Healthcare utilization, expenditures are high after ICD shocks

Nearly 15 percent of patients had a shock event after receiving an implantable cardioverter-defibrillator (ICD), according to a retrospective cohort study.

Of the shock events, 46 percent were associated with any healthcare utilization and 14 percent were associated with hospitalizations.

After the shocks, patients often underwent inpatient cardiovascular procedures, including echocardiography (59 percent), electrophysiology study or ablation (34 percent), stress testing (16 percent) and lead revision (11 percent).

Lead researcher Mintu P. Turakhia, MD, MAS, of the Stanford University School of Medicine and Veterans Affairs Palo Alto Health Care System, and colleagues published their results online in Circulation: Cardiovascular Quality and Outcomes on Feb. 14. Medtronic funded the study.

“ICDs cannot assess patients the way a doctor can,” Turakhia said in a news release. “The device doesn’t know, for instance, if the patient is unconscious or has a pulse. We wanted to see what happens after a shock, in terms of care and cost, to help define the potential benefit of smarter ways to program these devices.”

The researchers used the Medtronic Data Warehousing and Analytics Service (DWAS) and identified patients with ICDs and cardiac resynchronization therapy defibrillator (CRT-D) devices implanted between 2008 and 2010. They stripped the data of personal identifiers of name, date of birth, medical record number and social security number.

They also identified patients with ICDs and CRT-D implants from the Truven Health MarketScan commercial claims and encounters database. They then used a deterministic matching algorithm to link patients who had data in the DWAS and MarketScan databases. All of the patients had Medtronic ICDs and remote monitoring.

In all, the researchers linked 10,266 patients from the DWAS data set to the MarketScan database. Of those patients, 14.9 percent had at least one shock event, which the researchers defined as one or more spontaneous shocks delivered by an implanted device.

The mean age of patients with shock events was 61.3 years old, while 81 percent were male, 41 percent had Medicare supplemental coverage, 60 percent had preimplant ischemic heart disease or previous MI, 41 percent had ventricular arrhythmias and 37 percent had atrial arrhythmias. In addition, 23 percent had single chamber, 49 percent had dual chamber and 28 percent had cardiac resynchronization ICDs. Further, 43 percent had appropriate shocks, 42 percent had inappropriate shocks and 12 percent had appropriate and inappropriate shocks.

Of the shock events, 13.7 percent were followed by inpatient healthcare utilization and 32.2 percent were followed by outpatient healthcare utilization. The most frequently reported procedures during shock-related inpatient healthcare utilization were electrocardiogram (85.3 percent), chest x-ray (75.7 percent), cardiac catheterization (75.7 percent), echocardiogram (58.7 percent) and electrophysiology study/ablation (33.6 percent). Cardiovascular procedures were less common but still substantial after inappropriate shock events.

Cardiac catheterization occurred in 71 percent of cases after appropriate shocks and in 51 percent of cases after inappropriate shocks, while PCIs occurred in 6.5 percent and 5.0 percent of cases, respectively. The mean length of hospital stay was 3.6 days after appropriate shock events and 2.8 days after inappropriate shock events.

For all shock events, the mean expenditure was $5,887 and the median expenditure was $901. The mean and median expenditures for healthcare utilization after appropriate shock events were $5,592 and $940, respectively, while the mean and median expenditures for healthcare utilization after inappropriate shock events were $4,470 and $681, respectively. The mean and median expenditures for patients with inpatient healthcare utilization were higher than for patients with outpatient healthcare utilizations.

The researchers mentioned the study had a few limitations, including that the results could be affected by confounding factors such as patient behaviors and lifestyles, additional medical or pharmaceutical interventions or the indication for the ICD. The results might also not be generalizable to patients who are uninsured, have ICDs from a different manufacturer or do not use remote monitoring.

“[Healthcare utilization] and expenditures, including use of invasive cardiovascular procedures, were substantial, even after inappropriate shock events,” the researchers wrote. “Strategies to reduce the incidence of ICD shocks may result in significant reductions in [healthcare utilization] and expenditures. Further investigation is warranted to determine the impact on survival of these procedures after appropriate and inappropriate shocks.”