Coffee, Tea, and Cocoa and Risk of Stroke

Susanna C. Larsson

Susanna C. Larsson

From the Unit of Nutritional Epidemiology, Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden.

Search for more papers by this author

Originally published10 Dec 2013https://doi.org/10.1161/STROKEAHA.113.003131Stroke. 2014;45:309–314

Other version(s) of this article
- You are viewing the most recent version of this article. Previous versions:
  - January 1, 2013: Previous Version 1

Introduction

Coffee, tea, and cocoa are important dietary sources of polyphenols and have received much attention during the past years because of their potential beneficial effects on cardiovascular health. The polyphenols in these beverages and cocoa may reduce the risk of stroke through multiple mechanisms, including antihypertensive, hypocholesterolemic, antioxidant, and anti-inflammatory effects as well as through improvements of vascular endothelial function and insulin sensitivity. This review summarizes the available evidence from experimental studies, prospective studies, and meta-analyses of the potential role of coffee, tea, and cocoa in the prevention of stroke.

Methods

References for this review were identified through a literature search of the PubMed database through October 2013 by using the following search terms: coffee, tea, cocoa, chocolate, prospective study, cohort study, randomized trial, meta-analysis, review, stroke, cerebral infarction, and cerebrovascular disease. Moreover, the reference lists of pertinent publications were searched manually for further relevant articles. Priority was given to systematic reviews and meta-analyses published during the past 5 years. When >1 meta-analysis on the same topic was available, the most recent publication was included in the present review.

Coffee

Coffee is a complex beverage with hundreds of bioactive components with potential adverse or beneficial effects on the cardiovascular system. The most abundant bioactive compounds in coffee are caffeine, diterpenes (present in the oil), and polyphenols. The cardiovascular effects of coffee drinking depend in part on coffee preparation method and individual characteristics (eg, hypertension and hyperlipidemia).^1–3 There are 2 main methods of coffee preparation: filtered and unfiltered. Filtered coffee, also known as drip-brewed coffee, is the most common mode of preparation in the United States and involves brewing the coffee through a paper filter. Unfiltered coffee, often known as boiled coffee, do not use a filter and includes Scandinavian boiled, French press, Turkish/Greek, and espresso coffees. Espresso is often the base for other drinks, such as latte, cappuccino, macchiato, and caffè Americano.

Caffeine is a stimulant that induces a transient increase in blood pressure (BP). Findings from a meta-analysis of 5 randomized controlled trials (RCTs) of the acute effects of caffeine on BP in individuals with hypertension showed that intake of 200−300 mg caffeine (equivalent to ≈1.5–2 cups of coffee) produced a mean rise of 8.1 mm Hg in systolic blood pressure and of 5.8 mm Hg in diastolic blood pressure (Table 1).¹ The increase in BP was observed in the first hour after caffeine ingestion and lasted for ≥3 hours. However, a meta-analysis of 10 RCTs of the long-term effect of coffee consumption in mainly healthy, normotensive individuals found no significant changes in systolic blood pressure or diastolic blood pressure (Table 1).² Prospective studies of habitual coffee consumption and risk of hypertension have yielded inconsistent results, with a positive association found in 2 out of 4 studies.² Tolerance to the effects of caffeine on BP in some individuals may in part explain why the long-term effects of coffee consumption differ from the short-term effects. Moreover, other compounds present in coffee may counteract the BP-raising effect of caffeine. A study of 6 habitual and 9 nonhabitual coffee drinkers found that intravenous caffeine raised BP in both groups, whereas coffee consumption increased BP in nonhabitual drinkers only.⁴

**Table 1.** Summary of Recent Meta-Analyses of RCTs of the Effects of Coffee or Caffeine Intake on Cardiometabolic Biomarkers
References	Interventions	Duration	No. ofTrials	Outcomes	Effects	Mean Change(95% CI)	Heterogeneity
References	Interventions	Duration	No. ofTrials	Outcomes	Effects	Mean Change(95% CI)	P Value	I², %
Mesas et al, 2011¹	Coffee or caffeine (200–300 mg)	<60–180 min	5	SBP (mm Hg)	↑	8.14 (5.68, 10.61)	0.99	0
Mesas et al, 2011¹	Coffee or caffeine (200–300 mg)	<60–180 min	5	DBP (mm Hg)	↑	5.75 (4.09, 7.41)	0.57	0
Steffen et al, 2012²	Filtered, boiled, instant, or decaffeinated coffee	4–16 wk	10	SBP (mm Hg)	↔	–0.55 (–2.46, 1.36)	<0.001	72
Steffen et al, 2012²	Filtered, boiled, instant, or decaffeinated coffee	4–16 wk	10	DBP (mm Hg)	↔	–0.45 (–1.52, 0.61)	0.07	41
Cai et al, 2012³	Filtered coffee	14–79 d	10	Total cholesterol (mmol/L)^*	↑	0.09 (0.02, 0.17)	0.52	0
			9	LDL cholesterol (mmol/L)^*	↔	0.06 (–0.03, 0.15)	0.35	10
			5	Triglycerides (mmol/L)^*	↔	0.04 (–0.05, 0.13)	0.43	0
Cai et al, 2012³	Boiled/unfiltered coffee	14–79 d	12	Total cholesterol (mmol/L)^*	↑	0.33 (0.18, 0.49)	<0.0010	79
			6	LDL cholesterol (mmol/L)^*	↑	0.31 (0.08, 0.53)	0.002	73
			6	Triglycerides (mg/dL)^*	↑	0.21 (0.05, 0.37)	0.001	77

CI indicates confidence interval; DBP; diastolic blood pressure; LDL, low-density lipoprotein; RCTs, randomized controlled trials; and SBP; systolic blood pressure.

^*Values were converted from mg/dL to mmol/L by dividing the levels of cholesterol (total, LDL, and HDL) by 38.67; triglyceride levels by 88.57; and glucose levels by 18.02.

The diterpenes cafestol and kahweol have cholesterol-raising properties.⁵ The diterpenes are extracted from the coffee beans by hot water but are retained by a paper filter.⁵ Hence, unfiltered coffee, particularly Scandinavian boiled and Turkish coffees, contains much higher concentrations of diterpenes than filtered coffee, which only contains negligible amounts.⁶ Espresso coffee contains intermediate amounts.⁶ In a meta-analysis of 12 RCTs, including 1017 subjects, consumption of unfiltered coffee significantly increased total cholesterol, low-density lipoprotein (LDL) cholesterol, and triglyceride concentrations, whereas filtered coffee consumption produced a small change in total cholesterol concentrations only (Table 1).³ The meta-analysis further showed that those who had hyperlipidemia seemed to be more sensitive to the cholesterol-raising effect of coffee.³

Coffee is rich in various polyphenols, most notably chlorogenic acid (CGA), which possesses antioxidant activities in vitro.⁷ Studies in animals have demonstrated that coffee and caffeic acid, a primary CGA metabolite, can decrease lipid peroxidation, thus indicating also an in vivo antioxidant activity.⁷ However, there is controversy on whether chlorogenic acid and other polyphenols in coffee could suppress the oxidative modification of LDL particles in humans. Among 3 available studies on this topic, 2 studies reported a protective effect of 1 cup of boiled⁸ or filtered coffee⁹ on LDL oxidation, whereas 1 study found neither short-term nor long-term effects of filtered coffee consumption on lipid peroxidation.¹⁰ As opposed to caffeine, CGA have been demonstrated to have antihypertensive effects,^11,12 possibly via nitric oxide–mediated vasodilation.¹² Results from an RCT of 23 healthy adults showed that CGA ingestion significantly reduced systolic blood pressure by 2.41 mm Hg and diastolic blood pressure by 1.53 mm Hg.¹¹

Epidemiological Studies on Coffee and Stroke

In the past, coffee was generally viewed as a risk factor for cardiovascular disease. However, recent evidence suggests that moderate coffee consumption may reduce stroke risk. Results from a meta-analysis of 11 prospective studies (published through January 2011) involving 479 689 participants and 10 003 stroke cases showed a nonlinear relationship between coffee consumption and stroke risk (Figure 1).¹³ Compared with no coffee consumption, the overall relative risks (RRs; 95% CI) of total stroke were 0.87 (0.81–0.93) for 2, 0.84 (0.77–0.91) for 3−4, 0.88 (0.79–0.97) for 6, and 0.94 (0.80–1.10) for 8 cups/d of coffee.¹³ Risk estimates were similar for ischemic and hemorrhagic stroke and for men and women at lower levels of coffee consumption (≤2 cups/d).¹³

**Figure 1.** Relative risks of stroke by coffee consumption in prospective studies. The relative risks were extracted from the meta-analysis by Larsson and Orsini.¹³

Three prospective studies on coffee consumption and stroke^14–16 were published since the meta-analysis. Two of them confirmed an inverse association of moderate coffee consumption with stroke incidence¹⁴ or mortality.¹⁵ Findings from a large prospective cohort of 229 119 US men and 173 141 US women showed an inverse association between moderate coffee consumption and stroke mortality.¹⁵ In men, the multivariable RRs (95% CI) of total stroke death were 0.99 (0.79–1.24) for <1, 0.92 (0.73–1.15) for 1, 0.84 (0.68–1.02) for 2−3, 0.65 (0.51–0.84) for 4−5, and 0.83 (0.61–1.14) for ≥6 cups/d compared with no coffee consumption (P for trend=0.003).¹⁵ In women, compared with no coffee consumption, the RRs (95% CI) were 1.15 (0.91–1.45) for <1, 0.89 (0.70–1.13) for 1, 0.93 (0.75–1.15) for 2−3, 0.82 (0.62–1.09) for 4−5, and 0.84 (0.56–1.25) for ≥6 cups/d (P for trend=0.05).¹⁵ A prospective study of 82 369 Japanese adults also observed an inverse association between moderate coffee consumption and stroke risk (RR, 0.81; 95% CI [0.72–0.91], for ≥2 cups/d versus none).¹⁴ No association between caffeinated or decaffeinated coffee consumption and stroke risk was observed in a prospective study of 42 659 German adults, but the number of cases was small (n=310).¹⁶ Coffee consumption is usually associated with a less health conscious diet and lifestyle. Although most studies controlled for other dietary and lifestyle factors, residual confounding may in part explain the inconsistent results. Furthermore, because the relative composition of bioactive compounds in coffee varies by coffee preparation method, this could contribute to the heterogeneity among studies in different populations.

Green and Black Tea

Tea is the most frequently consumed beverage in the world after water. Tea is produced from the leaves of the plant Camellia sinensis and can be classified by degree of fermentation: black tea (fermented), predominantly consumed in Western countries; oolong tea (partially fermented), primarily consumed in Southern China and Taiwan; and green tea (unfermented), mainly consumed in Asia. All types of tea are rich in various flavonoids. Catechins are the main flavonoids in green tea, whereas black tea mainly contains condensed flavonoids, such as theaflavins and thearubigins.¹⁷ Tea and tea-derived flavonoids have been demonstrated to have a hypocholesterolemic effect and to reduce the development of atherosclerosis in animal models.^17,18 Tea flavonoids can enhance nitric oxide status and improve endothelial function, which could at least partly be responsible for the benefits of tea on cardiovascular health.^17,18

Studies in humans also indicate potential beneficial effects of consumption of green and black tea on cardiometabolic risk factors, including endothelial function (measured by flow-mediated dilatation), blood pressure, and cholesterol and blood glucose concentrations (Table 2). The most consistent findings are for endothelial function. In a meta-analysis of 9 RCTs (2 on green tea, 6 on black tea, and 1 on both types of tea), involving 213 participants, the overall absolute increase in FMD of tea consumption (median daily dose of 500 mL tea, equivalent to 2–3 cups) versus placebo was 2.6% of the arterial diameter.¹⁹ This is a relative improvement of ≈40% compared with the average FMD of 6.3% measured under placebo or baseline conditions.¹⁹ Results from a meta-analysis of 14 short-term (≤3 months) RCTs showed that green tea consumption lowered total and LDL cholesterol concentrations but had no effect on high-density lipoprotein cholesterol.²⁰ In another meta-analysis of RCTs of ≥3 months duration, both green and black tea consumption reduced LDL cholesterol concentrations as well as BP.²¹ With regard to glucose and insulin, 2 meta-analyses of several RCTs found that green tea consumption decreased fasting blood glucose concentrations, whereas results for insulin and hemoglobin A1c concentrations were inconsistent.^22,23

**Table 2.** Summary of Recent Meta-Analyses of RCTs of the Effects of Green and Black Tea Consumption on Cardiometabolic Biomarkers
References	Interventions	Duration	No. ofTrials	Outcomes	Effects	Mean Change(95% CI)	Heterogeneity
References	Interventions	Duration	No. ofTrials	Outcomes	Effects	Mean Change(95% CI)	P Value	I², %
Ras et al, 2011¹⁹	Green or black tea as a beverage	≤120 min–4 wk	9	FMD (%)	↑	2.6 (1.8, 3.3)	<0.001	75.8
Zheng et al, 2011²⁰	Green tea as a beverage or green tea extract	3 wk–3 mo	14	Total cholesterol (mmol/L)^*	↓	–0.19 (–0.21, –0.16)	0.35	9
			11	LDL cholesterol (mmol/L)^*	↓	–0.06 (–0.08, –0.03)	0.20	25
			12	HDL cholesterol (mmol/L)^*	↔	0.006 (–0.02, 0.03)	0.27	18
Hartley et al, 2013²¹	Green tea as a beverage or green tea extract	3–6 mo	2	SBP (mm Hg)	↓	–3.18 (–5.25, –1.11)	0.72	0
			2	DBP (mm Hg)	↓	–3.42 (–4.54, – 2.30)	0.39	0
			4	Total cholesterol (mmol/L)	↓	–0.62 (–0.77, –0.46)	0.28	21
			4	LDL cholesterol (mmol/L)	↓	–0.64 (–0.77, –0.52)	0.33	13
			4	HDL cholesterol (mmol/L)	↔	0.01 (–0.08, 0.11)	0.18	39
			4	Triglycerides (mmol/L)	↔	–0.08 (–0.24, 0.07)	0.41	0
Hartley et al, 2013²¹	Black tea extracts, in tablet form or as a drink	3–6 mo	2	SBP (mm Hg)	↓	−1.85 (−3.21, −0.48)	0.49	0
			2	DBP (mm Hg)	↔	−1.27 (−3.06, 0.53)	0.53	0
			2	Total cholesterol (mmol/L)	NA	NA^*	NA	84
			3	LDL cholesterol (mmol/L)	↓	−0.43 (−0.56, −0.31)	0.22	33
			3	HDL cholesterol (mmol/L)	↔	−0.01 (−0.06, 0.04)	0.20	36
			3	Triglycerides (mmol/L)	NA	NA^*	NA	64
Liu et al, 2013²²	Green tea extract or decaffeinated green tea extract	4 wk–3 mo	17	Fasting glucose (mmol/L)	↓	–0.09 (–0.15, –0.03)	0.73	0
			13	Fasting insulin (μIU/mL)	↓	–1.16 (–1.91, –0.40)	0.01	57
			7	Hb A_1c (%)	↓	–0.30 (–0.37, –0.22)	0.10	44
			5	HOMA-IR (units)	↔	–0.04 (–0.67, 0.59)	0.13	44
Zheng et al, 2013²³	Green tea extract or decaffeinated green tea extract	3–24 wk	22	Fasting glucose (mmol/L)^†	↓	–0.08 (–0.14, –0.02)	0.92	0
			16	Fasting insulin (μU/mL)	↔	0.04 (–0.36, 0.45)	0.35	9
			6	Hb A_1c (%)	↔	–0.04 (–0.15, 0.08)	0.12	40
			6	HOMA-IR (units)	↔	–0.05 (–0.37, 0.26)	0.31	16

CI indicates confidence interval; DBP; diastolic blood pressure; FMD, flow-mediated dilation; Hb A_1c, glycohemoglobin; HDL, high-density lipoprotein; HOMA-IR, homeostatic model assessment index for insulin resistance; LDL, low-density lipoprotein; NA, not available; RCTs, randomized controlled trials; and SBP; systolic blood pressure.

^*Meta-analysis was not performed because of significant heterogeneity between the trials.

^†Values were converted from mg/dL to mmol/L by dividing by 38.67 for total cholesterol, LDL cholesterol, and HDL cholesterol; by 88.57 for triglycerides; and by 18.02 for glucose.

Epidemiological Studies on Tea and Stroke

In a meta-analysis of 14 prospective studies of green or black tea consumption, the overall RR of total stroke for a 3-cup/d increment in tea consumption was 0.87 (95% CI, 0.81–0.94), with heterogeneity among studies (P=0.006).²⁴ There was no evidence of publication bias (Egger test: P=0.85).²⁴ The association was similar in men and women and among most subgroups, but was slightly stronger for green tea (RR=0.83; 95% CI [0.72–0.96]; P_{heterogeneity}<0.01; n=5 studies) than for black tea (RR=0.91; 95% CI [0.83–0.98]; P_{heterogeneity}=0.17; n=13 studies).²⁴ The heterogeneity may be because of differences in types of tea, tea preparation methods (amounts of tea leaves, cup size, brewing time, water temperatures, addition of milk or sugar, etc), stroke measures, and analysis strategies.²⁴

Two recent large prospective studies of green¹⁴ or black tea²⁵ consumption confirmed a reduction in stroke risk associated with high tea consumption. Results from a cohort of 82 369 Japanese men and women showed a significant 20% reduced risk of total stroke among those who consumed ≥4 cups/d of green tea.¹⁴ In a cohort of 74 961 Swedish men and women, consumption of ≥4 cups/d of black tea, compared with no consumption, was associated with a significant 21% lower risk of total stroke.²⁵ In both studies, the association was similar for ischemic stroke and intracerebral hemorrhage.

Cacao Products

Cacao products, such as chocolate, are rich sources of flavonoids, mainly flavan-3-ols (also referred to as flavanols), which are potent antioxidant and anti-inflammatory compounds. Both the flavan-3-ol content and the total antioxidant capacity in plasma increase after cocoa consumption.²⁶ Whether these effects are reduced when cocoa is ingested with milk or when cocoa is consumed as milk chocolate is controversial.²⁶ Flavanols found in cocoa have also been shown to increase the formation of endothelial nitric oxide, which promotes vasodilation and thus blood pressure reduction.²⁷

The potential benefits of cacao products on cardiovascular health have been examined in several short-term RCTs, and results from those trials have been summarized in meta-analyses. The overall results from 2 meta-analyses indicate that cocoa or chocolate intake may modestly reduce systolic blood pressure²⁷ and diastolic blood pressure,^27,28 but findings from individual trials were inconsistent (Table 3). A recent meta-analysis of 42 acute or short-term chronic (≤18 weeks) RCTs found that cocoa or chocolate interventions significantly reduced fasting insulin concentrations, insulin resistance, and mean arterial pressure as well as improved endothelial function measured by FMD (Table 3).²⁸ Cocoa or chocolate consumption had only marginally significant or no effects on blood concentrations of cholesterol (total, LDL, and high-density lipoprotein), triglycerides, glucose, hemoglobin A1c, and C-reactive protein.²⁸ In a recent 1-year trial comprising 93 postmenopausal women with type 2 diabetes mellitus, a combination of flavan-3-ols and isoflavones reduced LDL cholesterol (−0.1 mmol/L; P=0.04) and insulin (−0.8 mU/L; P=0.02) concentrations and the homeostatic model assessment index for insulin resistance (−0.3; P=0.004).²⁹

**Table 3.** Summary of Recent Meta-Analyses of RCTs of the Effects of Cocoa or Chocolate Consumption on Cardiometabolic Biomarkers
References	Interventions	Duration	No. ofTrials	Outcomes	Effects	Mean Change(95% CI)	Heterogeneity
References	Interventions	Duration	No. ofTrials	Outcomes	Effects	Mean Change(95% CI)	P Value	I², %
Ried et al, 2012²⁷	Dark or milk chocolate or flavanol-rich cocoa powder	2–8 wk*	20	SBP (mm Hg)	↓	–2.77 (–4.72, –0.82)	<0.001	83
Ried et al, 2012²⁷	Dark or milk chocolate or flavanol-rich cocoa powder	2–8 wk*	19	DBP (mm Hg)	↓	–2.20 (–3.46, –0.93)	<0.001	70
Hooper et al, 2012²⁸	Dark or milk chocolate, cocoa drinks, or cocoa supplements	≤18 wk	3	SBP (mm Hg), acute†	↔	–1.75 (–6.27, 2.77)	NA	85
			23	SBP (mm Hg), chronic†	↔	–1.50 (–3.43, 0.43)	NA	NA
			3	DBP (mm Hg), acute	↔	–1.38 (–4.14, 1.38)	NA	79
			22	DBP (mm Hg), chronic	↓	–1.60 (–2.77, –0.43)	NA	52
			4	MAP (mm Hg), chronic	↓	–1.64 (–3.27, –0.01)	NA	0
			11	FMD (%), acute	↑	3.19 (2.04, 4.33)	<0.001	84
			11	FMD (%), chronic	↑	1.34 (1.00, 1.68)	0.58	0
			11	Fasting glucose (mmol/L)	↔	–0.02 (–0.22, 0.17)	0.02	54
			5	Fasting insulin (μU/mL)	↓	–2.65 (–4.65, –0.65)	0.50	0
			3	Hb A_1c (%)	↔	0.02 (–0.09, 0.14)	NA	NA
			6	HOMA-IR (units)	↓	–0.67 (–0.98, –0.36)	0.61	0
			21	Total cholesterol (mmol/L)	↔	–0.04 (–0.11, 0.03)	NA	NA
			21	LDL cholesterol (mmol/L)	↓	–0.07 (–0.13, 0.00)	NA	NA
			21	HDL cholesterol (mmol/L)	↑	0.03 (0.00, 0.06)	NA	NA
			22	Triglycerides (mmol/L)	↓	–0.05 (–0.09, –0.01)	NA	NA
			10	C-reactive protein (mg/L)	↔	0.12 (–0.42, 0.66)	NA	NA

CI indicates confidence interval; DBP; diastolic blood pressure; FMD, flow-mediated dilation; Hb A_1c, glycohemoglobin; HDL, high-density lipoprotein; HOMA-IR, homeostatic model assessment index for insulin resistance; LDL, low-density lipoprotein; MAP, mean arterial pressure; NA, not available; RCTs, randomized controlled trials; and SBP; systolic blood pressure.

^*One trial was 18 wk.

^†Acute effect: studies of 90−150 min duration; chronic effect: studies of <3, 3−6, or 7−26 wk duration.

Several controlled intervention studies have found that flavanols present in cocoa may improve platelet function. Based on data from 5 trials, Ostertag et al³⁰ estimated that intake of 100 mg of flavanols induces a 3% to 11% reduction in platelet aggregation.

Epidemiological Studies on Chocolate and Stroke

The few prospective studies of chocolate consumption in relation to stroke risk have reported either a statistically significant^31–33 or a nonsignificant inverse association^34,35 (Figure 2). Results from a meta-analysis of those 5 studies (4 from Europe and 1 from the United States) showed a significant 19% lower risk of stroke when comparing the highest with the lowest category of chocolate consumption (Figure 2) and a significant 14% reduction in stroke risk for a 50-g/week increment in chocolate consumption, without heterogeneity among studies.³³ There was indication of potential publication bias in the meta-analysis for the highest versus lowest category of chocolate consumption (Egger test: P=0.03) but not in the dose–response meta-analysis (Egger test: P=0.26).³³

**Figure 2.** Relative risks (RRs) of stroke for the highest versus lowest category of chocolate consumption in prospective studies. Squares represent the study-specific RRs (size of the square indicates the study-specific statistical weight, that is, the inverse of the variance); the horizontal lines represent 95% confidence intervals (CIs); and the diamond represents the overall RR estimate with its 95% CI. Study-specific RRs were combined by using a random effects model. The RRs were extracted from the meta-analysis by Larsson et al.³³ Heterogeneity test: I²=0%; P=0.47. COSM indicates Cohort of Swedish Men³³; EPIC, European Prospective Investigation into Cancer³¹; IWHS, Iowa Women’s Health Study³⁴; SHEEP, Stockholm Heart Epidemiology Program³⁵; and SMC, Swedish Mammography Cohort.³²

Summary

Current evidence from experimental studies in animals and humans along with findings from prospective studies indicates beneficial effects of green and black tea as well as chocolate on cardiovascular health, and that tea and chocolate consumption may reduce the risk of stroke. The strongest evidence exists for beneficial effects of tea and cocoa on endothelial function, total and LDL cholesterol (tea only), and insulin sensitivity (cocoa only). The majority of prospective studies have reported a weak inverse association between moderate consumption of coffee and risk of stroke. However, there are yet no clear biological mechanisms whereby coffee might provide cardiovascular health benefits. Awaiting the results from further long-term RCTs and prospective studies, moderate consumption of filtered coffee, tea, and dark chocolate seems prudent.

Disclosures

None.

Footnotes

Correspondence to Susanna C. Larsson, PhD, Unit of Nutritional Epidemiology, Institute of Environmental Medicine, Karolinska Institutet, Box 210, SE- 17177 Stockholm, Sweden. E-mail [email protected]

References

1. Mesas AE, Leon-Muñoz LM, Rodriguez-Artalejo F, Lopez-Garcia E. The effect of coffee on blood pressure and cardiovascular disease in hypertensive individuals: a systematic review and meta-analysis.Am J Clin Nutr. 2011; 94:1113–1126.Crossref Medline Google Scholar
2. Steffen M, Kuhle C, Hensrud D, Erwin PJ, Murad MH. The effect of coffee consumption on blood pressure and the development of hypertension: a systematic review and meta-analysis.J Hypertens. 2012; 30:2245–2254.Crossref Medline Google Scholar
3. Cai L, Ma D, Zhang Y, Liu Z, Wang P. The effect of coffee consumption on serum lipids: a meta-analysis of randomized controlled trials.Eur J Clin Nutr. 2012; 66:872–877.Crossref Medline Google Scholar
4. Corti R, Binggeli C, Sudano I, Spieker L, Hänseler E, Ruschitzka F, et al.. Coffee acutely increases sympathetic nerve activity and blood pressure independently of caffeine content: role of habitual versus nonhabitual drinking.Circulation. 2002; 106:2935–2940.Link Google Scholar
5. Urgert R, Katan MB. The cholesterol-raising factor from coffee beans.Annu Rev Nutr. 1997; 17:305–324.Crossref Medline Google Scholar
6. Gross G, Jaccaud E, Huggett AC. Analysis of the content of the diterpenes cafestol and kahweol in coffee brews.Food Chem Toxicol. 1997; 35:547–554.Crossref Medline Google Scholar
7. Bonita JS, Mandarano M, Shuta D, Vinson J. Coffee and cardiovascular disease: in vitro, cellular, animal, and human studies.Pharmacol Res. 2007; 55:187–198.Crossref Medline Google Scholar
8. Yukawa GS, Mune M, Otani H, Tone Y, Liang XM, Iwahashi H, et al.. Effects of coffee consumption on oxidative susceptibility of low-density lipoproteins and serum lipid levels in humans.Biochemistry (Mosc). 2004; 69:70–74.Crossref Medline Google Scholar
9. Natella F, Nardini M, Belelli F, Scaccini C. Coffee drinking induces incorporation of phenolic acids into LDL and increases the resistance of LDL to ex vivo oxidation in humans.Am J Clin Nutr. 2007; 86:604–609.Crossref Medline Google Scholar
10. Mursu J, Voutilainen S, Nurmi T, Alfthan G, Virtanen JK, Rissanen TH, et al.. The effects of coffee consumption on lipid peroxidation and plasma total homocysteine concentrations: a clinical trial.Free Radic Biol Med. 2005; 38:527–534.Crossref Medline Google Scholar
11. Mubarak A, Bondonno CP, Liu AH, Considine MJ, Rich L, Mas E, et al.. Acute effects of chlorogenic acid on nitric oxide status, endothelial function, and blood pressure in healthy volunteers: a randomized trial.J Agric Food Chem. 2012; 60:9130–9136.Crossref Medline Google Scholar
12. Zhao Y, Wang J, Ballevre O, Luo H, Zhang W. Antihypertensive effects and mechanisms of chlorogenic acids.Hypertens Res. 2012; 35:370–374.Crossref Medline Google Scholar
13. Larsson SC, Orsini N. Coffee consumption and risk of stroke: a dose-response meta-analysis of prospective studies.Am J Epidemiol. 2011; 174:993–1001.Crossref Medline Google Scholar
14. Kokubo Y, Iso H, Saito I, Yamagishi K, Yatsuya H, Ishihara J, et al.. The impact of green tea and coffee consumption on the reduced risk of stroke incidence in Japanese population: the Japan public health center-based study cohort.Stroke. 2013; 44:1369–1374.Link Google Scholar
15. Freedman ND, Park Y, Abnet CC, Hollenbeck AR, Sinha R. Association of coffee drinking with total and cause-specific mortality.N Engl J Med. 2012; 366:1891–1904.Crossref Medline Google Scholar
16. Floegel A, Pischon T, Bergmann MM, Teucher B, Kaaks R, Boeing H. Coffee consumption and risk of chronic disease in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Germany study.Am J Clin Nutr. 2012; 95:901–908.Crossref Medline Google Scholar
17. Hodgson JM, Croft KD. Tea flavonoids and cardiovascular health.Mol Aspects Med. 2010; 31:495–502.Crossref Medline Google Scholar
18. Moore RJ, Jackson KG, Minihane AM. Green tea (Camellia sinensis) catechins and vascular function.Br J Nutr. 2009; 102:1790–1802.Crossref Medline Google Scholar
19. Ras RT, Zock PL, Draijer R. Tea consumption enhances endothelial-dependent vasodilation; a meta-analysis.PLoS One. 2011; 6:e16974.Crossref Medline Google Scholar
20. Zheng XX, Xu YL, Li SH, Liu XX, Hui R, Huang XH. Green tea intake lowers fasting serum total and LDL cholesterol in adults: a meta-analysis of 14 randomized controlled trials.Am J Clin Nutr. 2011; 94:601–610.Crossref Medline Google Scholar
21. Hartley L, Flowers N, Holmes J, Clarke A, Stranges S, Hooper L, et al.. Green and black tea for the primary prevention of cardiovascular disease.Cochrane Database Syst Rev. 2013; 6:CD009934.Google Scholar
22. Liu K, Zhou R, Wang B, Chen K, Shi LY, Zhu JD, et al.. Effect of green tea on glucose control and insulin sensitivity: a meta-analysis of 17 randomized controlled trials.Am J Clin Nutr. 2013; 98:340–348.Crossref Medline Google Scholar
23. Zheng XX, Xu YL, Li SH, Hui R, Wu YJ, Huang XH. Effects of green tea catechins with or without caffeine on glycemic control in adults: a meta-analysis of randomized controlled trials.Am J Clin Nutr. 2013; 97:750–762.Crossref Medline Google Scholar
24. Shen L, Song LG, Ma H, Jin CN, Wang JA, Xiang MX. Tea consumption and risk of stroke: a dose-response meta-analysis of prospective studies.J Zhejiang Univ Sci B. 2012; 13:652–662.Crossref Medline Google Scholar
25. Larsson SC, Virtamo J, Wolk A. Black tea consumption and risk of stroke in women and men.Ann Epidemiol. 2013; 23:157–160.Crossref Medline Google Scholar
26. Corti R, Flammer AJ, Hollenberg NK, Lüscher TF. Cocoa and cardiovascular health.Circulation. 2009; 119:1433–1441.Link Google Scholar
27. Ried K, Sullivan TR, Fakler P, Frank OR, Stocks NP. Effect of cocoa on blood pressure.Cochrane Database Syst Rev. 2012; 8:CD008893.Google Scholar
28. Hooper L, Kay C, Abdelhamid A, Kroon PA, Cohn JS, Rimm EB, et al.. Effects of chocolate, cocoa, and flavan-3-ols on cardiovascular health: a systematic review and meta-analysis of randomized trials.Am J Clin Nutr. 2012; 95:740–751.Crossref Medline Google Scholar
29. Curtis PJ, Sampson M, Potter J, Dhatariya K, Kroon PA, Cassidy A. Chronic ingestion of flavan-3-ols and isoflavones improves insulin sensitivity and lipoprotein status and attenuates estimated 10-year CVD risk in medicated postmenopausal women with type 2 diabetes: a 1-year, double-blind, randomized, controlled trial.Diabetes Care. 2012; 35:226–232.Crossref Medline Google Scholar
30. Ostertag LM, O’Kennedy N, Kroon PA, Duthie GG, de Roos B. Impact of dietary polyphenols on human platelet function–a critical review of controlled dietary intervention studies.Mol Nutr Food Res. 2010; 54:60–81.Crossref Medline Google Scholar
31. Buijsse B, Weikert C, Drogan D, Bergmann M, Boeing H. Chocolate consumption in relation to blood pressure and risk of cardiovascular disease in German adults.Eur Heart J. 2010; 31:1616–1623.Crossref Medline Google Scholar
32. Larsson SC, Virtamo J, Wolk A. Chocolate consumption and risk of stroke in women.J Am Coll Cardiol. 2011; 58:1828–1829.Crossref Medline Google Scholar
33. Larsson SC, Virtamo J, Wolk A. Chocolate consumption and risk of stroke: a prospective cohort of men and meta-analysis.Neurology. 2012; 79:1223–1229.Crossref Medline Google Scholar
34. Mink PJ, Scrafford CG, Barraj LM, Harnack L, Hong CP, Nettleton JA, et al.. Flavonoid intake and cardiovascular disease mortality: a prospective study in postmenopausal women.Am J Clin Nutr. 2007; 85:895–909.Crossref Medline Google Scholar
35. Janszky I, Mukamal KJ, Ljung R, Ahnve S, Ahlbom A, Hallqvist J. Chocolate consumption and mortality following a first acute myocardial infarction: the Stockholm Heart Epidemiology Program.J Intern Med. 2009; 266:248–257.Crossref Medline Google Scholar