Articles

The new approach to work

Methods

Data source

The data were drawn from the National Inpatient Sample and the Nationwide Inpatient Sample (NIS), the Healthcare Cost and Utilization Project (HCUP), and the Agency for Healthcare Research and Quality (AHRQ) (11,12). The NIS is the largest collection of all-payer data on inpatient hospitalizations in the United States. The dataset represents an approximate 20% stratified sample of all inpatient discharges from U.S. hospitals. The database provides de-identified information for each hospital stay. This information includes patient-level and hospital-level factors: patients’ demographic characteristics; primary and secondary diagnoses and procedures; AHRQ comorbidities; length of stay (LOS); hospital region; hospital teaching status; hospital bed size; and cost of hospitalization. National estimates can be calculated using the patient-level and hospital-level sampling weights that are provided by the NIS. For the purpose of this study, we obtained data for the years 2000 to 2013.

Study patients and variables

The International Classification of Diseases-9th Revision-Clinical Modification (ICD-9-M) is used for reporting diagnoses and procedures in the NIS database. For each index hospitalization, the database provides a principal discharge diagnosis and a maximum of 14 or 24 additional diagnoses (depending on the year), in addition to a maximum of 15 procedures. We identified patients 18 years of age or older who underwent catheter ablation (ICD-9-CM code 37.34) during 2000 to 2013. From this cohort of patients, we selected only those discharges that had 1 of the following arrhythmias as its principal diagnosis: 1) AF (ICD-9-CM code 427.31); 2) AFL (ICD-9-CM code 427.32); 3) SVT (ICD-9-CM codes 427.0, 427.89, 426.7, and 426.89); 4) VT (ICD-9-CM code 427.1). These are all well-known indications for ablation. Furthermore, we excluded patients with any of the following cardiac procedures during the index hospitalization, to avoid attributing their complications to the ablation procedure: 1) pacemaker implantation (ICD-9-CM 00.50, 00.52, 00.53, 37.71-37.79); 2) implantable cardioverter-defibrillator insertion (ICD-9-CM 37.94-37.98, 00.51, 00.54); and 3) open surgical ablation (ICD-9-CM 37.33). Detailed information on exclusion criteria are provided in the Online Appendix.


Study outcomes

We identified the common in-hospital complications of catheter ablation by using the ICD-9-CM diagnosis and procedures codes. These complications include the following: 1) cardiac complications (i.e., post-operative cardiac block, myocardial infarction); 2) pericardial complications (tamponade, hemopericardium, pericarditis, and pericardiocentesis); 3) vascular complications (arteriovenous fistula, blood vessel injury, accidental puncture, injury to the retroperitoneum, vascular complications requiring surgery, and other iatrogenic vascular complications); 4) post-operative hemorrhage or hematoma (including post-operative hemorrhage requiring blood transfusion); 5) post-operative stroke or TIA (transient ischemic attack); 6) pneumothorax or hemothorax; 7) diaphragm paralysis; 8) infections (fever, septicemia, and post-procedural aspiration pneumonia); and 9) in-hospital deaths. All codes used in identifying complications are shown in Online Table 1.

The following patient-related demographic features were collected from the database: age, sex, and race. The hospital identification number (HOSPID) was used to calculate hospital volumes. We defined high-volume hospitals as those with 100 or more annual catheter ablation procedures, medium-volume hospitals as those with 50 to 100 annual procedures, and low-volume hospital as those having fewer than 50 annual procedures. Associated comorbidities were identified by measures from the AHRQ. For the purpose of calculating the Deyo-Charlson comorbidity index (Deyo-CCI), an additional list of comorbidities was identified from the database by using ICD-9-CM codes (Online Table 2). Deyo-CCI is a modification of the Charlson comorbidity index. It contains 17 comorbid conditions (13). Higher Deyo-CCI indicates more severe comorbid conditions, and is an indicator of patient mortality 1 year after admission.

Statistical analysis

Trend weight files (TRENDWT) provided by the AHRQ were used to reflect national estimates. Detailed information on the implementation of sampling weights is provided in the Online Appendix. The chi-square test and the Wilcoxon rank sum test were used to compare categorical variables and continuous variables, respectively. Trends for continuous variables were tested using the nonparametric test for trend by Cuzick (14).

To account for hospital-level clustering of discharges, we generated a 2-level mixed-effects logistic regression model to identify independent predictors of complications. Congruent with the Healthcare Cost and Utilization Project NIS design, the hospital identification number was used as a random effect with patient-level factors clustered within hospital-level factors. Candidate variables included patient-level characteristics, Deyo-CCI, hospital-level factors, and hospital volume. For all analyses, we used survey estimation in Stata/SE software version 14.1 (StataCorp LP, College Station, Texas) to account for the complex survey design of the NIS database. A p value <0.05 was considered significant.

Results

After exclusion criteria were applied, a total of 108,273 unweighted observations were included in the final analysis. This represents an estimated 519,951 (95% confidence interval [CI]: 475,702 to 564,200) patients who underwent in-hospital catheter ablation procedures from 2000 to 2013. Among these patients, the annual number of ablation procedures increased by 2.5-fold from 18,096 in 2000 to 44,465 in 2013 (p = 0.002). More than one-half of the ablations were performed in low- and medium-volume hospitals (57.64%). There was an increase in the estimated number of hospitals performing ablations from 156 in 2000 to 216 in 2011 (p = 0.002). The percentage of high-volume hospitals showed a 2.5-fold increase from 15.19% in 2000 to 53.9% in 2011 (p = 0.016). Overall, AF was the most common indication for catheter ablation during these years (n = 190,398; 36.62%), followed by SVT (n = 159,895; 30.75%) (Figure 1). The increase in the total number of inpatient ablation cases over the study period was primarily the result of the increase in AF ablation, a nearly 10-fold rise from 2,644 procedures in 2000 to 21,345 procedures in 2013 (p = 0.001). Significant increases were also seen in the annual number of inpatient ablation procedures for AFL and VT: AFL from 4,046 in 2000 to 10,465 in 2013; and VT from 1,951 in 2000 to 4,735 in 2013 (Figure 2).

Back to blog