Infective Endocarditis in IV Drug Users Doesn’t Always Match Expectations

New data on infective endocarditis (IE) among people who inject drugs run counter to long-held stereotypes about the condition—for example, a sizeable proportion of cases affect the heart’s left side, occur in prosthetic valves, and arise from organisms other than Staphylococcus aureus.

The results, drawn from two international, prospective cohort studies, were published online today in the Journal of the American College of Cardiology.

According to the study authors, the series represents the largest of its kind to date in this patient group and should drive home the point that while there are well-known features of “classical” IE in injection-drug users, understanding some of the “not so typical” findings will be key to patient management.

Bernard Prendergast, MD (St Thomas' Hospital, London, England), commenting on the study for TCTMD, described the paper and its accompanying editorial as thought-provoking.

“I think there is a clinical preconception about what these patients look like,” he said. The “negative concept” is that they’re young, often have HIV or hepatitis C, present with tricuspid IE in the right heart, and “nearly always have a staphylococcal infection, because it comes from needle use.” There’s also reluctance to pursue surgery, he continued, because these patients are thought to have poor prognosis.

“This paper challenges some of those assumptions,” Prendergast said, urging that clinicians need to rethink how they approach IV drug users with IE.

There is quite a prolonged opportunity to involve drug addiction and rehabilitation teams to help them. Bernard Prendergast

Editorialists led by Larry M. Baddour, MD (Mayo Clinic, Rochester, MN), agree that “complex strategies” are needed to control the global problem of IE in people who inject drugs.

“Lifelong interventions will be needed to reduce . . . recidivism in [those] who have developed IE and survived, because they remain at high risk for subsequent bouts of IE, regardless of whether they continue to inject drugs or not,” they advise.

Unexpected Patterns

Led by Juan M. Pericàs, MD, PhD, and Jaume Llopis, MD, PhD (University of Barcelona, Spain), the researchers analyzed 8,112 IE episodes from two databases: ICE-PCS, which spanned 2000 to 2006 and 28 countries, and ICE-Plus, which covered 2008 to 2012 and 18 countries. Regions included Europe, North America, Oceania, Africa, Asia, and South America.

There were 7,616 patients in all, 591 of whom (7.8%) used IV drugs. Compared with people who did not inject drugs, those who did tended to be younger (median 37.0 vs 63.3 years), were more likely to be men (72.5% vs 67.4%), and had fewer comorbidities overall. They were, however, more likely to have HIV, liver disease, and prior IE.

In contrast to the expectation that IE associated with IV drug use affects the right side of the heart, 53% of the cases involved left-side valves and 34.5% were entirely located in the left heart.

Native valves were the most common sites of IE, particularly for patients who inject drugs (90.2% vs 64.4%). These individuals were less apt than nonusers to have IE occur in prosthetic valves (9.3% vs 27.0%) and cardiovascular implantable electronic devices (0.5% vs 8.6%; P < 0.001 for all).

S. aureus predominated among IV drug users, causing 65.9% of cases. Yet one-third of cases resulted from other sources, such as streptococci, or were polymicrobial. For people not on IV drugs, S. aureus caused 26.8% of IE episodes.

In terms of complications, the drug users were more apt to have systemic emboli (51.1% vs 22.5%) and persistent bacteremia (14.7% vs 9.3%). They were less likely than nonusers to undergo cardiac surgery (39.5% vs 47.8%). Thanks to their youth and relative health, the people whose IE occurred on the backdrop of IV drug abuse were less likely to die in the hospital (10.8% vs 18.2%) or by 6 months (14.4% vs 22.2%). However, they were more likely to see their IE recur (9.5% vs 2.8%; P < 0.001 for all).

Risk factors for mortality among the people who were injection-drug users were prior IE, left-sided IE, polymicrobial etiology, intracardiac complications, and stroke. Those who underwent cardiac surgery tended to have lower mortality.

The researchers also teased out geographic differences. North America had a higher proportion of IV drug users among their entire IE population compared with other regions, whereas South America showed the opposite. South America also stood out for having a higher rate of surgery among those who inject drugs versus those who don’t. In terms of mortality, Africa was the lone region where in-hospital mortality was higher for drug users than for nonusers.

Beyond Surgery

“Approaching people who inject drugs with a respectful attitude and offering the possibility of cardiac surgery when indicated, together with the adequate harm reduction services and drug abuse disorder treatment, are crucial management elements,” the investigators conclude, acknowledging that their study didn’t track details on the latter.

Prendergast also called for a broad perspective. Surgery is just one element of care for IE, he stressed, and the index hospital stay can provide time for a more-holistic approach.

“Their behavior can be challenging sometimes. They abscond from the ward, and they use drugs, and their friends bring drugs into the ward, and it’s all very difficult. But there is quite a prolonged opportunity to involve drug addiction and rehabilitation teams to help them,” Prendergast said, adding that dental care could also be a means of reducing recurrence.

The editorial also cites “several management conundrums” with regards to IV drug users who develop IE. Among them are “the financial impact on healthcare systems; limited patient healthcare coverage; when to initiate drug addiction treatment; how to manage the frequent discharge against medical advice; the perceived limitations of cardiovascular surgery, particularly in [people] with ongoing high-risk behavior and/or history of recurrent IE; the need for an ‘endocarditis team’ in individual patient management; and a desire to implement oral or long-acting parenteral antimicrobial therapy for treatment of IE as soon as feasible,” its authors note.

When interpreting the current study, there are some caveats to remember, they say. For example, the findings are derived from specialist centers with on-site surgery and patients whose IE could be managed medically may not been referred to these hospitals. Additionally, Prendergast noted, the study period ends before the opioid epidemic’s explosion in the United States and doesn’t explore how concomitant liver disease might have affected outcomes, which he said is an important question for future research.