渡邉 英一

BACKGROUND The incidence of various cardiovascular diseases is known to exhibit seasonal variations, but seasonal patterns of paroxysmal atrial fibrillation (AF) have not been well characterized. OBJECTIVE The objective of this study was to determine whether seasonal variation affects the incidence of paroxysmal AF and whether this pattern is affected by patient age. METHODS We identified 258 paroxysmal AF episodes in 237 patients (age 65 14 years, mean standard deviation; age range 16-95 years) among 12,390 consecutive 24-hour Hotter electrocardiogram recordings obtained from 2001 to 2005 at our institute. Seasonal variations were analyzed by both month and by season. The relative risk (RR) of AF for each period was determined as being high or low in relation to the overall mean incidence. The association among clinical covariates and risk of paroxysmal AF was tested by logistic regression analysis. RESULTS The incidence of paroxysmal AF was highest in September (RR = 1.40, 95% confidence interval [CI] 1.36-1.44) and lowest in June (RR = 0.52, 95 to CI 0.50-0.54), with an RR difference of-63 /0 (P <.001) among all patients. Patients aged >= 65 years demonstrated a peak incidence in September (RR = 1.46, 95 to CI 1.41-1.51) and a minimum in June (RR = 0.55, 95% CI 0.52-0.58), while those aged <65 years showed a peak incidence in December (RR = 1.33, 95% CI 1.27-1.39) and a minimum in June (RR = 0.49, 95% CI 0.45-0.53). The incidence of paroxysmal AF also showed an autumn peak (RR = 1.21, 95% CI 1.16-1.27) and a summer minimum (RR = 0.66, 95% CI 0.62-0.70), with an RR difference of 53% (P <.001) among all patients. This seasonal variation in paroxysmal AF did not differ between patients of different age ranges. Clinical covariates including underlying disease or medications did not influence the monthly or seasonal variation in paroxysmal AF. There was a significant inverse relationship between the incidence of paroxysmal AF and the length of daylight in patients aged <65 years (r = -0.57, P <.05). CONCLUSION There was a significant seasonal variation in paroxysmal AF, with maximum and minimum incidences in autumn and summer, respectively, and this pattern was not age dependent.

Backgrounds: Cardioversion for atrial fibrillation (AF) is the most effective treatment for the restoration of sinus rhythm (SR). Recently, an elevated level of hs-CRP has been shown to be associated with AF burden, suggesting that inflammation increases the propensity for persistence of AF. We examined whether the level of high-sensitivity C-reactive protein (hs-CRP) was predictive of the outcome of cardioversion for AF. Methods and results: One hundred and six patients with a history of symptomatic AF lasting >= 1 day (age 63 14 years, mean +/- S.D.) underwent cardioversion. Echocardiography and hs-CRP assay were performed immediately prior to cardioversion. SR was restored in 84 patients (79%). By using selected cutoff values, multiple discriminant analysis revealed significant associations between successful cardioversion and a shorter duration of AF (AF duration <= 36 days, odds ratio (OR), 0.98; 95% confidence interval (CI), 0.97-0.99), smaller left atrial diameter (left atrial diameter <= 40 mm, OR 0.82, 95% CI 0.71-0.94), better-preserved left ventricular ejection fraction (left ventricular ejection fraction >= 60%, OR 0.92, 95% Cl 0.86-0.99), and lower hs-CRP level (hs-CRP <= 0.12 mg/dL, OR 0.33, 95% Cl 0.21-0.5 1). During a follow-up period of 140 144 days, AF recurred in 64 patients (76%). By using a cutoff value of hs-CRP >= 0.06 mg/dL, Cox proportional-hazards regression model found that only hs-CR-P level was an independent predictor of AF recurrence (OR 5.30, 95% Cl 2.46-11.5) after adjustment for coexisting cardiovascular risks. When patients were divided by the hs-CRP level of 0.06 mg/dL, percentage of maintenance of SR below and above the cutoff was 53% and 4%, respectively (log-rank test, p < 0.0001). Conclusions: hs-CRP level determined prior to cardioversion represents an independent predictor of both successful cardioversion for AF and the maintenance of SR after conversion. (c) 2005 Elsevier Ireland Ltd. All rights reserved.