The first sign was not a fever. It was bleeding in the brain.
A 5-year-old girl with a surgically repaired heart defect arrived at the hospital with a ruptured aneurysm in a branch of her left middle cerebral artery, bleeding into the frontal lobe. An echocardiogram found a growth on her mitral valve, the kind of finding that points strongly toward infective endocarditis, an infection of the heart's inner lining.
Then the blood cultures came back. Negative. Repeatedly.
The case, reported in JACC: Case Reports, describes what happened next. Rather than guessing at an antibiotic, clinicians ordered plasma microbial cell-free DNA sequencing, a test that scans a blood sample for fragments of microbial genetic material shed by an infection anywhere in the body. It named the organism: Gemella haemolysans.
The Blind Spot in Standard Testing
Blood cultures are the foundation of endocarditis diagnosis, and they fail more often than most people would guess.
According to a scientific statement from the American Heart Association, up to 30 percent of infective endocarditis cases produce negative blood cultures. The statement identifies two main reasons. Antibiotics given before cultures are drawn is the single biggest cause. The other is infection by organisms that simply do not grow well in a culture bottle.
That second category has a name in microbiology: fastidious. These are bacteria with demanding growth requirements, and Gemella species sit squarely in it. They are Gram-positive cocci that normally live harmlessly in the mouth and upper airway, and only occasionally turn opportunistic.
The practical cost of a negative culture is that treatment becomes a hedge. Without a named organism and its drug susceptibilities, clinicians reach for broad-spectrum coverage and hope it lands.
A Mouth Bacterium with an Underappreciated Reach
Gemella endocarditis is uncommon enough that most of what is known comes from stitching together individual reports.
A systematic review of 52 published cases found that Gemella morbillorum accounted for 46.3 percent and Gemella haemolysans for 25.9 percent, with a striking male predominance of nearly 80 percent. The clinical presentation, the authors found, looked like ordinary endocarditis, which is part of the problem. What set these cases apart was delay, driven by the organism's slow, fussy growth and the quiet course of the illness.
There is a wrinkle. A Swedish registry analysis identified 29 episodes among 7,677 registered cases of endocarditis, or 0.4 percent, and found a different species mix: Gemella morbillorum at 47 percent but Gemella bergeriae at 27 percent, a species that barely registers in the case report literature. That gap is a useful reminder that what gets written up is not the same as what actually circulates. Case reports favor the unusual, while registries capture the routine.
The mouth connection recurs. In a case published in 2023, a 37-year-old man with two months of unexplained fever turned out to have Gemella morbillorum endocarditis, identified by sequencing after cultures grew nothing usable. He had undergone root canal therapy a month before symptoms began.
What Sequencing Can and Cannot Do
The technology at the center of this case works differently from culture. Instead of trying to grow an organism, it looks for genetic fragments circulating in plasma and matches them against reference databases. Nothing has to survive a trip to the lab.
That is a real advantage for fastidious organisms and for patients already on antibiotics. It is not a replacement for blood cultures, and the American Heart Association writing group is careful on this point. Molecular tests can detect genetic material from organisms that are dead, irrelevant, or contaminating the sample, and a positive result still has to be interpreted against the clinical picture. Cultures also provide something sequencing does not: a living isolate to run antibiotic susceptibility testing on.
The identification changed this child's management directly. She underwent emergency neurosurgical and endovascular treatment for the brain bleed, then received antibiotics aimed at the organism that had been named rather than at a broad guess. The report describes subsequent neurological recovery.
Why Repaired Hearts Stay on the Watch List
The underlying condition matters. She had a repaired partial atrioventricular canal defect, and infective endocarditis is a recognized complication of repaired congenital heart disease.
That is not a reason for alarm among families of children with corrected heart defects. Endocarditis remains uncommon, and repair generally lowers risk rather than raising it. But it does explain why cardiologists keep these patients under long-term follow-up, and why dental and oral health guidance features so prominently in their care, given that the usual culprits are mouth organisms.
The limits of this evidence should be stated plainly. This is a single pediatric case. It demonstrates that molecular testing changed the course for one patient. It does not establish how often the test finds an answer, how often it misleads, or which patients should get it. A retrospective cohort of 141 patients with suspected endocarditis found a more mixed picture: sequencing identified a pathogen in about 61 percent of confirmed cases, compared with 39 percent for routine workup, but clinically insignificant detections turned up in nearly 29 percent of patients who did not have endocarditis at all.
For anyone with a heart valve problem or a repaired congenital defect, the actionable piece is old-fashioned. A fever that will not resolve, or unexplained symptoms lasting weeks, is worth mentioning to a clinician along with the cardiac history, because that history changes what gets looked for.
Key Questions Answered
What is culture-negative endocarditis?
A heart valve infection in which standard blood cultures grow nothing. The American Heart Association says it accounts for up to 30 percent of cases.
Why did the cultures fail here?
Gemella species are fastidious, meaning they have demanding growth requirements and often do not grow in routine culture bottles.
What is microbial cell-free DNA sequencing?
A blood test that searches plasma for fragments of microbial genetic material, allowing identification without growing the organism.
Where does Gemella normally live?
In the mouth and upper respiratory tract, where it is usually harmless. It causes disease only opportunistically.
Does this test replace blood cultures?
No. Cultures still yield an isolate for antibiotic susceptibility testing, and molecular results require careful clinical interpretation.
Are children with repaired heart defects at risk?
Endocarditis is a recognized complication of repaired congenital heart disease, which is why these patients stay under long-term cardiology follow-up.
