Takotsubo cardiomyopathy (TC) is believed to be an increasingly diagnosed syndrome; however, data on its incidence are limited. The purpose of this study was to determine the reported incidence of TC in the United States and to examine its trend over several years. Data was obtained from the Nationwide Inpatient Sample, created by the Agency for Healthcare Research and Quality, for each of the years from 2006 to 2012. Hospital discharges with principal diagnosis of TC, identified using Internal Classification of Diseases, Ninth Revision , code 429.83, were included. We tabulated estimated total numbers of discharges, incidence per 100,000 persons, mean length of stay, inhospital death rates, and diagnoses stratified by age group and gender. The reported incidence of TC based on principal diagnosis at hospital discharge increased significantly over the study period, with 315 cases ± 43 (standard error) in 2006 and 6,230 cases ± 232 (standard error) in 2012 (p <0.001 for trend). Mean length of hospital stay was stable over the study period (3.4 days in 2006 vs 3.6 days in 2012; p = 0.74 for trend). The diagnosis was most frequent in patients aged 65 to 84 years (50% of all diagnoses in 2012), followed by those aged 45 to 64 years (39% of all diagnoses in 2012). Women accounted for >90% of diagnoses throughout the study period. In conclusion, the reported incidence of TC has increased significantly from 2006 to 2012, most likely because of increasing recognition of the syndrome.
Although previously thought to be rare, takotsubo cardiomyopathy (TC), or stress-induced cardiomyopathy, is an increasingly recognized syndrome. Some studies have reported an incidence of 1.7% to 2.5% in patients presenting with acute coronary syndrome (ACS). Other studies more recently reported that TC may be even more common, even constituting up to 10% of women presenting with ACS. However, no studies to date have ascertained the incidence of TC in the US population. The correct diagnosis of TC has important implications for both management and prognosis. The aim of this study was to examine the reported incidence of TC in the United States and to examine its trend over several years.
Methods
Data were obtained from the Nationwide (now renamed National) Inpatient Sample (NIS), created by the Agency for Healthcare Research and Quality. The NIS is the largest publicly available all-payer inpatient health care database in the United States. We obtained data for each year from 2006 to 2012. Before 2012, the NIS database drew from 100% of hospital discharges from approximately 1,000 hospitals. Beginning in 2012, the database represents a 20% sample of discharges from all community hospitals participating in HCUP (Healthcare Cost and Utilization Project). The NIS covers all patients, including subjects covered by Medicare, Medicaid, or private insurance and those who are uninsured.
Using the NIS data, we obtained data for hospital discharges with principal diagnosis of TC, identified using Internal Classification of Diseases, Ninth Revision , code 429.83. We elected not to obtain data for discharges with any diagnosis of TC but only with a principal diagnosis of TC, to avoid overestimating the number of cases so that only patients admitted with TC-associated chief complaints would be included. It was at the discretion of each hospital to diagnose TC; therefore, information regarding the methods of diagnosis was not standardized and was not ascertainable from the database used for this study.
Total number of discharges with the principal diagnosis of TC was tabulated for each of the years from 2006 through 2012; these data were then used to calculate the estimated incidence (number/year/100,000 subjects). We also evaluated mean length of hospital stay and inhospital mortality associated with a principal diagnosis of TC. In addition, we examined the breakdown by gender and age in each year. Patient ages were grouped as follows: 18 to 44 years, 45 to 64 years, 65 to 84 years, and ≥85 years. The HCUP database provides standard error measurements associated with its estimates. Differences between years were compared using analysis of variance testing for linear trend over time, with a p value of <0.05 considered significant.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
