Rapid Communications
Trend in the Prevalence of Atrial Fibrillation During the Past 15 Years in Iwate (Northeastern Area of Japan)
2017 Volume 81 Issue 10 Pages 1537-1539
Details
2017 Volume 81 Issue 10 Pages 1537-1539
Background: The trend in age-specific prevalence of atrial fibrillation (AF) in Japan has not been reported.
Methods and Results: Age-specific prevalence (40–49, 50–59, 60–69, 70–79 and 80–89 years old) of AF in Iwate Prefecture was determined in 1997, 2002, 2007 and 2012 (n=818,577). A positive linear trend in the prevalence of AF across calender years was observed only in males in their 60 s and 70 s. The direct age-standardized rate in males increased from 1.55% to 1.85%, while the rate in females remained around 0.5%.
Conclusions: The age-specific prevalence of AF has increased only in elderly males.
The prevalence of atrial fibrillation (AF) greatly increases with advancing age both in Western countries1,2 and in regions outside North America and Europe.3 Recently, several studies have revealed that not only aging society but also increasing age-specific prevalence of AF has contributed to the increasing number of individuals with AF in Western countries.4–6 On the other hand, the age-specific prevalence of AF has remained constant during the period from 1980 to 2000 in Japan.7 However, it is not known whether there has been an increase in the age-specific prevalence rate of AF in Japan in recent years. Thus, we attempted to estimate age-specific prevalence rates of AF in men and women using a dataset for community dwellers who underwent annual health checkups in Iwate Prefecture (northeastern area of Japan).
Data for individuals who had participated in 4 checkups (1997, 2002, 2007, and 2012) out of a total of 42 annual checkups and aged 40–89 years (n=818,577, accounting for 26% of the entire population of the prefecture) were used for analysis (Table). The study was approved by the Medical Ethics Committee of Iwate Health Service Association and conducted in accordance with the guidelines of the Declaration of Helsinki.
| Age category (years) |
1997 | 2002 | 2007 | 2012 | Trend P | ||||
|---|---|---|---|---|---|---|---|---|---|
| Total | AF (%) | Total | AF (%) | Total | AF (%) | Total | AF (%) | ||
| Male subjects | |||||||||
| 40–49 | 37,956 | 108 (0.28) | 34,851 | 107 (0.31) | 34,137 | 95 (0.28) | 33,491 | 109 (0.33) | 0.472 |
| 50–59 | 29,792 | 336 (1.13) | 34,495 | 390 (1.13) | 37,522 | 426 (1.14) | 34,835 | 370 (1.06) | 0.448 |
| 60–69 | 23,077 | 525 (2.27) | 22,170 | 564 (2.54) | 19,996 | 578 (2.89) | 24,959 | 737 (2.95) | <0.001* |
| 70–79 | 7,863 | 328 (4.17) | 11,702 | 496 (4.24) | 12,806 | 633 (4.94) | 11,819 | 666 (5.63) | <0.001* |
| 80–89 | 1,059 | 50 (4.72) | 1,467 | 82 (5.59) | 2,284 | 140 (6.13) | 2,333 | 139 (5.96) | 0.158 |
| 40–89 | 99,747 | 1,347 (1.34) | 104,685 | 1,639 (1.56) | 106,745 | 1,872 (1.75) | 107,437 | 2,021 (1.88) | <0.001* |
| 95% CI | (1.33–1.34) | (1.56–1.57) | (1.75–1.76) | (1.87–1.89) | |||||
| Entire population | 351,022 | 365,511 | 367,627 | 366,192 | |||||
| Participation rate | 28.4% | 28.6% | 29.0% | 29.3% | |||||
| Direct age- standardized rate (95% CI) |
1.55% (1.49–1.61) | 1.65% (1.57–1.72) | 1.81% (1.63–1.79) | 1.85% (Ref.) | <0.001** | ||||
| Female subjects | |||||||||
| 40–49 | 33,276 | 10 (0.03) | 28,502 | 7 (0.02) | 25,867 | 10 (0.04) | 23,947 | 4 (0.02) | 0.603 |
| 50–59 | 31,099 | 52 (0.17) | 32,150 | 42 (0.13) | 31,694 | 30 (0.09) | 26,006 | 28 (0.11) | 0.017 |
| 60–69 | 28,712 | 152 (0.53) | 28,811 | 153 (0.53) | 23,186 | 137 (0.59) | 23,155 | 124 (0.54) | 0.684 |
| 70–79 | 9,671 | 139 (1.44) | 14,463 | 186 (1.29) | 16,221 | 223 (1.37) | 13,940 | 222 (1.59) | 0.170 |
| 80–89 | 1,067 | 32 (3.00) | 1,577 | 47 (2.98) | 2,730 | 83 (3.04) | 2,921 | 110 (3.77) | 0.113 |
| 40–89 | 103,825 | 385 (0.37) | 105,503 | 435 (0.41) | 99,698 | 483 (0.48) | 89,969 | 488 (0.54) | <0.001* |
| 95% CI | (0.37–0.37) | (0.41–0.41) | (0.48–0.49) | (0.54–0.54) | |||||
| Entire population | 413,353 | 425,759 | 429,127 | 424,184 | |||||
| Participation rate | 25.1% | 24.8% | 23.2% | 21.2% | |||||
| Direct age- standardized rate (95% CI) |
0.51% (0.46–0.56%) | 0.47% (0.44–0.51%) | 0.50% (0.46–0.54%) | 0.54% (Ref.) | 0.768** | ||||
*P for trend by logistic regression with no adjustment. **P for trend by logistic regression after adjusting for age. AF, atrial fibrillation; CI, confidence interval.
Resting 12-lead ECG was performed in each of the participants. Medical doctors in the Iwate Health Service Association independently evaluated the ECG findings according to an original coding system. Cases of AF were determined on the basis of the presence of chronic or paroxysmal atrial fibrillation/flutter. Participants were divided into 5 age categories (40–49, 50–59, 60–69, 70–79 and 80–89 years). We determined sex- and age-specific prevalence rates (95% confidence intervals) of AF in each year. A linear trend in the age-specific prevalence rate of AF across the calender year was tested by logistic regression analysis. The direct age-standardized prevalence rate of AF in each year was estimated using the population in 2012 as a reference. A linear trend test was also performed for the prevalence of AF from 1997 to 2012 after adjusting for age by logistic regression analysis.
Table shows the number of subjects and prevalence of AF. In the male subjects, the range of prevalence rates of AF was 0.28–0.33% in subjects in their 40 s, 1.06–1.14% in subjects in their 50 s, 2.27–2.95% in subjects in their 60 s, 4.17–5.63% in subjects in their 70 s and 4.72–6.13% in subjects in their 80 s. In female subjects, the corresponding rates were 0.02–0.04%, 0.09–0.17%, 0.53–0.59%, 1.29–1.59% and 3.00–3.77%. Age-stratified linear trend tests revealed significant linear trends between prevalence of AF and calender year in male subjects aged 60–69 years and 70–79 years. Non-age-adjusted linear trend tests showed significant positive associations between prevalence of AF and calender year in both male and female subjects.
Figure S1A shows that prevalence rates of AF were all around 0.3% in subjects in their 40 s and around 1.10% in subjects in their 50 s, but from 1997 to 2012 the rates progressively increased from 2.27% to 2.95% in subjects in their 60 s and from 4.17% to 5.63% in subjects in their 70 s. The rate in subjects in their 80 s increased from 4.72% to 5.96%, but the difference was not statistically significant. Figure S1B shows that the prevalence rate of AF did not increase during the period in female subjects in their 40 s, 50 s, 60 s or 70 s. The rates in female subjects in their 80 s increased from 3.00% to 3.77%, but the difference was not statistically significant. Figure S1C shows the trend in age-standardized prevalence rates of AF, which in male subjects progressively increased from 1.55% to 1.85% during the period (P for trend <0.001), but remained around 0.5% in female subjects during the same period (P for trend=0.768).
Significant increases in the prevalence of AF from 1997 to 2012 were observed in male subjects in their 60 s and 70 s. Overall, the age-standardized prevalence rate of AF progressively increased from 1.55% to 1.85% in male subjects during the period, but there was no significant increase in the prevalence of AF among female participants in any age category, and the age-standardized prevalence rate of AF remained at about 0.5%.
Studies in European countries have shown that the age-specific prevalence of AF has been increasing during the past decade in both men and women aged 65 years or older.4,5 Among Medicare beneficiaries, non-sex-stratified analysis revealed that the age-specific prevalence of AF increased in every age category from 1993 to 2007.6 The age-specific prevalence of AF in elderly men in Iwate has also been increasing during the past 15 years, but the age-specific prevalence of AF has remained very low in women in Iwate.
The low prevalence of AF in female subjects might be attributable to a low prevalence of the predisposing factors to AF in the area or it might be attributable to underestimation of the prevalence of AF, because persons who did not participate in health checkups were probably in a poor health and might have had AF. This might have reduced the number of cases of AF and is a limitation of this study. Some cases of paroxysmal AF might have been underdiagnosed and this is also an important limitation.
Our data indicated that the age-standardized prevalence of AF had increased only in male subjects. Thus, the associations between known predisposing factors to AF1 and prevalence of AF should be examined using individual datasets. Whether low prevalence rates of predisposing factors are contributing to the low prevalence rates of AF in the area, especially in female individuals, should also be examined.
Supplementary File 1
Figure S1. Age-specific prevalence of atrial fibrillation (AF) in (A) male participants and (B) female participants. Age-standardized prevalence rate of AF (C).
Please find supplementary file(s);
http://dx.doi.org/10.1253/circj.CJ-17-0535
