Circadian variation in the timing of stroke onset: a meta-analysis.
Level 3 - non-randomized controlled study
Meta-analysis of observational/cohort studies
PubMed 9596248 · doi:10.1161/01.str.29.5.992
What was done
A meta-analysis was conducted on 31 publications reporting the circadian timing of 11,816 strokes, subdivided by stroke type based on symptom onset time. When precise timing was missing, strokes were distributed evenly across time intervals to bias toward the null hypothesis of no circadian variation.
What was found
All stroke subtypes demonstrated significant (P < 0.001) circadian variation across 3-, 4-, or 6-hour intervals. For stroke of all types, there was a 49% increase (95% CI, 44% to 55%) in onset between 6 AM and noon compared with a uniform distribution, representing a 79% increase (95% CI, 72% to 87%) over the normalized risk of the other 18 hours. Between midnight and 6 AM, strokes were 29% lower than uniform expectation (a 35% decrease compared to the other 18 hours). Significantly elevated risk between 6 AM and noon occurred across all subtypes: 55% for 8,250 ischemic strokes, 34% for 1,801 hemorrhagic strokes, and 50% for 405 transient ischemic attacks.
Why it matters
The study quantitatively synthesizes evidence confirming a consistent morning peak in the onset of both ischemic and hemorrhagic strokes, paralleling patterns seen in myocardial infarction.
Limits
The analysis relies on pooled observational reports that may suffer from inaccurate timing of onset, particularly for unwitnessed nocturnal strokes noticed upon waking. Underlying patient characteristics, medications, and potential confounding factors across individual studies are not detailed in the abstract.
Cited by
- context Stroke rates between 6:00 AM and 10:00 AM are approximately 80% higher than during the rest of the day.