Participants – ethics statement
A total of 161 male participants were recruited for the study. All had moderate physical activities (2–6 hours per week). None were engaged in any physical activity aimed at increasing muscle strength and mass for the six months before the experiment. All were healthy and free of injury during the three months preceding the study. The study excluded subjects who had asthma with potentially steroids treatment, consumption of high-protein diet, steroids treatment, current consumption of drugs or during the previous month, consumption of dietary supplement, sports drink, special dietary food or functional food, of any kind, liable or presented as liable to enhance physical performances and especially to increase muscle mass. Moreover, subjects with known hypersensitivity to any of the constituents of the studied products were excluded. Throughout the study, subjects maintained their usual training routine and diet. All gave their written consent after being carefully informed about the experimental protocol. The study was conducted in accordance with the Helsinki Declaration without any deviation from the protocol approved by the East I ethics committee (East I committee, France, number: 2011–47, 9 November 2011). The authors confirm that all ongoing and related trials for this intervention are registered before at the French agency for the safety of health products (AFSSAPS number: 2011-A01211-40) and at ClinicalTrials.gov (NCT02128516).
After inclusion, participants were randomly divided into three experimental groups: Pea (n = 53), Whey (n = 54), and Placebo (n = 54). Table 1 sets out their characteristics. Product randomization was balanced by block sizes of 10. The randomization code was not made available to anyone involved in conducting or evaluating the study and was released after the blind review and the freezing of the final database. The sample size was calculated a priori using Nquery Advisor software (version 6.01) based on the primary criterion (muscle thickness) and allowing for a power >90%. This statistical analysis indicated a minimum of 34 participants per experimental group.
The objectives of this randomized, double-blind study conducted with parallel arms, was to evaluate the effects of oral Pea protein supplementation, versus Placebo and versus Whey proteins associated with a 12-week resistance training program, on elbow flexors muscle thickness (main outcome) and muscle strength (secondary outcome). The trial consisted in an inclusion visit (D0), an intermediate 6-week follow-up visit (D42) and a final 12-week visit (D84) (Figure 1). Visits were separated by weight training periods with three sessions per week. Testing sessions were conducted on non-training days and always at the same time of the day for the same subject. D0, D42 and D84 sessions included measurements of (i) muscle thickness using ultrasonography, (ii) arm circumference and (iii) maximal muscle strength in isokinetic conditions (concentric, isometric and eccentric). After the initial evaluation (D0), each subject was given a batch of products, according to randomization, and began weight training for a 12-week period. The same tests were repeated, in the same order, half-way through and at the end of the training period (D42 and D84, respectively). Tolerance, collected from adverse events and compliance with product intake (determined by counting products not consumed) was evaluated at D42 and D84 too.
All subjects followed the same training routine, three times per week with a rest day between each session. Training was based on three exercises involving the elbow flexor and extensor muscles. The exercises soliciting the flexor muscles were arm curl and lateral pull-down. In the arm curl exercise, subjects sat with weights in their hands with a ~40° trunk/arm angle. They had to flex/extend the forearm over the arm. For the lateral pull-down, subjects sat with a bar in their hands above the head. They had to flex/extend the forearm over the arm with a vertical movement. The exercise soliciting the extensors was the bench press. Subjects were lying on their backs with a bar in their hands with a 90° trunk/arm angle, arms extended, and had to flex and extend their upper limbs vertically. Throughout the training program, the number of sets was progressively increased from 2 to 5 while the number of repetitions was reduced in parallel from 15 to 5 repetitions maximum (RM). The final week, training was composed of three sets of 5 RM in order to preclude any fatigue for the D84 tests. Recovery between sets was 2–3 minutes. The load used for each exercise was regularly adapted during training depending on individuals’ maximum load (1-RM, one maximum repetition, evaluated every two weeks).
The three products under study were presented as 45 g sachets of banana-flavored cocoa powder to be diluted in 300 mL of cold water at each intake. The diluted drinks were identical in appearance, texture and taste and were isoenergetic (Table 2). Products were taken twice a day for 12 weeks. On training days, one dose was taken in the morning and the second just after training. On non-training days, one dose was taken in the morning and the second dose in the afternoon. The general food intake was not monitored over the experimental procedure but participants were instructed to maintain their diet habits throughout the experimental protocol.
A dose of powder contained either 25 g of vegetable Pea protein isolate (NUTRALYS®) or 25 g of Whey protein concentrate. The placebo, with no added protein, was composed of maltodextrin. The nutritional composition of each product and the amino acids content of Pea and Whey proteins are shown in Tables 2 and 3, respectively. The other components (fat-reduced cocoa, flavouring, aspartame, salt, silica dioxide) were identical in nature and in quantity in all three products. NUTRALYS® pea protein (ROQUETTE, Lestrem, France) is a vegetable protein isolate from the yellow pea (Pisum sativum). Peas are cleaned and ground in a dry process to produce pea flour. Flour is then hydrated and the pea starch and internal fiber are extracted separately. The protein fraction is then coagulated for further purification and, finally, carefully dried in a multi-stage spray dryer. The resulting highly purified pea protein isolate contains 85% protein, 7% fat, 3% carbohydrate, and 5% ash on a dry matter basis.
Biceps brachii muscle thickness
Right-side biceps brachii muscle thickness was measured in real time using an ultrasound machine (AU5; Esaote Biomedica, Florence, Italy) coupled to a 50 mm probe at a frequency of 7.5 MHz. Subjects were lying supine with arms and legs completely relaxed. The right upper limb was positioned supine with a 45° angle with respect to the trunk. The probe was placed perpendicular to the skin surface at two-thirds of the distance between the acromion process of the scapula and the lateral epicondyle of the humerus . The probe was coated with a water-soluble transmission gel to provide acoustic contact without depressing the dermal surface. Thickness was calculated as the distance between superficial and deep aponeuroses measured at the ends and middle of each 3.8 cm-wide sonograph. Three images were independently obtained for each point. The average value of these nine measures was calculated. To favor reproducibility, probe placement was carefully noted for reproduction during the other test sessions and measurements were always performed by the same operator.
The circumference of the right arm was measured using a constant tension tape during maximal elbow extension at rest and during a maximal voluntary contraction (with maximal elbow flexion). Three measurements were made (at rest and contracted) along the length of the biceps, namely ¼, ½ and ¾ of the length of the upper arm (distance between the acromion process of the scapula and the lateral epicondyle of the humerus). Averaging was performed to obtain mean values for the circumference at rest and contracted.
Maximal voluntary torque
The maximal voluntary torque was measured on a Biodex (Biodex, Shirley, USA) isokinetic dynamometer during isometric, concentric and eccentric elbow flexions. The right-hand side was tested. Subjects were seated upright with a 95° hip angle. The upper limb was placed horizontally with the elbow rotation axis coinciding with the axis of rotation of the ergometer and aligned with the shoulder axis. The chest, shoulder and forearm were firmly attached to avoid perturbing contributions. Movements were made in the horizontal plan through a 120° elbow range of motion (from 10 to 130°, 0° = full extension). After a standardized warm-up consisting of submaximal contractions, measurements were made in concentric and eccentric conditions at an angular velocity of 60°.s−1. Subject had to accelerate or resist the ergometer lever arm, respectively. Five maximal voluntary contractions were performed consecutively for each condition. In isometric condition, the position was set at 80° elbow flexion and the subject had to produce a maximal voluntary contraction lasting 5 s. Three isometric contractions were requested with 60 s rest between contractions. Isometric, concentric and eccentric solicitations were presented in a random order and separated by five minutes of passive recovery. These various parameters were recorded for further analysis. The maximum value for each condition was retained for the statistical analysis.
Arm curl 1-RM
The maximum load (1-RM in kg) that could be lifted during elbow flexions was measured during an arm curl movement performed with both arms. For this, the load was progressively increased through successive sets (the first set being considered as warm up). Then, subjects were requested to lift each load only once. Care was taken to lift the load with the largest range of motion (~100 °). One minute rest was permitted between trials. In case of failure, a second try was allowed. The maximal load lifted was considered as the 1-RM. It was regularly evaluated (every two weeks) in order to adjust resistance training intensity.
Twenty four subjects left the study early due to personal reasons. At the end of the experimental procedure, 137 subjects were considered for analysis with 47 in the Pea group, 46 in the Whey group and 44 in the Placebo group (see Figure 2 for the CONSORT Diagram). Quantitative variables were presented as mean values and standard deviations. Values were tested using a repeated measures analysis of variances (ANOVA). Groups were used as independent variables and time (D0, D42, and D84) was used as the repeated variable. A sensitivity analysis was also conducted and considered subjects with a 1-RM at inclusion <25 kg (median value of study sample). Sixty eight subjects were considered for this sensitivity analysis. In the case of significant main effects or interactions, Scheffé post-hoc tests were conducted. Qualitative variables (supplementation compliance or adverse effects) were presented as absolute and relative frequencies and were tested by using a Chi square test. Statistics were conducted using SAS software (Ver. 9.2, SAS Institute, Inc., Cary, NC). P < 0.05 was taken as the level of statistical significance for all procedures.