Reference
Mostofsky E, Schlaug G, Mukamal KJ, Rosamond WD, Mittleman MA. Coffee and acute ischemic stroke onset: The Stroke Onset Study. Neurology. 2010;75(18):1583-1588.
Design
Multicenter case-crossover designed study
Participants
The study was conducted in 3 medical centers (Boston, Chapel Hill, and Victoria, BC) between January 2001 and November 2006. Trained interviewers questioned 390 subjects (209 men, 181 women) a median of 3 days after acute ischemic stroke. Each subject's coffee consumption in the hour before stroke symptoms appeared was compared with his or her usual frequency of consumption in the prior year.
Study “Medication and Dosage”
The portion size for 1 serving of coffee was defined as 8 ounces. Similar questions were asked about consumption of caffeinated tea and cola.
Primary Outcome Measures
In those who suffered an ischemic stroke, consumption and timing of coffee consumption.
Key Findings
Seventy-eight percent of the subjects drank coffee in the prior year, 59% within 24 hours and 9% within 1 hour of stroke onset. Thus the relative risk of stroke doubled in the hour after consuming coffee. No apparent increase in risk was seen for caffeinated tea or cola. The stroke and coffee association was only seen in those consuming ≤1 cup per day but not in patients who drank more coffee more regularly.
The stroke and coffee association was only seen in those consuming 1 cup per day but not in patients who drank more coffee more regularly.
Practice Implications
Although this study has received wide coverage in the public media, the relationship is more complex than usually reported. The study is only one of several related papers to be published recently.
The timing of death from heart attack and stroke follows a circadian pattern and peaks after waking in the morning. Mental stress, physical exertion, and simply waking up influences physical parameters such as blood pressure, heart rate, blood flow, endothelial function, and epinephrine levels; all upregulate sympathetic catecholamine production, increasing oxygen demand by the heart at the same time as reducing oxygen supply. The combination leads to an increased risk of stroke in the morning.1 The authors of the current paper believe that they have designed the study to disallow these circadian patterns from affecting their findings.2
Other studies provide inconsistent data regarding coffee consumption and risk of stroke. Ester Lopez-Garcia et al in their 2009 paper reported that coffee consumption lowered risk of stroke. They analyzed data from a cohort of 83,076 women in the Nurses' Health Study across categories of coffee consumption. Those drinking 2 to 3 cups per day had a 19% decreased risk of stroke [RR 0.81 (95% CI, 0.70 to 0.95)]. This association was stronger among non-smokers. Those who drank 4 or more cups of coffee a day had a 43% lower risk than those who didn’t drink coffee regularly.3
Studies have generally also associated coffee drinking with lowering risk of heart attacks. A paper by Dutch researchers published in August 2010 told us that coffee was clearly associated with lowered cardiovascular morbidity and mortality. In their paper, 37,514 participants were observed for 13 years. Drinking 2–3 cups of coffee a day lowered risk of risk of heart disease by 21%. Drinking less or more coffee had less protective effect.4
Not all studies report this protection. A study on Swedish coffee drinkers found that coffee increased risk of heart attacks. Lena Nilsson et al compared coffee consumption in 375 cases of MI against 1,293 matched controls and found a statistically significant positive association between consumption of filtered coffee and MI risk in men. Those who drank coffee 4 or more times a day had a 73% increased risk of having a heart attack.5
An earlier Swedish study from 2003 study found similar results in men. Those who drank 10 dL (~4 cups) had about double the risk of having a heart attack: a relative risk of 1.93 for filtered coffee and 2.20 for boiled coffee.6
Sorting through these various studies, it seems as if the coffee ‘toxicity dose response’ curve may actually be U-shaped. Low and high consumption levels possibly increase risk, while moderate consumption, in the 2–3 cups per day range, may be beneficial.
There’s another possible explanation for these inconsistent findings. Back in the March 2006 issue of JAMA, Cornelis et al from the University of Toronto asked a reasonable question. Coffee does not affect all people the same way. Some people are very sensitive—a cup in the morning will leave them sleepless all night. Others drink coffee with dinner and at bedtime with no ill effect. This variation is due to genetics. The CYP1A2 gene codes for the enzyme that breaks down caffeine. Individuals with one version of this gene are “rapid” caffeine metabolizers. Those who carry another variation are “slow metabolizers.”
In the 2006 Cornelis study, drinking coffee had a different effect on risk for myocardial infarction that depended on whether people were fast or slow metabolizers. For slow metabolizers, drinking coffee increased their risk of having a heart attack. For fast metabolizers, coffee lowered their risk:
For people who metabolize coffee quickly due to their genes, drinking a cup of coffee a day cuts their risk of a heart attack by a little more than half. On the other hand, that same cup of coffee will increase a slow metabolizers risk by about 25%.7
Could it be in the current coffee and stroke risk study, as well as these other coffee risk studies, we are being confused because we are not paying attention to the varying metabolism of the study participants? For example, in the stroke study under consideration, does it seem more likely that slow metabolizers would be in the group that drinks less coffee less often? Fast metabolizers are more likely to drink more coffee more regularly and less likely to suffer from adverse effects from the habit?
Though it is tempting to encourage patients to drink coffee regularly to reduce risk of stroke and heart disease based on the study by Mostofsky et al, there may be more to this story yet to be published.
