Right-Sided Endocarditis From Injection Drug Use: Why Outpatient Antibiotic Step-Down Isn't the Default Here
His infection likely doesn't need surgery, and the antibiotics that will cure it are well established — the actual disagreement is how to deliver a month or more of inpatient treatment to a man who has already left before completing one.
K.D., a 33-year-old man, is back in the same hospital for the third time in eighteen months for the same underlying reason — an infection following intravenous drug use, this time growing methicillin-susceptible Staphylococcus aureus from blood cultures drawn on admission for fever and malaise. He left against medical advice partway through his second admission, roughly a year ago, after eight days of a planned four-week antibiotic course, a fact his current team knows because it's written plainly in his chart and because he brought it up himself before anyone asked.
This admission's echocardiogram found a mobile vegetation on his tricuspid valve, consistent with right-sided infective endocarditis — the pattern most closely associated with repeated IV injection, and mechanistically and prognostically distinct from left-sided disease. Right-sided endocarditis in this population is more often managed successfully with antibiotics alone; surgery is reserved for a smaller minority with large vegetations, recurrent septic pulmonary emboli, or right heart failure, none of which he has so far. The harder question isn't whether he needs surgery — he likely doesn't, not yet — but how to actually deliver the four-to-six weeks of antibiotics his infection requires to a man who has already shown, once, that a long inpatient stay doesn't reliably hold him, and who is still using the drug that caused this. He has said this admission, without being asked directly, that he wants to finish treatment this time — a statement his team takes seriously without treating it as a guarantee, given what happened at his last admission under what felt, at the time, like the same intention.
On the wards, third admission
Right-sided endocarditis from MSSA in someone without heart failure or large vegetations typically responds well to a full course of IV nafcillin, 2 g every four hours, for four to six weeks depending on complicating features. I want to name the obvious objection before anyone raises it: there is an older abbreviated two-week regimen for uncomplicated right-sided MSSA disease in people who inject drugs, and it would solve half of what we're about to argue over. Current guidance has moved away from it — the 2022 AHA statement on endocarditis in this population advises against abbreviated courses, and the European guideline puts MSSA endocarditis at four to six weeks regardless of which valve is involved. My concern with any step-down plan is that the evidence supporting oral or outpatient options, like POET, was built almost entirely in stable left-sided disease in patients not actively using intravenous drugs — extrapolating that evidence to his situation isn't supported by what the trial actually tested.
If he had left-sided disease, no ongoing IV drug use, and a clearly reliable home situation, I'd be having a completely different conversation — oral step-down therapy is genuinely well-supported in that population, and I don't want anything I'm saying here to sound like blanket skepticism of it generally.
I don't disagree with the antibiotic evidence, but I want to name what actually happened last time, because it matters for what we plan differently now rather than just repeating the same admission and hoping the outcome changes. He left after eight days, not because the medical plan was wrong, but because nothing in that admission addressed his substance use itself alongside the infection. If we're keeping him inpatient for four to six weeks again without a real plan for his addiction during that time, I think the risk of a repeat AMA departure is high regardless of how sound the antibiotic reasoning is.
I'm not proposing to send him home with a PICC line and self-administered antibiotics — that would be a real risk given active use, both for line infection and for potential misuse of the access itself. What I am proposing is starting medication for opioid use disorder now, during this admission, as a real part of the treatment plan, not an afterthought — his stated wish to complete treatment this time is worth taking seriously and building an actual structure around, not just noting.
From the valve side, there's no indication for surgery today, and I don't expect that to change unless his echocardiogram findings worsen or he develops a complication over the coming weeks — right-sided IE without heart failure or large-vegetation criteria is managed medically far more often than not. I'd keep inpatient IV antibiotics as the plan, not because outpatient options are never appropriate for right-sided IE in principle, but because the specific combination here — active use, a prior AMA departure, no established safe home infusion plan — doesn't meet the bar those options were shown safe for.
Agreed: full inpatient course of IV nafcillin, buprenorphine started this admission for opioid use disorder, and weekly echocardiograms to watch for any complication that would change the surgical picture.
This becomes the first admission where treatment for both the infection and the underlying substance use happened together, not separately.
The team has committed in advance to discussing harm-reduction-oriented options directly with him rather than only presenting the choice as complete-the-course-or-leave-against-advice.
Addiction medicine will follow him throughout this admission regardless of which branch it becomes, not only if a crisis point is reached.