Burden of arrhythmia in hospitalized HIV patients

Abstract Background The improved life expectancy observed in patients living with human immunodeficiency virus (HIV) infection has made age‐related cardiovascular complications, including arrhythmias, a growing health concern. Hypothesis We describe the temporal trends in frequency of various arrhythmias and assess impact of arrhythmias on hospitalized HIV patients using the Nationwide Inpatient Sample (NIS). Methods Data on HIV‐related hospitalizations from 2005 to 2014 were obtained from the NIS database using International Classification of Diseases, 9th Revision (ICD‐9) codes. Data was further subclassified into hospitalizations with associated arrhythmias and those without. Baseline demographics and comorbidities were determined. Outcomes including in‐hospital mortality, cost of care, and length of stay were extracted. SAS 9.4 (SAS Institute Inc., Cary, NC) was utilized for analysis. A multivariable analysis was performed to identify predictors of arrhythmias among hospitalized HIV patients. Results Among 2 370 751 HIV‐related hospitalizations identified, the overall frequency of any arrhythmia was 3.01%. Atrial fibrillation (AF) was the most frequent arrhythmia (2110 per 100 000). The overall frequency of arrhythmias increased over time by 108%, primarily due to a 132% increase in AF. Arrhythmias are more frequent among older males, lowest income quartile, and nonelective admissions. Patients with arrhythmias had a higher in‐hospital mortality rate (9.6%). In‐hospital mortality among patients with arrhythmias decreased over time by 43.8%. The cost of care and length of stay associated with arrhythmia‐related hospitalizations were mostly unchanged. Conclusions Arrhythmias are associated with significant morbidity and mortality in hospitalized HIV patients. AF is the most frequent arrhythmia in hospitalized HIV patients.

pitalizations. This has led to an overall improved life expectancy in the HIV patient population. 2 However, in the aging HIV-infected population, cardiovascular complications such as hypertension, coronary artery disease (CAD) and heart failure have become a growing health concern. [3][4][5][6] Other contributors such as the HIV infection itself, immune dysfunction, chronic inflammation, ART exposure and toxicity are also implicated in heart disease and can lead to complications such as myocardial infarction and cardiomyopathy. 7,8 In addition, arrhythmias are also important contributors to cardiovascular morbidity and mortality in patients with HIV. 9,10 The temporal trends of the frequency and outcome of arrhythmias in patients with HIV have not been adequately described.

| METHODS
The primary objective of this study was to describe the temporal trends in the frequency of arrhythmias among hospitalized HIV patients. Secondary objectives included identifying comorbidities associated with arrhythmias in this cohort as well as determine the outcomes associated with arrhythmias including in-hospital mortality, length of stay, and cost of care.

| Data source
The data were obtained from the Nationwide Inpatient Sample (NIS) data set from 2005 to 2014. 11 The NIS is a nationally representative survey of hospitalizations conducted by the Healthcare Cost and Utilization Project in collaboration with the participating states. It is the largest inpatient data set in the United States and includes a sample of US community hospitals that approximates 20% of all US community hospitals. 12 No institutional review board approval was sought because of the publicly available de-identified data set used in this research.

| Study population
Our target population consisted of HIV-related hospitalizations from January 1, 2005 to December 21, 2014. We included hospital admissions with a diagnosis of HIV infection in primary and secondary diagnostic field during our study period. We subclassified this group into hospitalizations with associated cardiac arrhythmia and those without arrhythmia for trend analysis using ICD codes for cardiac arrhythmias such as ventricular tachycardia (VT), ventricular fibrillation (VF), supraventricular tachycardia (SVT), AF, and atrial flutter (AFL). Interventions such as ICD implantation, use of vasopressors, cardiac catheterization, endotracheal intubation, and CPR were identified.

| Definition of variables
We used NIS variables to identify patient level and hospital level variables. We divided age into 4 subgroups: 18 to 49 years of age, 50 to 64 years of age, 65 to 79 years of age, and 80 years of age and older.
We defined the severity of comorbid conditions by using the Deyo modification of the Charlson Comorbidity Index. Co-morbidities associated with hospitalization for HIV infection were identified using AHRQ comorbidity measures, that is, by using ICD-9-CM diagnoses and the Diagnosis Related Group (DRG) in effect on the discharge date. 13

| In hospital mortality, cost, and length of stay
In-hospital mortality was defined as death from any cause during the same hospital stay. LOS was already provided by the Healthcare Cost and Utilization Project (HCUP) for each entry. The HCUP NIS contains data on total charges for each hospital in the databases, which represents the amount that hospitals billed for services. To calculate estimated cost of hospitalizations, the NIS data were merged with Cost to Charge Ratios (CCR) available from HCUP. 14,15 Using the merged data elements from the CCR files and the total charges reported in the NIS database, we converted the hospital total charge data to cost estimates by simply multiplying total charges with the appropriate CCR.
These costs are essentially standardized, can be measured across hospitals, and are used for the remainder of this report.

| Demographics
Baseline characteristics of HIV-related hospitalizations with arrhythmias are summarized in Table 2

| Hospital course in patients with any arrhythmia
Hospitalization course among patients with arrhythmias is summarized in Table 2  in the pathogenesis of AF in HIV infected patients. 24 The severity of the HIV infection also correlates with risk of developing AF. A previous analysis identified low CD4+ cell count and high HIV RNA viral load as independent variables for the development of AF in HIV patients. 9 The exact mechanism underlying this correlation is difficult to establish given that many patients in the HIV population share similar risk factors for AF, such as CAD and heart failure, as demonstrated by our study.
Additionally, components of highly active antiretroviral therapy (HAART) such as protease inhibitors are associated with development of metabolic syndrome, which is a risk factor for AF. 25 In our study, the frequency of malignant arrhythmias such as VF and

| CONCLUSION
Among hospitalized HIV patients, cardiac arrhythmias are associated with significant morbidity and mortality. AF is the most frequent arrhythmia among hospitalized HIV patients. The presence of arrhythmias is associated with adverse outcomes in HIV patients, including a higher in-hospital mortality rate and cost of care. The in-hospital mortality among patients with any arrhythmia is significant but has decreased over the years.

CONFLICT OF INTEREST
The authors declare no potential conflict of interests.

DATA AVAILABILITY STATEMENT
The following information was supplied regarding data availability: The NIS database raw files were purchased through online HCUP (health care cost and utilization project) distributor. All purchasers and users of HCUP data must complete the online Data Use Agreement (DUA) training so that they are familiar with the rules and restrictions for the use of HCUP data. If this raw data is made public, then the likelihood of maintaining standards to complete the DUA training might be violated. Moreover, anyone who uses the database files we purchased must complete an HCUP Data Use Agreement form (a form which the owner of the data has to generate and acknowledge). Therefore, given these limitations, we have provided raw data for review purposes and not for publishing (i.e., to be shared with the public). Additionally, as per AHRQ-HCUP, before the reviewers/editors may access any HCUP data (the raw files), they are required to com-