Rapid Communications
Trends in Hospitalizations for Cardiac Sarcoidosis in the United States, 2005–2011
2017 Volume 81 Issue 11 Pages 1739-1741
Details
2017 Volume 81 Issue 11 Pages 1739-1741
Background: Cardiac sarcoidosis (CS) is a life-threatening disease that is frequently under-diagnosed.
Methods and Results: We used a nationwide inpatient sample to identify CS patients from 2005 to 2011 in the USA. The annual admissions of CS increased from 1,108 in 2005 to 2,182 in 2011, representing a 2-fold rise over a short time. The proportions of CS patients with severe comorbidities, ventricular tachycardia, ventricular fibrillation and heart failure all increased from 2005 to 2011. However, the in-hospital mortality rate declined.
Conclusions: An increasing trend of CS was observed. Cardiologists should notice that CS is not as rare as thought.
Cardiac sarcoidosis (CS) is a life-threatening disease that is diagnosed clinically in only about 5% of patients with sarcoidosis, but is detected in up to 25% of patients with systemic sarcoidosis at autopsy and with advanced imaging.1 Recent studies on the epidemiology of CS suggest increasing disease prevalence, possibly related to advanced imaging technology and/or greater awareness of the disease.2 The largest such study, a nationwide survey of Finland, has reported that the detection rate of CS increased more than 20-fold between 1988 and 2014.3 However, contemporary data on trends in the clinical characteristics, hospitalization and mortality rates of CS in the USA are currently lacking. The objective of this study was to estimate these trends using a nationally representative database of hospital admissions.
Data from the Healthcare Cost and Utilization Project Nationwide Inpatient Sample (HCUP-NIS) administrative files for the years 2005–2011 were analyzed. The NIS is the largest publicly available all-payer inpatient care database in the USA. It is composed of discharge-level data from approximately 8 million hospitalizations per year and approximates a stratified sample of 20% of inpatient admissions in the USA.4 We identified patients using the International Classification of Diseases-9th Revision (ICD-9) code 135 (sarcoidosis), combined with code 425.8 (cardiomyopathy in other diseases including sarcoidosis and sarcoid heart muscle disease). Patients’ characteristics were presented in total and by different years. We used sampling weights to estimate the annual number of admissions and all percentages presented by groups over time on a national level. A two-tailed P-value <0.05 was considered significant. Statistical analyses were performed using Stata/SE 14.0 (StataCorp, College Station, TX, USA). Certification to use this de-identified HCUP data was obtained from the Committee on Human Research, University of California, San Francisco.
We found 13,045 hospitalizations for CS from 2005 to 2011 (Table). Annual admissions increased from 1,108 in 2005 to 2,182 in 2011, representing a 2-fold rise over a short time. The majority of CS patients were 35–64 years old, and 48% were classified as African-American. The prevalence of atrioventricular block, ventricular tachycardia and heart failure were 6.1%, 22.5% and 50.6%, respectively. The overall in-hospital mortality rate among CS patients was 2.3%. The proportion of CS patients who were male increased from 39.3% in 2005 to 50.1% in 2011 (P for trend <0.001). The proportions of CS patients with severe comorbidities (index ≥2), ventricular tachycardia, ventricular fibrillation, and heart failure all increased from 2005 to 2011 (P for trend <0.001). In contrast, in-hospital deaths declined (P=0.0462). Implantation of implantable cardioverter-defibrillator (ICD) during hospitalization decreased from 2.1% in 2005 to 1.3% in 2010 and 1.8% in 2011 (P=0.0037).
| Overall | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | P-value for trend |
|
|---|---|---|---|---|---|---|---|---|---|
| Discharges, weighted | 13,045 | 1,108 | 1,108 | 1,426 | 1,615 | 1,661 | 1,930 | 2,182 | |
| Age (years), % | |||||||||
| <34 | 5.5 | 10.3 | 3.4 | 8.2 | 8.0 | 6.4 | 2.8 | 4.7 | <0.001 |
| 35–49 | 35.2 | 35.1 | 40.7 | 37.2 | 33.6 | 35.3 | 36.3 | 34.9 | 0.1096 |
| 50–64 | 42.6 | 45.1 | 33.3 | 40.9 | 41.3 | 38.6 | 46.8 | 45.7 | <0.001 |
| ≥65 | 16.7 | 9.6 | 22.5 | 13.7 | 17.2 | 19.8 | 14.1 | 14.7 | 0.8295 |
| Sex, % | |||||||||
| Male | 49.1 | 39.3 | 49.4 | 46.3 | 44.2 | 54.5 | 50.1 | 50.1 | <0.001 |
| Female | 50.9 | 60.7 | 50.6 | 53.7 | 55.8 | 45.5 | 49.9 | 49.9 | <0.001 |
| Race, % | |||||||||
| White | 32.4 | 27.1 | 27.1 | 30.9 | 27.2 | 38.1 | 33.3 | 36.6 | <0.001 |
| Black | 48.0 | 41.8 | 43.5 | 44.0 | 46.1 | 37.4 | 53.2 | 55.2 | <0.001 |
| Hispanic | 1.7 | 0.4 | 1.2 | 0.8 | 2.8 | 1.7 | 2.4 | 1.3 | 0.0076 |
| Other* | 3.0 | 2.1 | 1.6 | 3.1 | 2.4 | 4.9 | 1.3 | 1.8 | 0.3918 |
| Income percentile, % | |||||||||
| 0–25th | 30.4 | 30.5 | 27.7 | 26.4 | 36.8 | 26.4 | 33.5 | 29.3 | 0.2938 |
| 26–50th | 21.2 | 15.2 | 29.7 | 20.6 | 22.6 | 19.0 | 21.4 | 20.6 | 0.5935 |
| 51–75th | 23.1 | 25.2 | 19.6 | 22.1 | 17.1 | 30.7 | 23.0 | 22.1 | 0.3194 |
| 76–100th | 22.6 | 25.1 | 22.2 | 27.1 | 21.8 | 19.9 | 19.7 | 25.2 | 0.1039 |
| Primary payer, % | |||||||||
| Medicare | 36.5 | 31.6 | 42.9 | 34.2 | 36.5 | 34.4 | 34.8 | 36.1 | 0.7657 |
| Medicaid | 14.5 | 13.9 | 9.8 | 14.1 | 11.5 | 18.7 | 14.0 | 15.3 | <0.001 |
| Private | 41.9 | 49.0 | 39.8 | 44.6 | 44.4 | 39.2 | 44.1 | 42.2 | 0.0191 |
| Self-pay/no charge/other | 6.7 | 5.5 | 7.5 | 6.7 | 7.6 | 5.9 | 7.1 | 5.9 | 0.7007 |
| Charlson/Deyo comorbidity index, % | |||||||||
| 0 | 22.9 | 32.9 | 25.0 | 27.9 | 24.4 | 23.3 | 19.5 | 20.5 | <0.001 |
| 1 | 31.6 | 36.3 | 40.7 | 30.4 | 31.0 | 32.6 | 31.3 | 26.9 | <0.001 |
| ≥2 | 45.5 | 30.8 | 34.3 | 41.7 | 44.5 | 44.1 | 49.2 | 52.6 | <0.001 |
| Cardiac manifestations, % | |||||||||
| 2nd- or 3rd-degree AV block | 6.1 | 6.9 | 6.0 | 8.3 | 3.2 | 6.1 | 5.5 | 6.4 | 0.2961 |
| Ventricular tachycardia | 22.5 | 17.2 | 18.4 | 21.5 | 22.4 | 20.6 | 21.5 | 27.7 | <0.001 |
| Ventricular fibrillation | 2.2 | 0.9 | 1.3 | 2.4 | 2.0 | 3.1 | 2.4 | 2.5 | <0.001 |
| Cardiac arrest | 1.5 | 1.2 | 1.3 | 0.7 | 0.9 | 0.9 | 2.6 | 1.9 | <0.001 |
| Premature beats | 3.2 | 2.2 | 1.8 | 1.9 | 1.6 | 4.6 | 5.0 | 4.5 | <0.001 |
| Heart failure | 50.6 | 40.8 | 50.0 | 47.2 | 49.8 | 51.3 | 51.7 | 55.6 | <0.001 |
| In-hospital procedures, % | |||||||||
| ICD implantation | 2.1 | 2.1 | 2.5 | 3.9 | 2.1 | 2.4 | 1.3 | 1.8 | 0.0037 |
| Pacemaker implantation | 11.3 | 10.4 | 10.3 | 9.6 | 10.9 | 12.1 | 14.1 | 11.7 | <0.001 |
| In-hospital death, % | 2.3 | 2.4 | 2.6 | 2.7 | 3.7 | 1.8 | 2.2 | 1.9 | 0.0462 |
Percentages may not sum to 100% given missing data. Overall, missing data were <4% except for race. *Includes Asian or Pacific Islander, Native American and other. AV, atrioventricular; ICD, implantable cardioverter-defibrillator.
We report contemporary trends of CS hospitalization and patients’ characteristics in the USA. As expected, CS admissions increased from 2005 to 2011. The demographic characteristics and clinical manifestations of CS patients also changed during this period, with increasingly frequent arrhythmia and heart failure but declining numbers of deaths.
This 2-fold increase in CS cases in the USA from 2005 to 2011 was remarkably consistent with the Finnish findings, where the number of CS cases also approximately doubled from 2003–2007 to 2008–2012.3 Al-Kindi et al reported a similar increasing trend of heart transplantation for CS.5 Overall, we found a higher percentage of African-Americans patients with CS in the present study, which is consistent with the higher prevalence of sarcoidosis found in this group.6 However, we observed lower percentages of both female patients and advanced atrioventricular block compared with previous studies in other countries.3,7,8 These differences in clinical characteristics might be explained by different sample approaches and geographies. Further studies are expected.
The limitations of our study are primarily related to the use of the NIS, a large administrative database. We did not have enough information to report the misdiagnosis percentage of CS. Also, the validity of ICD-9 codes for CS diagnoses is unknown. Because the ICD-9 code 425.8 also includes cardiomyopathy caused by other uncommon diseases such as amyloidosis or scleroderma, these data may have overestimated the frequency of CS in hospitalized patients. However, it is rare to simultaneously have systematic sarcoidosis and cardiomyopathy caused by other uncommon diseases. In addition, the sample did not include outpatients in the early stage of CS, which may also contribute to the higher percentage of heart failure we observed.
Overall, our finding that annual hospitalizations for CS have increased since 2005 in the USA confirmed previous studies and provides evidence for improved detection and awareness of this life-threatening disease.
None declared.
