Monday preference in onset of ischemic stroke.

The onset of stroke has a specific temporal pattern, characterized by a higher frequency in winter and in the mornings. According to a meta-analysis of more than 11,000 patients, an estimated 37% of strokes occurred during morning hours. Variability of the day of week has not been extensively investigated, however. To verify a temporal weekly pattern, we reviewed all cases of stroke admitted to our hospital during a 4-year period. Methods Ferrara is a town in northeastern Italy with a population of 150,000. The only hospital is St. Anna Hospital, the teaching hospital for the University of Ferrara. House calls are made by family physicians during the daytime hours and by Emergency Department physicians during nighttime hours and on holidays at no charge to patients. Key physicians, including neurologists and neurosurgeons, are active 24 hours a day. We ascertained the occurrence of stroke from January 1, 1994, to December 31, 1997. The diagnosis of stroke, which was always made by a neurologist, was defined according to World Health Organization criteria as rapidly developing clinical symptoms or signs of focal, and at times global, loss of cerebral function, with symptoms lasting more than 24 hours or leading to death, with no apparent cause other than that of vascular origin. In all patients, diagnostic and laboratory investigations included computed tomography (CT), blood tests, 12-lead electrocardiography, chest radiography, carotid duplex imaging, transcranial doppler imaging, cerebral angiography, echocardiography (transthoracic or transesophageal), and assessment of prothrombotic syndromes were done. We observed 1610 consecutive cases of stroke (987 [61%] men). The mean (± SD) age of patients was 70 ± 12 years. There were 1,395 (87%) ischemic strokes and 315 (13%) hemorraghic strokes. Stroke onset was defined as the earliest time the patient or witness noted definite neurologic symptoms or signs; this information was obtained from patients, their relatives, or bystanders. Precise determination of the day of onset was possible in all cases. We performed the main statistical analysis using a partial Fourier series, using a least-squares minimization fitting, and fitting a cosine function to the data by means of a regression method using a 168-hour period. Analyses were made using Chronolab software. This method selects the harmonic, or the combination of harmonics, that best explains the variance of the data. The percentage of overall variability of the data about the arithmetic mean that is attributable to the fitted rhythmic function estimates goodness of fit; and an F statistic is used to test the hypothesis of zero amplitude. Significance levels were set at P