Hospitalizations, heart valve surgeries for drug-associated infective endocarditis increase

From 2007 to 2017, annual hospitalizations for drug use-associated infective endocarditis increased by about 12-fold in one state, with opioids implicated in nearly two-thirds of cases.


Rates of overall hospitalizations, as well as hospitalizations with surgery, for drug use-associated infective endocarditis markedly increased from 2007 to 2017, according to a recent study of North Carolina hospitals.

Using a statewide hospital discharge database, researchers analyzed all adult patients hospitalized for infective endocarditis from July 1, 2007, through June 30, 2017. Drug use-associated infective endocarditis was identified by a diagnosis code indicating recreational drug use, withdrawal, dependence, poisoning, or a diagnosis of hepatitis C virus infection in patients born after 1965 (codes for marijuana use or unspecified drug use were excluded). Results were published online on Dec. 4 by Annals of Internal Medicine.

Of 22,825 infective endocarditis hospitalizations during the 10-year period, 2,602 (11%) were associated with drug use. Valve surgery was performed in 1,655 (7%) infective endocarditis hospitalizations, including 285 (17%) for drug use-associated infective endocarditis.

Annual hospitalizations for drug use-associated infective endocarditis increased from 64 (0.92 per 100,000 persons) in 2007 to 2008 to 867 (10.95 per 100,000 persons) in 2016 to 2017, an increase of about 12-fold. Until mid-2013, fewer than 10 drug use-associated infective endocarditis hospitalizations with surgery occurred each year in North Carolina. By 2016 to 2017, there were 109 (1.38 per 100,000 persons), an increase of 13-fold. Similar increases were not observed in cases of infective endocarditis without drug use. The most common drugs observed in drug use-associated infective endocarditis were opioids, which were involved in 1,626 (62%) cases. Of all patients hospitalized with drug use-associated infective endocarditis, 333 (13%) were discharged against medical advice.

Compared to patients with non-drug use-associated infective endocarditis, those hospitalized for drug use-associated infective endocarditis were younger (median age, 35 years vs. 67 years), were more often female (51% vs. 48%) and of non-Hispanic white ethnicity (81% vs. 61%), and had longer lengths of stay (median, 11 vs. 7 days) with higher median charges ($60,333 vs. $34,968). The same differences were found in comparisons of patients having surgery for infective endocarditis that was or was not drug use-associated.

The authors noted that the study used administrative data lacking granular clinical information and may have been subject to inaccurate coding of drug use or other diagnoses. They added that using hepatitis C virus infection as an indicator of drug use (which accounted for 18% of cases) may have artificially increased the number of drug use-associated infective endocarditis cases detected.

Clinicians must take action to improve care for patients with drug use-associated infective endocarditis, according to an accompanying editorial. For instance, they should know that no evidence exists to support concerns around providing postdischarge outpatient parenteral antibiotic therapy to people with a history of substance use, according to the editorialist. In addition, to help avoid discharge against medical advice and readmissions, addiction medicine clinicians must be involved soon after infected patients are admitted, the editorialist said. “Initiating medical treatment for substance use disorder during hospitalization is acceptable, feasible, and sustainable after discharge,” the editorialist wrote.

ACP Hospitalist covered the challenges of treating drug use-associated endocarditis in an October 2017 article.