Medical case narrative
A 14-year-old called schizophrenic for 18 months — the clue was on his skin
What happened
A 14-year-old boy in the Midwestern US (unnamed in the published report — previously high-functioning: national history and geography competitions, lead actor in a school play, a fencing award) developed psychosis in October 2015 that ran for about 18 months and four hospitalisations. His first admission brought a diagnosis of major depressive disorder with psychotic features. Two psychiatrists then diagnosed schizophrenia, and an 11-week admission at a large out-of-state psychiatric teaching hospital in the summer of 2016 discharged him with diagnoses of schizophrenia and obsessive-compulsive disorder. His mother quit her job in December 2015 to supervise him full-time.
What it looked like
Persecutory delusions (he believed a family cat wanted to kill him), auditory, visual, and tactile hallucinations, rage outbursts — alongside fatigue, daily headaches, chest pains, shortness of breath, and urinary frequency. He had sustained cat bites and scratches before the onset. And after the August 2016 discharge, his parents noticed red, stretch-mark-like skin lesions on his thighs and armpit — the flag that finally redirected the workup.
What the labs found
The extensive summer-2016 workup — inflammation markers, an autoimmune encephalopathy panel, EEG, brain MRI — was unremarkable. A consulting neurologist proposed autoimmune encephalitis in October 2016 on a positive ANA (1:320) and a rising CRP, and rituximab reduced how often his episodes came, though daily disruption continued. PANS (paediatric acute-onset neuropsychiatric syndrome) was diagnosed in January 2017, and the evaluating physician, seeing the skin lesions, proposed neurobartonellosis — infection with Bartonella, the bacteria behind cat-scratch disease. In March 2017, research testing found Bartonella henselae DNA, PCR-amplified and sequenced, in his blood, serum, and two of three seven-day enrichment blood cultures — despite nearly two months of doxycycline. The ordinary antibody test, notably, was negative: his blood test told the story through DNA, not antibodies.
What changed
On a combination antimicrobial regimen, his psychoses decreased from January 2017 and none were reported by November 2017; the lesions resolved, all antipsychotic drugs were withdrawn, and he returned to all his pre-illness activities. The honest complications travel with this entry. Causation is contested, in the authors' own register: the report concludes only that B. henselae infection "may contribute to" such symptoms "in a subset of patients," and that a case report cannot establish causation. The positive microbiology came from research-basis testing at the authors' own specialist laboratory; there is no independent family-told account — the family's voice reaches the record only inside the medical paper. Repeat testing in June 2017 was mixed (enrichment-culture testing negative across nine extractions, but two of three samples direct-PCR positive), with a fully negative repeat in October 2017. And in June 2017, abrupt withdrawal of clozapine and tramadol was followed by severe illness and a 20.5 kg weight loss over 40 days, which resolved with treatment for reflux oesophagitis. Improvement was also gradual, over months, on a five-drug combination, after rituximab had already reduced his episodes — this entry does not credit the antibiotics alone.
The test that told the story
Bartonella henselae PCR and DNA sequencing from enrichment blood cultures — after the standard antibody test came back negative.
Sources: Breitschwerdt et al., Journal of Central Nervous System Disease 2019 (published case report, all values); NC State University news release via Medical Xpress (March 2019), summarising the same report in the lead author's words.