CT-detected lymphadenopathy and prognostic markers in severe fever with thrombocytopenia syndrome: a retrospective study

Severe fever with thrombocytopenia syndrome (SFTS) is a tick-borne viral disease with substantial mortality in endemic regions. CT findings, particularly lymphadenopathy with perinodal fat stranding, may be under-recognized. We evaluated CT findings and factors associated with in-hospital mortality in an endemic setting. We retrospectively reviewed 16 RT-qPCR–confirmed SFTS patients treated at Ehime Prefectural Central Hospital and Kagawa University Hospital (January 2013–December 2024). CT images were reviewed by a board-certified diagnostic radiologist and an infectious diseases physician for lymphadenopathy and perinodal fat stranding. Clinical and laboratory variables were compared between non-survivors and survivors using permutation tests with bootstrap 95% confidence intervals; categorical factors were assessed by univariable analysis. Median age was 69 years (44–86), and 11 (68.8%) were male. CT showed lymphadenopathy with perinodal fat stranding in 15/16 (93.8%), with one case showing no lymphadenopathy. Two patients (12.5%) died in hospital. Lymph node size was not associated with mortality (difference [non-survivors − survivors], − 0.28 cm; 95% CI, − 0.54 to 0.84; P = 0.549). Non-survivors had higher aspartate aminotransferase (difference, 190.21 U/L; 95% CI, 37.36 to 343.50; P = 0.040) and lower platelet counts (difference, − 5.99 × 10³/µL; 95% CI, − 7.52 to − 4.40; P = 0.009). No categorical factor reached statistical significance (altered mental status, borderline; P = 0.050). In this cohort, CT frequently revealed lymphadenopathy with perinodal fat stranding, an under-recognized pattern that may support earlier recognition of SFTS in patients with compatible clinical and epidemiological features. Additionally, elevated aspartate aminotransferase and decreased platelet counts were associated with in-hospital mortality in this cohort and may be useful as exploratory markers for early risk stratification.