Scrub Typhus Can Imitate Typhoid Fever in Bone Marrow and Mislead Diagnosis
A 79-year-old man in Shandong Province arrived with fever, syncope, cytopenias, and a marrow picture that looked like enteric fever. The final diagnosis was scrub typhus, not typhoid fever, and the mismatch between morphology and microbiology carried a clear bedside lesson: bone marrow findings alone can point clinicians in the wrong direction when an intracellular pathogen is doing the damage.
The patient had 4 days of spiking temperatures to 39.5°C with rigors, but his history reached back a month because of repeated transient loss of consciousness. He lived in a rural setting, tended a backyard garden, and had daily outdoor exposure that would later fit the epidemiology of Orientia tsutsugamushi. On arrival, he was pale and febrile, with no murmur, no focal neurologic deficit, and no abdominal organ enlargement. Cranial imaging did not explain the syncopal spells, and chest and abdominal imaging found only scattered lymphadenopathy in the mediastinum and groins.
Laboratory results quickly moved the case beyond a simple febrile illness. He had pancytopenia, macrocytic anemia, thrombocytopenia, elevated transaminases, increased lactate dehydrogenase, renal dysfunction, prolonged activated partial thromboplastin time, and elevated D-dimer. Ferritin was also elevated. The pattern suggested systemic inflammation and possible marrow involvement rather than an isolated bloodstream infection. Tests for brucellosis, HIV, bunyavirus, hepatitis, cytomegalovirus, and SARS-CoV-2 were negative, narrowing the differential but not resolving it.
Bone marrow aspirate revealed 23% ring sideroblasts, and sequencing found an SF3B1 mutation with a 45,X,-Y karyotype, supporting a diagnosis of myelodysplastic syndrome. Yet the marrow abnormality did not explain the abrupt fevers, the recurrent syncope, or the inflammatory laboratory profile. A second process was clearly operating. The question was which one.
Typhoid Cells, Hemophagocytosis, and a Pathogen Hiding in Plain Sight
The marrow contained clusters of large macrophages packed with fine eosinophilic granules. In standard hematopathology, those cells resemble the classic “typhoid cells” associated with Salmonella Typhi infection. Hemophagocytosis of red cells and platelets was also present, and the platelet count fell further over a matter of days. In a patient with fever, pancytopenia, and these marrow findings, enteric fever seemed plausible at first glance.
The rest of the case did not fit neatly. He had no abdominal pain, no diarrhea, and no rose spots. Blood and marrow cultures remained negative, and the Widal test was negative as well. The combination of marrow morphology and absent microbiologic support made the apparent typhoid diagnosis unstable. This is where scrub typhus entered the frame in an unexpected way.
Targeted next-generation sequencing of the bone marrow aspirate identified O. tsutsugamushi. A repeat physical examination then uncovered a 5-mm eschar in the left popliteal fossa, with a black necrotic center and surrounding erythema. The lesion had been easy to miss, tucked behind the knee and outside the line of routine attention. When asked again, the patient did not recall a bite, but he did confirm that he gardened barefoot and in shorts every day. That exposure history matched a mite-borne infection far better than enteric fever.
The marrow findings now made biological sense. Scrub typhus is an intracellular infection that can trigger intense macrophage activation and secondary hemophagocytic lymphohistiocytosis. In that inflammatory state, marrow histiocytes can become conspicuous, overfilled, and misleading. The resemblance to typhoid cells is striking enough to create a diagnostic trap, especially in older adults or in patients with another hematologic disorder already present.
Doxycycline Reversed the Acute Inflammatory Syndrome Fast
Intravenous doxycycline 100 mg twice daily produced rapid clinical change. The fever resolved within 48 hours. By day 7, the platelet count had risen to 152,000/µL, and aminotransferases had normalized. At 3-week follow-up, the syncopal episodes had not returned, and treatment for MDS had begun.
The speed of recovery matters here. If the marrow abnormalities had been driven mainly by MDS, that platelet rebound would not be expected so quickly after an anti-rickettsial agent alone. The pattern instead argues that scrub typhus was responsible for the acute pancytopenia and organ dysfunction, while MDS remained a separate chronic diagnosis. The case therefore separated the background clonal marrow disease from the acute infectious syndrome layered on top of it.
This distinction is clinically important because scrub typhus is treatable and time-sensitive. Delay increases the chance of multiorgan dysfunction, and the disease is no longer confined to its historical geography. Scrub typhus has expanded across China, including into provinces such as Shandong, so a narrow geographic assumption can no longer be relied upon. In a febrile patient with cytopenias, a careful skin examination can be as valuable as advanced molecular testing.
Marrow Morphology Alone Cannot Exclude Scrub Typhus
The case broadens the hematopathologic spectrum of scrub typhus. Typhoid-like macrophages have long been treated as a hallmark of enteric fever, but this patient shows that the same marrow appearance can arise from another intracellular pathogen. When that pattern appears without microbiologic evidence of Salmonella infection, clinicians should widen the differential to include rickettsial disease.
Secondary hemophagocytic lymphohistiocytosis also belongs in the differential when fever, pancytopenia, ferritin elevation, and marrow hemophagocytosis cluster together. Scrub typhus can drive that syndrome, and the inflammatory phenotype may improve with antibiotics alone. In practical terms, that means a marrow report should not be read in isolation from the fever pattern, exposure history, and skin exam.
For confirmation, 47 kDa polymerase chain reaction remains the preferred first-line molecular test when scrub typhus is suspected. Targeted next-generation sequencing proved useful here only because the infection had not been recognized at the bedside. The patient’s concealed eschar, the outdoor exposure, and the brisk doxycycline response fit the diagnosis more cleanly than the initial marrow morphology ever could.
The unresolved issue is not whether scrub typhus can mimic typhoid fever at the marrow level; this case shows that it can. The remaining challenge is diagnostic discipline: in an endemic or expanding region, a hidden eschar and a short antibiotic response window may matter more than a striking but nonspecific histologic pattern.