The authors of this study note that the majority of studies evaluating the effects of caffeine have focused on coffee consumption and that the association of caffeine in pharmaceutical products and haemorrhagic stroke warrants further study.
Their multicentre matched case-control study included 940 patients aged 30-84 years with acute non-traumatic haemorrhagic stroke, who were able to communicate and complete an interview within 30 days of stroke onset. Each patient was matched to one hospital control (hospitalised in the same institution for diseases other than stroke) and one community control (recruited from siblings, friends or neighbours of the patient) according to age and sex. Information on all medications taken within 14 days of the index date was collected as part of the questionnaire.
A total of 4.9% of patients with a haemorrhagic stroke were users of caffeine-containing medicines compared to 2.3% of controls (adjusted odds ratio [aOR] 2.28; 95% CI 1.41-3.69), indicating that use of caffeine-containing medicines is associated with an increased risk of haemorrhagic strokes. The aOR was higher in those individuals who drank less than one cup of coffee per day (2.95; 95% CI 1.45-5.98). For caffeine-free medicines (drugs with the same indication as the caffeine-containing medicines – pain relievers, cold medicines, fatigue restoratives), the aOR was 0.71 (95% CI 0.39-1.28).
The study has a number of important limitations that should be considered when interpreting the results:
• It included patients who had the mental capacity to respond to an interview – so excluding those with severe neurological deficits or those who died before reaching hospital.
• The selection of non-population-based controls risks the possibility that selection may be related to some of the factors under study
• Information on potential confounders and drug exposure was collected after the event and recall bias may have affected results
• Information on the use of non-prescription medication depended on the participant’s memories
• Temporal-precedence bias may have affected the study (e.g. caffeine-containing medicines may have been used to treat the early symptoms of a haemorrhagic stroke)
Although the results are suggestive of a link, they do not prove causality and further research in this area is required.