Back to Authors

Schultz, Marie

0009-0008-7931-354X

Mayo Clinic

Mayo Clinic Health System

Citations 1

Comparative Clinical Characteristics of ICU and Non-ICU Hospitalized Patients With Human Anaplasmosis
Schultz, Marie, Benny, Christopher , Rich, Jennifer, Shiari, Aryan , Dumic, Igor , Schultz, Marie , Benny, Christopher , Rich, Jennifer R. , Shiari, Aryan , Dumic, Igor

Open Forum Infectious Diseases | May 20, 2026 | Volume: 13 | Issue: 5 | Pages: No page available

10.1093/ofid/ofag313 Get It

Abstract Background Human granulocytic anaplasmosis (HGA) can cause severe illness requiring intensive care, but factors associated with critical illness among hospitalized patients are incompletely described. Method We performed a retrospective multicenter cohort study of adults hospitalized with PCR-confirmed HGA at Mayo Clinic Rochester and affiliated Mayo Clinic Health System sites in Minnesota and Wisconsin from 2014 through 2024. Patients who required ICU-level care were compared with those managed on general medical wards using exploratory univariate analyses (Fisher exact and Kruskal–Wallis tests). Results Among 101 hospitalized patients, 12 (12%) required ICU-level care. Symptoms documented at hospital presentation/admission were more frequent in ICU patients, including vomiting (64% vs 30%) and abdominal pain (55% vs 26%). On admission, ICU patients had lower sodium (mean 128.8 vs 132.3 mmol/L) and higher total bilirubin (mean 1.3 vs 0.8 mg/dL). Platelet nadir during hospitalization was lower in ICU patients (mean 54.8 × 109/L vs 84.1 × 109/L). Peak ferritin, measured in a subset, was markedly higher in ICU patients (median 21 666 vs 961 µg/L). The composite endpoint of HGA-related complications or serious treatment-related adverse events occurred more often in ICU patients (75% vs 17%); one death occurred in the ICU group. Conclusions In hospitalized adults with HGA, ICU-level care was associated with gastrointestinal symptoms at presentation/admission, admission hyponatremia and hyperbilirubinemia, lower platelet nadir, and marked hyperferritinemia when measured. Because these findings were derived from univariate group-level comparisons with substantial overlap between groups, they should be viewed as hypothesis-generating rather than individually predictive or sufficient as stand-alone triage criteria.