids, habitual use of vitamin products containing single vitamins in strengths beyond the recommended daily intake, habitual use of herbal remedies, participation in any clinical trial during the previous 2 months and any other condition or medication that in the opinion of the investigator would interfere with the evaluation of the study results.

The subjects were randomized in a cross-over design to receive placebo or the apple polyphenol extract rich in epicatechin and flavan-3-ol oligomers for 4 weeks (“Treatment Period 1”), followed by a four to five-week wash-out period and finally 4 weeks of treatment with the product that they did not receive during the first Treatment Period of the study (“Treatment Period 2”). There were altogether five scheduled visits to the study site - one at screening (“Screening Visit”), and two visits in each Treatment Period. In the Treatment Periods 1 and 2 first visit was conducted when the investigational product intake was started (“Visit 1” in Treatment Period 1 and “Visit 3” in Treatment Period 2) and the second visit when the last dose was to be taken (“Visit 2” and “Visit 4”). For practical reasons, it was allowed that each subject could take the investigational product for 26-30 days in each Treatment Period. The study design is shown in Fig. 2.

Prior to all visits 10 h of fasting was required. Use of alcohol and strenuous physical exercise were forbidden for 24 h prior to all visits. Cigarette smoking or use of other nicotine products was not allowed for 10 h prior to each visit and during the visits. Otherwise the subjects were to follow their habitual diet and lifestyle during the entire study.

A written informed consent was obtained. The subjects were interviewed for collection of demographic data (date of birth, age, race, sex, special diets) and information on substance use (use of nicotine, alcohol and caffeine, and drug abuse), use of epicatechin -containing foods and drinks and participation in other clinical trials. Body weight, height and BMI were recorded, blood samples were collected for haematology, clinical chemistry and serology after fasting for at least 10 h, a urine sample was collected for drug abuse and pregnancy tests, an alcohol breath test was performed, a standard physical examination was performed, medical history and current medical conditions were recorded, use of medications (including prescription and over-the-counter medications, dietary supplements, herbal products, micronutrients and vitamins) within two weeks prior to the screening visit was recorded, a 12-lead ECG was recorded in the supine position after 10 min rest, systolic and diastolic BP and HR were recorded twice from the right arm with 1-2 min interval in the sitting position after 5 min rest. Diet diaries with instructions were given to the subjects with an instruction to fill the diary on four successive days, after the Screening Visit and at the end of both Treatment Periods. A subject was invited to the Visit 1 if all of the inclusion criteria and none of the exclusion criteria were met. The study events in five different visits are shown in Table 1.

The apple polyphenol extract rich in epicatechin and flavan-3-ol oligomers had the following composition: high content of polyphenols (around 90% total polyphenols as catechin equivalents), specifically at minimum 30% (−)-epicatechin and at least 20% flavan-3-ol oligomers (degree of polymerisation (dp)2-dp7) and minor amounts of oligomers with dp > 7, other catechins and polyphenols. The apple extract was provided by Coressence Ltd. The source apples (Evesse™ Apples) for the extract have the botanical classification Malus pumila MILLER, more generally known as Malus domestica or the domestic modern apple. Evesse™ Apples are selected from the Herefordshire Pomona in the UK, a nineteenth Century catalogue of variety type / names of apples and pears known to have been grown in Herefordshire [16]. The apples are harvested at maturity in a period between August and November each year and either freeze dried and powdered or stored in controlled atmosphere and/or coldstores at chilled conditions (2-5 °C) until processed. Whole fresh apples are de-waxed by an ethanol wash then combined with a proportion of freeze dried apple granules and hot water. While continually stirring, the extraction mixture is rapidly cooled and enzymes are added to assist in the liberation of flavan-3-ols from the apple granule matrix. Denaturing of the enzymes by heating and filtration yields a sugar-rich syrup. Residual carbohydrate polymers are further degraded with enzymes before the sugar-rich intermediate extract is passed over an ion-exchange column to remove sugars. The apple polyphenol extract rich in epicatechin is precipitated using food-grade ethanol and dried. The apple extract is stable at ambient temperatures for at least 18 months from date of manufacturing. Epicatechin and flavan-3-ol content was determined by Reversed Phase HPLC.

In a study by Schroeter et al. [3] vascular response measured by the flow mediated dilation and pulse amplitude tonometry was improved with oral administration of 1 or 2 mg/kg bw of epicatechin in a small proof-of-concept study with healthy volunteers. Similarly, significant increases in vascular response were reported after oral administration of 1.5. mk/kg bw of epicatechin [17]. These studies provide rationale for a target dose of dietary supplement delivering 1-2 mg/kg bw epicatechin to improve vascular function. Thus, in this study the daily dose of 100 mg epicatechin was administrated.

The study product was given orally as a single dose of 330 mg/day, containing 100 mg of epicatechin, and encapsulated in hard gelatine capsule. The study subjects were instructed to take the study product preferably in the morning, with breakfast, approximately at 8 a.m. (between 6 and 10 a.m.). The first and the last study product dose of each treatment period were administered at the study site after a standard breakfast.

The placebo capsules had similar appearance as verum/epicatechin capsules but contained 335 mg Microcrystalline cellulose (MCC) (Tabulose Blanver, Florida, USA). Both placebo and verum capsules contained 3 mg Mg-Stearat. Epicatechin capsules were balanced with placebo capsules by adding 50 mg MCC.

Endothelium-dependent and -independent vasodilatation of the left brachial artery was monitored with ultrasonography (Acuson Sequoia 512 mainframe instrument; Acuson, Mountain View, CA, USA) with a 13.0-MHz linear-array transducer. The method was based on the measurement of brachial arterial diameter at baseline and after a vasodilatory stimulus.

Brachial artery scans were obtained at rest and during reactive hyperaemia to study endothelium-dependent FMD. The luminal diameter of the artery was first measured at rest at a fixed distance from an anatomical marker at end-diastole, incident with the R wave on the ECG (3 measurements; the mean was used in the analyses). The artery was then occluded by inflating a blood pressure cuff placed around the forearm to a pressure of 250 mmHg. After 4.5 min, the cuff was deflated, causing increased blood flow in the artery, i.e. reactive hyperaemia. The arterial diameter was now recorded at 40, 60 and 80 s after cuff release, and dilatation of the artery from baseline was measured offline at end-diastole, incident with the R wave. From these data, the maximal FMD response was calculated as per cent increase of the arterial diameter from the baseline [18].

The pre and post FMD tests were performed on all Visits 1 to 4. The pre FMD test was performed at baseline, prior to study treatment administration, and the post FMD test was repeated 1.5 h after the administration. The timing of the second FMD measurement was based on a recent human bioavailability study showing that the time for maximum concentration (T_max) of the active moiety (epicatechin and its metabolites: 3’methylepicatechin, 4’methylepicatechin, epicatechin sulfates and methyl-epicatechin sulfates) observed for a drink fortified with the same apple polyphenol extract rich in epicatechin and flavan-3-ol oligomers was at 1.0 -1.5 h [19].

Long-term changes in endothelial function were evaluated by comparing the pre FMD test values measured at the beginning and end of each Treatment Period. Acute changes in endothelial function were evaluated by comparing the pre and post FMD test values measured on the first visit of each treatment period (Visits 1 and 3). Additionally, acute changes in endothelial function were assessed by comparing the pre and post FMD test values measured on the last visit of each Treatment Period (visits 2 and 4), and taking into account the possible long-term change.

The NMD test was performed on all Visits (1-4), after the post FMD test. A new baseline measurement of the luminal diameter was performed 10 min after the postFMD test, and a dose of 1.25 mg of isosorbide dinitrate (Dinit, Leiras, Finland, 1.25 mg/dose; 1 spray) was administered sublingually, and the arterial diameter was measured 4 min later. From this measurement, the NMD% (per cent increase of the arterial diameter from the baseline) was calculated. Possible changes in vascular function were evaluated by comparing the two NMD values measured at the start and end of each Treatment Period and the two NMD values measured at the end of each Treatment Period.

Changes in systolic and diastolic BP values between the visits were also evaluated.

Serum and plasma samples were collected for analysis of markers of inflammation, adhesion molecules and coagulation markers. These included soluble E-selectin (sE-selectin), soluble vascular cellular adhesion molecule-1 (sVCAM-1), soluble intercellular adhesion molecule-1 (sICAM-1), von Willebrand factor (vWF), plasminogen activator inhibitor-1 (PAI-1), asymmetric dimethylarginine (ADMA) and C-reactive protein (CRP) (sensitive assay). Concentrations of sE-selectin and ADMA were determined from serum samples, whereas the other markers were determined from plasma samples. The biomarkers were analyzed with commercial enzyme linked immunosorbent assay (ELISA) kits according to instructions provided by the manufacturers: ADMA (DLD Diagnostika Gmbh, Hamburg, Germany), vWf (RayBiotech, Norcross, GA, USA), sVCAM, sICAM-1, PAI-1, sE-selectin, CRP (Aushon Biosystem, Billerica, MA, USA). Each individual subject’s samples were analyzed on the same ELISA plate to reduce analytical variation. Biomarker samples were collected at the same time points as those collected for the determination of epicatechin concentrations, prior to and 2 h after dosing on each Visit.

Concentrations of epicatechin and its metabolites were analysed by IFR Extra Ltd. (www.​ifrextra.​co.​uk) using liquid chromatography-tandem mass spectrometry, as described previously [20].

At the screening visit, blood samples were collected for haematology, clinical chemistry and serology. Urine samples were also collected for drug abuse and pregnancy tests.

Systolic and diastolic BP and HR were recorded at all Visits. At the last visits of both Treatment Periods (Visits 2 and 4), blood samples were collected for safety determinations. Any clinically significant abnormal laboratory or other safety values were followed up.

Laboratory safety variables measured included haemoglobin, haematocrit, reticulocytes, mean corpuscular volume, mean corpuscular haemoglobin, leucocytes, differential blood count (neutrophils, lymphocytes, monocytes, eosinophils, basophils), plasma albumin, plasma bilirubin, plasma alanine aminotransferase, plasma aspartate aminotransferase, fasting plasma creatinine, plasma glutamyl transferase, plasma sodium, plasma potassium, fasting plasma calcium, fasting plasma urea, plasma uric acid, fasting plasma total cholesterol, fasting plasma HDL-cholesterol, fasting plasma LDL-cholesterol (calculated), fasting plasma triacylglycerols, fasting plasma glucose and plasma CRP.

Adverse events registered during the treatment periods were isolated occurences of nasopharyngitis, dyspepsia, headache, migraine, dizziness, ligament sprain and gammaglutamyltransferase increase, with no obvious association to either treatment.

Diet diary data were collected for the evaluation of the possible effects of diet on the study results. Dietary intakes were analysed three times using four-day food intake records; once after the Screening Visit and once during both Treatment Periods to estimate possible changes in the subject’s diet during the course of the interventions. A nutritionist checked the food records for accuracy at each occasion. The nutrient intakes were calculated with Micro-Nutrica software based on the Food and Nutrient Database of the Social Insurance Institution, Finland. The programme has been continuously updated with new foods and recipes in conjunction with the STRIP study at the Research Centre of Applied and Preventive Cardiovascular Medicine, University of Turku [21].

Based on the results of previous studies [5, 22] a mean (SD) difference between the treatments in FMD% change from baseline of 1.0% (SD 2.5) was assumed. With 80% power (β = 0.20) and two-sided type I error risk of 5% (α = 0.05), 26 subjects per sequence was calculated to be needed. Assuming a 15% lost-to-follow-up rate, a total of 60 subjects were needed to be recruited for the study, i.e. 30 subjects per the cross-over treatment sequence.

The primary efficacy variable was the change from baseline in FMD%, including both the long-term and the acute response. Subjects who completed the both Treatment Periods and had a full set of FMD measurements were used in the evaluation of the primary hypothesis.

The primary efficacy variable (maximal FMD% response) was analyzed with two separate statistical models; one assessed the long-term change in the FMD response and one assessed the acute change. An analysis of covariance (ANCOVA) model was used to analyze the long-term effect based on the changes in FMD%. The model included baseline FMD% as covariate, and the main effects of cross-over sequence, Treatment Period and treatment as fixed effects and subject as random effect. The difference in the change of FMD% (apple extract - placebo) at end-of-period and a two-sided 95% confidence interval (CI) for the difference were estimated from the ANCOVA model using contrasts. Estimates of within-group changes were provided, as well.

The acute effects were analyzed using the changes in FMD% between the measurements on the first visits of both treatment periods (Visits 1 and 3). A similar model was applied in the statistical analysis as for the long-term effects. The acute changes were also characterized by comparing the changes in FMD% between the measurements on the last visits of both treatment periods (Visits 2 and 4).

In addition, as no long-term effects were detected, a Wallenstein-Fisher model was used to compare the acute changes in FMD% using both the first and the last visits of each period. This model included main effects of treatment, sequence, Treatment Period and Visit (within the Treatment period) as well as sequence by visit, period by visit and treatment by visit interaction effects as fixed factors. The secondary efficacy variables were the brachial NMD% test result, systolic and diastolic BP and the seven investigated biomarkers (sICAM-1, sVCAM-1, PAI-1, CRP, ADMA, vWf and sE-selectin). The analyses of the secondary variables were based on data from all subjects who completed both Treatment Periods and had full sets of measurements available.

A similar Wallenstein-Fisher model was applied for the NMD% as in the primary analysis. Similar ANCOVA models were used for systolic and diastolic BP as for the long-term effect in the primary FMD% analysis. For the biomarkers, both long-term and acute effects were evaluated. Long-term effects were evaluated with similar ANCOVA models as the FMD% and acute effects with similar Wallenstein-Fisher models as the FMD%. A natural logarithmic transformation was used for the following biomarkers: vWf, E-selectin, CRP, sVCAM-1 and PAI-1. This was done to normalize the distributions to be able to use analysis methods assuming normal distributions. The analysis results for these variables are presented in the log-scale.

All statistical analyses were performed and tables, figures and subject data listings were prepared with SAS software version 9.3 (SAS Institute Inc., Cary, NC, USA).