The rates of infection involving cardiac implanted electronic devices (CIEDs), like pacemakers and cardioverter defibrillators (ICDs), are substantial, but only a minority of patients in the United States receive the guideline-directed recommendation of device removal, according to data from a Medicare population.
The study was conducted on the hypothesis that adherence to guidelines were low, “but we were surprised by how low the extraction rates turned out to be,” Sean D. Pokorney, MD, an electrophysiologist at the Duke Clinical Research Institute, Durham, N.C., reported at the annual scientific sessions of the American College of Cardiology.
The major U.S. and European guidelines are uniform in recommending complete extraction for a CIED infection. The American Heart Association and the Heart Rhythm Society and two out of the three other guidelines cited by Pokorney not only recommend extraction but specify prompt extraction.
Neither complete extraction nor prompt extraction are typical.
Of the 11,619 CIED infection cases identified in the Medicare database, 18.2% underwent extraction within 30 days of diagnosis. Only 13% were extracted within 6 days.
Lack of Extraction May Cause Avoidable Mortality
The result is likely to be avoidable mortality. Among those with extraction within 30 days, 80% were still alive 1 year later. Survival at 1 year fell to 67.6% in those without an extraction within this time frame.
This translated to a 22% lower rate of death at 1 year (hazard ratio, 0.78; P = .008) in those who underwent extraction within 30 days.
For those in whom the device was extracted within 7 days, the associated HR for death at 1 year was more than 40% lower (HR, 0.59; P < .001), reported Pokorney, who characterized these reductions as occurring in “a dose-response fashion.”
The very high risk of relapse despite antibiotics is the reason that “there is a class 1 indication for complete hardware removal,” Pokorney. He cited five studies that addressed this question. With partial device removal or medical therapy alone, relapse was consistently 50% or greater. In one study, it was 67%. In another it was 100%.
With complete removal, the rate of infection relapse was 1% or lower in four. In the fifth, the rate was 4.2%.
Infections can occur early or late after implantation, but cases accumulate over time. In the Medicare data sample, infection rates climbed from 0.3% at 1 year to 0.6% at 2 years and then to 1.1% at 3 years, Pokorney reported.
Other studies have also shown a steady increase in the proportion of implanted devices associated with infection over time. In a cohort study conducted in Olmstead County, Minnesota, the cumulative probability of a CIED infection reached 6.2% after 15 years and 11.7% after 25 years. While about half of these were infections localized to the device pocket, the others were potentially life-threatening systemic infections, according to Pokorney, who cited this study.
In his analysis of the Medicare data, all fee-for-service patients receiving a first CIED implant over a period of 14 years were included. The 14-year period ended just before the COVID-19 epidemic.
The more than 11,000 CIED infections were identified in 1,065,549 total CIED patients. Most (72%) had received a pacemaker. Of the others , more than half received an ICD and the others received a cardiac resynchronization device. The median age was 78 years.
Female and Black Patients Even Less Likely to Undergo Extraction
About half (49.1%) of the overall study population was female, but females represented only about 40% of those who developed an infection. Blacks represented just under 8% of the population but nearly 16% of the CIED infections. Both females and Blacks were significantly less likely than the overall study population to undergo extraction for their infection (P < .001 for both).
Perhaps predictably, patients with comorbidities were more likely to develop CIED infections. For example, 87% of those with infection, versus only 64.9% of the overall population, were in heart failure at the time of implantation. Diabetes (68.3% vs. 49.3%), ischemic heart disease (91.9% vs. 79.4%), renal disease (70.5% vs. 37.9%), and chronic obstructive pulmonary disease (70.6% vs. 55.0%) were also more common at baseline in those who went on to a CIED infection than in the overall population.
Based on the evidence that there is a large unmet need to improve adherence to the guidelines, Pokorney called for care pathways and other quality initiatives to address the problem.
The reasons that so many patients are not undergoing prompt device extraction at the time of infection is unclear, but Pokorney offered some hypotheses.
“There appears to be a false belief in the efficacy of antibiotics for treating CIED infections,” Pokorney said.
Comorbidities Shouldn’t Delay Extraction
It is also possible that clinicians are concerned about performing extractions in patients with multiple comorbidities. If clinicians are delaying extractions for this reason, Pokorney suggested this behavior is misdirected given the fact that delays appear to increase mortality risk.
Several experts, including Rachel Lambert, MD, an electrophysiologist and professor of medicine at Yale University, New Haven, Conn., agreed that these data deserve a response.
“I was not surprised by the mortality data, but I was surprised at this low extraction rate,” said Lambert, who concurs with the guidelines. She indicated this study provides teeth to prompt action.
“It is great to have these data about the increased mortality risk to back up the guidelines,” she said.
More information is needed to understand exactly why CIED infection is not now leading to guideline-directed care. Pokorney said: “Where do we go from here is a key question.”
While several different types of initiatives might be needed, Pokorney called for regionalization of care to address the fact that not every center that places CIEDs has the capability to perform extractions.
“Extraction is not available at every center, and it probably should not be available at every center, so mechanisms are need to get patients with infection to the specialized centers that provide care,” he said.
Pokorney has financial relationships with Boston Scientific, Bristol-Myers Squibb, Gilead, Janssen, Medtronic, Pfizer, and Philips. Lambert reported financial relationships with Abbott, Amgen, and Medtronic.
This article originally appeared on MDedge.com, part of the Medscape Professional Network.