Hernández-González T, González-Barrio R, Escobar C, et al. Timing of chocolate intake affects hunger, substrate oxidation, and microbiota: A randomized controlled trial. FASEB J. 2021;35(7):e21649.
A randomized, crossover design
To examine whether the timing of chocolate consumption may differentially affect energy balance and impact body weight due to changes in energy intake, substrate oxidation, microbiota (composition/function), and circadian-related variables.
Nineteen postmenopausal Caucasian females from Murcia, Spain, aged 52 ± 4 years were studied. Their mean initial weight was 65.5 kg, body mass index (BMI) was 25.0 kg/m2, and body fat was 32.7%. They were following their usual dietary habits (ad libitum).
Study Medication and Dosage
The participants shifted between 3 patterns of chocolate consumption. They:
- Consumed 100 grams of milk chocolate daily for 2 weeks within an hour of waking up together with breakfast (BC).
- Consumed 100 grams of milk chocolate for 2 weeks within the hour before bedtime (EC).
- Refrained from all chocolate consumption for 2 weeks (N).
The order of this sequence was randomized. There was a week of washout between each 2-week period of chocolate consumption during which participants followed their regular diets but without any chocolate.
The 100-gram chocolate portion was ~30% of habitual total daily caloric intake in the volunteers studied. For reference, a typical Hershey milk chocolate bar weighs 43 grams, so the participants were eating the equivalent of about 2.5 chocolate bars.
The volunteers underwent an extensive series of tests and measurements at multiple points during the study that included:
- Anthropometry: weight, height, BMI, body fat%, waist circumference.
- Dietary intake: a daily intake diary that was confirmed with a photograph of all food consumed and which were used to calculate nutritional intake. These records were used to calculate dietary scores for Mediterranean Diet and other dietary patterns.
- Hunger and appetite for sweets: visual scale on multiple points.
- Sleep patterns: a diary confirmed by objective measurements.
- Indirect calorimetry: objective data on respiratory quotient, energy consumption, and fat oxidation.
- Salivary samples: cortisol measured 3 times per day.
- Fasting glucose: measured on the last day of each experimental period.
- Fecal samples: collected on the first and last day of each experimental period to measure fecal microbiota and short-chain fatty acids.
The participants wore a wristwatch that measured body temperature and acted as an accelerometer. This instrument also provided objective data on sleep including onset, duration, awakenings, etc.
Analysis of data showed that 14 days of chocolate consumption did not increase body weight. Chocolate consumption decreased hunger and desire for sweets (P<0.005), and reduced ad libitum energy intake by ~300 kcal/day when participants consumed chocolate in the morning and by ~150 kcal/day when the chocolate was eaten at night (P=0.01). This decrease did not fully compensate for the extra energy contribution of chocolate (542 kcal/day).
Eating chocolate at night increased physical activity by 6.9%, heat dissipation after meals 1.3%, and carbohydrate oxidation by 35.3% (P<0.05). Morning chocolate consumption reduced fasting glucose (4.4%) and waist circumference (1.7%) and increased lipid oxidation (25.6%).
Principal component analyses showed that chocolate intake resulted in differential microbiota profiles and function regardless of when it is consumed (P<0.05). Heat map of wrist temperature and sleep records showed that evening chocolate consumption improved sleep, inducing regular timing of sleep episodes with lower variability of sleep onset day to day than eating chocolate in the morning did (60 min vs 78 min; P=0.028).
Having chocolate in the morning or in the evening resulted in different effects on hunger and appetite, substrate oxidation, fasting glucose, microbiota, and sleep and temperature rhythms.
In recent years epidemiological studies have suggested that chocolate consumption is associated with reducing cardiovascular disease morbidity and mortality. An umbrella review of meta-analyses published in June 2019 by Veronese et al reported that “among observational studies, including a total of 1,061,637 participants, the best available evidence suggests that chocolate consumption is associated with reduced risk of cardiovascular disease (CVD) death (n=4 studies), acute myocardial infarction (n=6), stroke (n=5) and diabetes (n=6), although this was based on a weak evidence of credibility.”1 This is adequate evidence that we should be encouraging our patients—or at least those at risk of CVD—to consume chocolate.
Yet convincing patients to incorporate chocolate into their diets has proven difficult, despite the fact that the majority of people will readily admit to enjoying it. The obstacle has been their belief that eating chocolate every day will contribute to weight gain or disrupt sleep. The results of this study call into question the validity of both of these beliefs, at least in the short term.
We should note that the 100 grams of chocolate per day may be overkill. According to a 2019 meta-analysis of 23 studies on chocolate consumption and risk of cardiovascular disease (including 405,304 participants and 35,093 cases of CVD), “a non-linear dose-response (Pnon-linearity=0.001) indicated that the most appropriate dose of chocolate consumption for reducing risk of CVD was 45 g/week (RR: 0.890; 95% CI: 0.849-0.932).”2
Note though that this current study under examination used milk chocolate. Past studies have generally considered dark chocolate, which contains higher concentrations of polyphenols. Hernández-González et al chose to give participants milk chocolate because it is more common to consume chocolate in this form, would be more tolerable, and would more closely reflect actual conditions in the real world.
In this study of menopausal women, eating a generous serving of chocolate everyday did not have a significant impact on weight over a 2-week period. In part this may be because they compensated by decreasing their total daily calories. Daily calories dropped more when they ate the chocolate in the morning than before bed. The milk chocolate in the morning induced 26% more lipid oxidation than in the evening. These findings confirm what had been observed in mice: Chocolate decreases fat accumulation in adipocytes3 and decreases lipogenic enzymes.4
Chocolate eaten in the evening induced 35% greater carbohydrate oxidation, with neither eating pattern changing basal metabolic rates. Evening chocolate increased fasting glucose by about 5%. Morning chocolate did the opposite, decreasing fasting glucose by 4.4% and reducing waist circumference.
Chocolate consumption increased abundances of Actinobacteria and reduced relative abundance of Firmicutes (P=0.003). There were nonsignificant trends noted with chocolate toward an increased abundance of Akkermansia (Ptrend=0.09), Ruminococcus (P=0.029), and Dorea genus (P=0.020).
Chocolate increased short-chain fatty acid levels in the stool samples, more so when the chocolate was eaten at night.
These results bring to mind the early studies on nut consumption that reported improved lipid profiles and weight loss. Even as study after study confirmed these findings, patients kept telling me their primary care providers were cautioning them about eating too many nuts, warning them nuts would cause weight gain.
That wasn’t true about nuts and may not be true about chocolate either.
Morning chocolate consumption in general had greater impact on gut microbiota than evening consumption.
Nighttime chocolate consumption may increase endogenous carbohydrate reserves of glycogen in muscle and liver the next morning. Thus, chocolate at night might be an asset during high intensity or prolonged exercise the next morning. This study did not test this idea though.
With the link between chocolate and weight gain unlikely, we should continue to encourage patients to consume chocolate regularly. The only caveat, as mentioned in Ren et al, is that the dose response to chocolate is nonlinear; it appears to be an inverted U-shaped curve. This phenomenon was elaborated in a 2019 meta-analysis of cocoa flavanols and human endothelial function. Combining data from 15 published studies with 18 intervention arms, Sun et al calculated that endothelial function could be improved by 1.17% (95% CI: 0.76%-1.57%). The benefit peaked and then decreased with higher flavanol dosing. Participants in these intervention groups received 80 to 1,248 mg (mean: 704 mg) more flavanols than control groups. The dose mattered: The most improvement was seen with 710 mg total flavanols, 95 mg (-)-epicatechin or 25 mg (+)-catechin. Numbers like this are confusing as there is no easy way to access the amount of flavanols in most chocolate products.5
The measurement of cocoa solids, the number most often reported on chocolate packages, has no correlation with flavanol content. In general though, milk chocolate has lower flavanol content than dark chocolate: 0.703 mg/g compared to 1.156 mg/g.6
Flavanols are often called catechins, and this too leads to confusion when interpreting these study results.7 Hernández-González et al do not report flavanol content of the milk chocolate they provided to their study participants. They do say their chocolate “contained 215 mg of theobromine, 2.06 mg of caffeine, and 854 mg of total polyphenols (mainly epicatechin and catechin) per 100 g of chocolate.” We can assume that the milk chocolate used in the study provided adequate catechins/flavanols to reach the desired target range.
Past retrospective studies have that suggested 1-2 servings per week for women8 and 3-6 servings per week for men would have the greatest benefit in lowering heart disease.9
It isn’t just that dose matters with chocolate. This new data from Hernández-González suggest that time of day matters as well. Yet this new study does not offer a prediction as to which time of day eating the chocolate will be more protective for the heart. In an email correspondence, the authors concede that they “didn’t assess flow-mediated dilatation, forearm blood flow or other cardiovascular measures.”
One might hazard a guess that morning would be best as morning chocolate consumption is associated with a greater reduction in blood sugar and waist circumference. But if this study has taught us nothing else, we should assume that our best guess when it comes to chocolate and health is still only a guess.