Cardiac Device Implantation in the United States from 1997 through 2004: A Population-based Analysis

OBJECTIVE: Use of cardiac devices has been increasing rapidly along with concerns over their safety and effectiveness. This study used hospital administrative data to assess cardiac device implantations in the United States, selected perioperative outcomes, and associated patient and hospital characteristics. METHODS: We screened hospital discharge abstracts from the 1997-2004 Healthcare Cost and Utilization Project Nationwide Inpatient Samples. Patients who underwent implantation of pacemaker (PM), automatic cardioverter/defibrillator (AICD), or cardiac resynchronization therapy pacemaker (CRT-P) or defibrillator (CRT-D) were identified using ICD-9-CM procedure codes. Outcomes ascertainable from these data and associated hospital and patient characteristics were analyzed. MEASUREMENTS AND MAIN RESULTS: Approximately 67, 000 AICDs and 178, 000 PMs were implanted in 2004 in the United States, increasing 60% and 19%, respectively, since 1997. After FDA approval in 2001, CRT-D and CRT-P reached 33, 000 and 7, 000 units per year in the United States in 2004. About 70% of the patients were aged 65 years or older, and more than 75% of the patients had 1 or more comorbid diseases. There were substantial decreases in length of stay, but marked increases in charges, for example, the length of stay of AICD implantations halved (from 9. 9 days in 1997 to 5. 2 days in 2004), whereas charges nearly doubled (from 66, 000 in 1997 to 117, 000 in 2004). Rates of in-hospital mortality and complications fluctuated slightly during the period. Overall, adverse outcomes were associated with advanced age, comorbid conditions, and emergency admissions, and there was no consistent volume-outcome relationship across different outcome measures and patient groups. CONCLUSIONS: The numbers of cardiac device implantations in the United States steadily increased from 1997 to 2004, with substantial reductions in length of stay and increases in charges. Rates of in-hospital mortality and complications changed slightly over the years and were associated primarily with patient frailty.