The study was aimed at further investigating the circadian and circann
ual patterns of stroke onset. Study design and type of participants: 9
77 strokes (475 in men and 502 in women) concerning 926 subjects (457
men and 469 women) admitted to Ferrara Hospital in two calendar years
(1990-1991), were prospectively investigated. The strokes were classif
ied as based on cerebral infarction (CI), transient ischemic attack (T
IA) and cerebral hemorrhage (CH: subarachnoid and intracerebral hemorr
hage). Two statistical models of analysis were used. The assessment of
circadian and circannual periodicity was performed utilizing the sing
le cosinor method. A separate analysis was performed after distributio
n of events into 6-hour intervals, and chi-square test for fit was app
lied to the number of observed versus expected cases. The majority of
strokes occurred in the morning between 7 a.m. and noon (35% of cases)
and the hypothesis of a uniform distribution of the time onset was re
jected on the basis of the chi-square for all subtypes of stroke. A ci
rcadian rhythm was found for CI and TIA with acrophase at the 11.56 an
d 12.41 respectively. Also a circannual periodicity was found for CI w
ith a prevalent peak in October. The spectral analysis detected a circ
adian cycle for CH having a period of 4 h, and a circannual cycle for
TIA with a period of 4 months. This study confirms that stroke is a hi
gh-chronorisk disease, with specific circadian and circannual rhythms.
This is very important for a better understanding and control of the
underlying factors and in terms of prevention.