Methods

The subjects were enrolled from the National Cardiovascular Data Registry’s ICD Registry. Collection of data in the ICD Registry has been mandated by the Centers for Medicare and Medicaid Services as a stipulation of payment for all primary prevention ICD implantations in the United States since 2006. The registry captures detailed clinical, demographic, and procedural information and includes medications prescribed at discharge using standardized data elements and definitions. The Data are submitted by the participating hospitals using certified software. Data quality is examined using a formal data quality reporting process. Institutions with submissions passing the inclusion and exclusion quality criteria for data completeness were included in the data analytic file used for the present analysis. All patients in the ICD Registry who had received a first time (de novo) ICD (including cardiac resynchronization therapy-defibrillator [CRT-D]) from January 1, 2006 to September 30, 2011 (n = 500,995) were included in the present analysis.

The main outcome of interest was the prescription of an antiarrhythmic medication at discharge, classified as follows: class IA (quinidine, disopyramide, procainamide), class IB (mexiletine), class IC (propafenone, flecainide), and class III (amiodarone, sotalol, dofetilide). Information on dronedarone was not routinely captured by the data collection forms used for the ICD Registry. β Blockers and calcium channel blockers (Vaughan-Williams class II and IV, respectively) were not considered antiarrhythmic drugs, because these drugs are more often prescribed for indications other than arrhythmia.

The patient characteristics were collected in the ICD Registry and included as potential correlates of the prescription of antiarrhythmic drug at discharge: year of implant, age, gender, race, reason for admission (procedure only, hospitalized with a cardiac indication, or hospitalized with a noncardiac indication), insurance payer, syncope, family history of sudden death, history of heart failure, New York Heart Association class, history of cardiac arrest, atrial fibrillation or flutter, ventricular tachycardia (VT), cardiac transplantation, nonischemic cardiomyopathy, ischemic heart disease, previous myocardial infarction, previous coronary bypass surgery, previous percutaneous coronary intervention, previous valvular surgery, previous pacemaker implantation, cerebrovascular disease, chronic lung disease, diabetes, hypertension, renal failure requiring dialysis, ejection fraction, QRS duration, PR interval, creatinine, blood urea nitrogen, serum sodium, systolic blood pressure, ICD indication (primary, secondary), and ICD type (single, dual chamber, CRT-D). The provider characteristics and implanting physician’s training (board-certified or nonboard-certified electrophysiologist, surgeon, pediatric cardiologist, Heart Rhythm Society-credentialed, none of these) were also collected. The facility characteristics that were collected included geographic location, owner, urban/rural location, number of patient beds, and teaching status. We used the US Census division as a geographic variable.

All baseline variables collected in the ICD Registry were analyzed and compared between the patients with and without an antiarrhythmic prescription. Univariate associations were examined using t tests for continuous variables and chi-square testing for categorical variables. The factors independently associated with the use of antiarrhythmic drugs were identified using hierarchical logistic regression analysis, accounting for the potential clustering of patients among hospitals using a backward selection method. Most of the covariates had low rates of missing values (<1%); in such cases, the missing values were imputed as the median for continuous variables and the most common category for the categorical variables for the model analyses. For the few variables with missing rates >1%, the missing values were considered as a category for categorical variables. For continuous variables, a dummy variable indicating the missing values was created, and the missing values were imputed as the median. Both variables were considered as candidate variables in all multivariate models.

We prespecified several subgroups of interest to stratify our analysis of antiarrhythmic prescription: primary prevention ICD, secondary prevention ICD, atrial fibrillation or flutter, the absence of atrial fibrillation or flutter, VT, and ischemic and nonischemic cardiomyopathy. In addition, we included a subgroup of patients without any clearly documented indication for antiarrhythmic therapy (i.e., primary prevention ICD indication and absence of atrial fibrillation or VT).

To examine the prescription of antiarrhythmic at discharge during different periods, we examined the rate of prescription of antiarrhythmic agents in each year and tested the trend using the Cochran-Armitage trend test for different types antiarrhythmic medications and among the different subgroups.

To assess the institutional-level variation in use, we assessed the distributions and interquartile ranges of discharge antiarrhythmic prescriptions in the entire sample and in each year. To quantify the extent to which the variation in antiarrhythmic drug use was explained by hospital level effects, the hospital-specific median odds ratio (MOR) was determined, adjusting for the patient, clinical, and demographic characteristics. The MOR will always be >1.0. An MOR of 1.0 signifies no clustering effects, and an MOR of 2.0 indicates that an individual patient would have twofold greater odds of receiving an antiarrhythmic medication at discharge if cared for at another randomly selected hospital.

All analyses were performed using the Statistical Analysis Systems statistical package, version 9.1 (SAS Institute, Cary, North Carolina) and approved by the Yale Human Subjects Investigations Committee. Variables with p <0.05 were considered statistically significant.

Results

Of the entire cohort, 15.1% received an antiarrhythmic drug at discharge. The patients receiving an antiarrhythmic drug were older and more likely to have a history of cardiac arrest, atrial fibrillation or flutter, VT, or syncope ( Table 1 ). They also were more likely to have received an ICD for secondary prevention or to have received a dual-chamber device. However, they were less frequently prescribed an angiotensin-converting enzyme inhibitor or angiotensin receptor blocker or β blocker. Modest differences were present in the prevalence of coexisting illnesses between the 2 groups.

Of the antiarrhythmic drugs used, the class III agents (14.1% of cohort and 93.3% of all antiarrhythmic drugs) were the most commonly prescribed. Of these, amiodarone was the most commonly prescribed (81.5% of all antiarrhythmics) followed by sotalol (9.9%; Figure 1 and Supplemental Table 1 ).

Temporal trends in antiarrhythmic prescribing at hospital discharge in ICD recipients. (A) Class III antiarrhythmic drugs prescribed at discharge. From 2006 to 2011, a modest increase was seen in use of class III drugs (13.9% to 14.9%, p <0.01). Each of the class III drugs was significantly more likely to be prescribed in 2011 than in 2006. Information on dronedarone was not routinely captured in the ICD Registry, (B) Class I and other antiarrhythmic drugs prescribed at discharge. Minimal change was seen in class I antiarrhythmic drug use from 2006 to 2011.

From 2006 to 2011, a modest increase occurred in antiarrhythmic prescriptions (15.0% to 16.1%, p <0.01), with greatest increase for patients who had received an ICD for secondary prevention (26.2% to 30.5%, p <0.01) or with a history of VT (23.0% to 27.1%, p <0.01; Figure 2 ). Also, an increase occurred in the use of antiarrhythmic drugs among patients with nonischemic cardiomyopathy (15.4% to 17.2%, p <0.01) but not in patients with ischemic cardiomyopathy or those receiving primary prevention ICD or CRT-D therapy. Although a modest increase in use was observed for patients with atrial fibrillation (28.0% to 29.4%, p <0.01), a comparable increase in use was also observed for those without atrial fibrillation (9.1% to 9.7%, p <0.01). Among the patients with a primary prevention ICD and without a history of atrial fibrillation or VT, a small decline occurred in antiarrhythmic prescriptions (4.4% to 3.8%, p <0.01).

Trends in antiarrhythmic drug prescriptions at hospital discharge in subgroups of ICD recipients. From 2006 to 2011, the greatest increase in antiarrhythmic drug prescribing was in the subgroups of patients who had received an ICD for secondary prevention (26.2% to 30.5%, p <0.01) or with a history of VT (23.0% to 27.1%, p <0.01). A modest increase was also seen in the use of antiarrhythmic drugs for patients with nonischemic cardiomyopathy (15.4% to 17.2%, p <0.01).

From 2006 to 2011, a modest increase was seen in the use of class III drugs (13.9% to 14.9%, p <0.01), with minimal changes in class I use. Each of the class III drugs was significantly more likely to be prescribed in 2011 than in 2006.

The patient, provider, and facility level variables associated with antiarrhythmic prescriptions are listed in Table 2 . The patients who were acutely hospitalized for cardiac or noncardiac indications were more likely to be prescribed antiarrhythmic therapy at discharge than were patients electively hospitalized for ICD implantation. The other patient characteristics associated with increased antiarrhythmic prescription included atrial fibrillation or flutter, a history of cardiac arrest, and VT. Compared with patients whose ICDs were implanted by an electrophysiologist, those whose ICD was implanted by physicians other than electrophysiologists were more likely to be discharged with an antiarrhythmic agent. The characteristics associated with a lower likelihood of antiarrhythmic prescription are also listed in Table 2 . These included black race and renal failure requiring dialysis. Patients undergoing implantation of a primary prevention ICD were less likely to receive an antiarrhythmic drug than were patients who had received an ICD for secondary prevention. Also, patients who had received a single-chamber ICD were less likely to receive an antiarrhythmic drug prescription than were patients who had received a dual-chamber or CRT-D device. Significant regional differences in use were found when stratified by US Census division.

Table 2

Characteristics associated with use of antiarrhythmic drugs at discharge

Characteristic	Adjusted OR	95% CI	p Value
Year
2006	Reference
2007	0.96	0.93–0.99	0.01
2008	0.93	0.91–0.96	<0.01
2009	0.96	0.94–0.99	0.01
2010	1.04	1.01–1.07	0.02
2011	1.11	1.07–1.14	<0.01
Age	1.002	1.001–1.003	<0.01
Female gender	0.91	0.89–0.92	<0.01
Race
White	Reference
Black	0.87	0.85–0.90	<0.01
Other	0.91	0.88–0.95	<0.01
Reason for hospitalization
Elective for procedure	Reference
Hospitalized, cardiac	1.57	1.53–1.60	<0.01
Hospitalized, noncardiac	1.72	1.68–1.76	<0.01
Previous syncope	0.86	0.84–0.88	<0.01
Family history of sudden death	0.91	0.87–0.95	<0.01
Previous heart failure	0.96	0.94–0.98	<0.01
NYHA class
I	Reference
II	0.99	0.96–1.02	0.57
III	1.06	1.02–1.09	<0.01
IV	1.22	1.16–1.28	<0.01
Cardiac arrest	1.57	1.53–1.62	<0.01
Atrial fibrillation or flutter	3.90	3.83–3.97	<0.01
Ventricular tachycardia	1.97	1.93–2.01	<0.01
Ischemic heart disease	1.09	1.06–1.12	<0.01
Previous myocardial infarction	1.06	1.04–1.08	<0.01
Previous coronary bypass	0.93	0.91–0.95	<0.01
Previous percutaneous intervention	0.95	0.94–0.97	<0.01
Previous valvular surgery	1.18	1.15–1.21	<0.01
Previous pacemaker	0.96	0.93–0.98	<0.01
Cerebrovascular disease	0.95	0.92–0.97	<0.01
Diabetes mellitus	0.92	0.90–0.93	<0.01
Hypertension	0.97	0.95–0.99	<0.01
Renal failure requiring dialysis	0.88	0.84–0.92	<0.01
Ejection fraction (%)	0.995	0.994–0.996	<0.01
QRS duration (per 10 ms)	1.025	1.022–1.028	<0.01
PR interval (per 10 ms)	1.030	1.027–1.032	<0.01
Creatinine (mg/dl)	1.06	1.05–1.07	<0.01
Blood urea nitrogen (per 10 mg/dl)	1.012	1.006–1.019	<0.01
Systolic blood pressure (per 10 mm Hg)	0.994	0.990–0.998	<0.01
Primary prevention indication	0.56	0.55–0.58	<0.01
Implantable cardioverter defibrillator type
CRT-D	Reference
Single chamber	0.63	0.61–0.65	<0.01
Dual chamber	1.23	1.20–1.25	<0.01
Training of implanting physician
Board-certified electrophysiologist	Reference
Nonboard-certified electrophysiologist	0.98	0.95–1.02	0.31
Surgeons	1.23	1.15–1.32	<0.01
Pediatric cardiologist	0.87	0.60–1.26	0.45
Credentialed ∗	1.12	1.08–1.17	<0.01
None of these	1.08	1.03–1.12	<0.01
Geographic location †
Pacific	Reference
New England	0.62	0.55–0.71	<0.01
Mid-Atlantic	0.73	0.66–0.80	<0.01
South-Atlantic	0.85	0.79–0.93	<0.01
East North Central	0.90	0.83–0.98	0.01
East South Central	1.18	1.06–1.32	<0.01
West North Central	0.91	0.82–1.00	0.05
West South Central	1.04	0.95–1.13	0.44
Mountain	0.84	0.75–0.94	<0.01
No. of patient beds (per 10 beds)	0.998	0.997–0.999	<0.01

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Prognostic Utility of the SYNTAX Score in Patients With Single Versus Multivessel Disease Undergoing Percutaneous Coronary Intervention (from the Acute Catheterization and Urgent Intervention Triage StrategY [ACUITY] Trial)
2. Fatty Acids and Recovery During First Hours of Acute Myocardial Ischemia
3. Outcome of Transcatheter Aortic Valve Implantation in Patients With Low-Gradient Severe Aortic Stenosis and Preserved Left Ventricular Ejection Fraction
4. Recognizing Pregnancy-Associated Cardiovascular Risk Factors

Stay updated, free articles. Join our Telegram channel

Join