Effect of tea catechins with caffeine on energy expenditure in middle-aged men and women: a randomized, double-blind, placebo-controlled, crossover trial

Thirty healthy volunteers (17 men and 13 women; mean age 52 ± 4 years; mean BMI 21.9 ± 2.2 kg/m²) participated in this crossover trial. The inclusion criterion was 45–65 years of age. Exclusion criteria were as follows: treatment for severe disease; history or signs of a heart or cerebrovascular disease; taking therapeutic medicine for diabetes, hyperlipidemia or hypertension; smoker; heavy drinker; anemia; food allergies; and hypersensitivity to the test beverages. This trial was performed in accordance with the Declaration of Helsinki, and the trial plan was reviewed and approved by the local ethics committee (Kao Corporation, Tokyo, Japan). All subjects provided their informed, written consent to participate in the study. Recruitment of participants was conducted in January 2017. The treatment period was from February to April 2017.

The test beverages were prepared by the Kao Corporation. The beverage referred to as the catechin beverage contained 611-mg catechins and 88-mg caffeine in 350 mL, and the caffeinated-placebo beverage contained 0-mg catechins and 81-mg caffeine in 350 mL (Table 1). The catechin beverage contained catechins comprising catechin (6%), epicatechin (5%), gallocatechin (22%), epigallocatechin (19%), catechin gallate (4%), epicatechin gallate (6%), gallocatechin gallate (18%), and epigallocatechin gallate (EGCg, 19%). The catechin beverage used in this study is commercially available in Japan, with the production standard of total tea catechins set to 174–190 mg/100 mL to guarantee a total tea catechin content of 540 mg/350 mL.

This study was a two-phase, randomized, double-blind, placebo-controlled, crossover trial with a 2-week washout period between measurement periods (Supplemental Fig. 1). Subjects were randomized by stratified randomization to balance for BMI. The primary endpoint was energy metabolism after a 2-week intervention. After baseline measurements, subjects consumed one test beverage per day for 2 weeks while maintaining their usual lifestyle. Measurements on day 14 of phase 1 were obtained between 08:15 and 12:00. After a washout period of 2 weeks, in phase 2, subjects consumed the other test beverage per day for 2 weeks, again while maintaining their usual lifestyle. On day 14 of phase 2, measurements were obtained according to the same time-course used in phase 1. At each measurement period, energy metabolism and body temperature were measured before and after the test beverage ingestion and blood samples were obtained. During the trial, subjects were instructed to avoid intake of drinks/foods that contain a high amount of catechins, caffeine, and red pepper, and to maintain their usual exercise and dietary habits. Subjects also ate prescribed meals starting with breakfast up to dinner the day before each measurement period (10113 kJ/day; protein, 13E%; fat, 25E%, carbohydrates, 62E% for men and 8163 kJ/day; protein, 13E%; fat, 26E%, carbohydrates, 62E% for women, respectively) and were prohibited from performing strenuous exercise and drinking alcohol. The day before the measurement, subjects ate their prescribed meal by 20:00 and fasted thereafter.

Body weight was measured using an automatic scale (TF-780, Tanita, Tokyo, Japan) and body composition was measured using whole-body dual-energy X-ray absorptiometry (DEXA, Hologic Inc., QRD 4500 W, Waltham, MA). Measurements of the fat-free mass (FFM), fat mass (FM), and fat ratio (%) were obtained using Hologic Systems Software according to the Hologic QRD 4500 User’s Guide.

On measurement days, subjects were rested in a supine position on a bed, awake, in a fasting condition and in a room with a fixed environment (room temperature, 25 °C; humidity, 50%) for at least 30 min (8:30–9:00) before measurement of fasting RMR [16]. Using a hood-type respiratory gas analyzer (ARCO2000, Arco System, Inc., Chiba, Japan) [17], we assessed oxygen consumption (\(\dot{V}\)O₂) and carbon dioxide production (\(\dot{V}\)CO₂), and measured fasting RMR between 9:00 and 9:30 before the test beverage ingestion (Fig. 1). At 09:50, subjects consumed the test beverage within 5 min. Following this, EE after ingestion of the test beverage was measured three times at intervals of 15–30 min, between 10:00 and 11:30 as described above and the values from 10:00 to 11:30 were averaged to determine the EE after ingestion of the test beverage. \(\dot{V}\)O₂ and \(\dot{V}\)CO₂ were used to determine the EE [18] and respiratory quotient (RQ). The technical validity of the indirect calorimeter was repeatedly assessed by an alcohol combustion test throughout each trial on a weekly basis, repeated nine times. The coefficient of variation calculated from repeated measurement of the combustion rate was 1.2%.

In parallel with measurement of the EE, the forehead temperature (proxy for core temperature) was monitored by applying a temperature sensor (SpotOn™ Temperature Monitoring System, 3M, St Paul, USA) [19] to the forehead and using a data logger (LT-200, Gram, Saitama, Japan). The skin temperature was monitored using skin temperature sensors (LT-2N-12, Gram, Saitama, Japan) on the left subclavicular region and the left foot sole. A motion logger (Actigraph, MicroMini, AMI, NY, USA) was used to detect activity during the measurement of fasting RMR and EE after ingestion of the test beverage [20].

Blood was drawn at 9:30 am under fasting conditions. Fasting blood glucose and glycated hemoglobin (HbA1c) were measured using enzymatic methods. Fasting serum insulin, free triiodothyronine (T3), free thyroxine (T4), thyroid-stimulating hormone (TSH), progesterone, luteinizing hormone (LH), follicle-stimulating hormone (FSH) and estradiol were measured using a chemiluminescence immunoassay. Fasting serum total cholesterol, LDL cholesterol, HDL cholesterol, and triglycerides were measured using enzymatic methods. Fasting plasma noradrenaline was measured by HPLC and cortisol was measured by a chemiluminescence immunoassay. Measurements of parameters were performed by LSI Medience Corporation (Tokyo, Japan).

A significant positive correlation was observed between the fasting RMR and the FFM of subjects in both the placebo treatment (r = 0.891, P < 0.001) and the catechin treatment (r = 0.856, P < 0.001). Table 3 shows fasting RMR, EE after ingestion of the test beverage and RQ following the 2-week intervention. Before ingestion of the test beverage, the fasting RMR did not differ significantly between the catechin treatment and the placebo treatment. On the other hand, the EE after ingestion of the test beverage was significantly higher in the catechin treatment than in the placebo treatment [placebo: 5502 ± 757 kJ/day; catechin: 5598 ± 800 kJ/day; P = 0.041; mean difference: 96 kJ/day (95% CI 4–188), Table 3 and Fig. 2]. A repeated-measures ANOVA revealed a significant treatment effect (P = 0.008), while there was no significant interaction between time and treatment (P = 0.584). Although incremental area under the curve (iAUC) at 45–90 min tend to be significantly higher in the catechin treatment (placebo: 8 ± 6 kJ/45 min; catechin: 11 ± 8 kJ/45 min; P = 0.099), no significant differences were observed at 0–90 min (placebo: 15 ± 10 kJ/90 min; catechin: 18 ± 13 kJ/90 min; P = 0.181, Supplemental Fig. 2). No significant differences were observed in RQ (Table 3).

Table 4 shows changes in body temperature during energy metabolism measurements. Before ingestion of the test beverage, the forehead temperature and skin temperature did not differ significantly between the catechin treatment and the placebo treatment. Similarly, after ingestion of the test beverage, the forehead temperature and skin temperature did not differ significantly between the catechin treatment and the placebo treatment. Regression analyses showed no association between changes in RMR and changes in forehead temperature [placebo: y = 0.0001x + 0.117, R² = 0.029, catechin: y = 0.0001x + 0.132, R² = 0.063, x; changes in RMR (kJ/day), y; changes in forehead temperature (°C), R²; coefficient of determination].

Fasting blood parameters after the 2-week intervention are shown in Table 5. Glucose, insulin, total cholesterol, LDL cholesterol, HDL cholesterol, triglycerides, free T3, free T4, TSH, noradrenaline, and cortisol levels did not differ significantly between treatments. Although we could not confirm the menstruation cycles in female subjects, the progesterone, LH, FSH, and estradiol levels in the female subjects did not differ significantly between the treatments. Body weight and body fat were measured using an automatic scale after the 2-week intervention, and there are no differences between catechin and placebo treatment (body weight; placebo treatment: 61.4 ± 9.6 kg; catechin treatment: 61.5 ± 9.6 kg; P = 0.463, body fat; placebo treatment: 24.9 ± 5.6%; catechin treatment: 24.7 ± 5.6%; P = 0.289).