Infection Type and Class: Exploring Thrombotic and Renal Complications in a SICU Cohort

Authors:
Ioannis Karikis, Yasmin Arda, John O. Hwabejire, Michael P. DeWane, Casey M. Luckhurst, Lydia Maurer, Joshua S. Ng-Kamstra, Haytham M. Kaafarani, George C. Velmahos, Galit H. Frydman

Body of Abstract:
Background:
Infection-related thrombosis and organ injury are increasingly recognized in critically ill patients. Surgical ICU (SICU) patients may be particularly vulnerable because they combine severe physiological stress, complex operations, and high rates of invasive devices and broad-spectrum antimicrobials. We aimed to examine how pathogen burden and diversity relate to clinically significant adverse events in SICU patients with suspected infection.

Methods:
We conducted a single-center retrospective study of 137 SICU patients with suspected infection and available microbiologic data, who underwent longitudinal microbiologic sampling from 2020-2024. Organisms were grouped into four pathogen classes (Gram-positive bacteria, Gram-negative bacteria, fungi, and viruses). For each patient, we derived (1) the number of pathogen classes involved (0–4), (2) the number of distinct bacterial isolates (0, 1, 2–3, ≥4), and (3) combinations of bacteria, fungi, and viruses. The primary outcome was a major thrombotic event (deep venous thrombosis, pulmonary embolism, line thrombosis, myocardial infarction, or stroke). Secondary outcomes were AKI and hospital LOS. Associations were evaluated descriptively with univariate analysis.

Results:
Median age was 63 (51–73) years and median hospital LOS 12 (6–25) days. Overall, 22/137 patients (16%) developed an MTE and 41/132 (31%) developed AKI. Across pathogen-class strata (N=132 with complete data), MTE rates rose from 1/44 (2.3%) with no detected class to 5/15 (33.3%) with three classes and 4/5 (80.0%) with all four classes involved (p<0.001); AKI showed a similar pattern (13.6%, 19.5%, 51.9%, 60.0%, and 80.0%, respectively; p<0.001). Median LOS increased from 12 (5.5–22.5) days with no pathogen class to 23.5 (11–30) days with three classes and 60 (50–96.5) days with all four classes (p=0.003). When stratified by bacterial isolates, MTE rose from 4/80 (5.0%) with no bacterial growth to 6/20 (30.0%) with ≥4 isolates, and AKI from 14/80 (17.5%) to 8/15 (53.3%) (both p<0.001), with LOS increasing from 8.5 (5–18) to 29 (14–54) days (p=0.002). The highest rates of MTE and AKI were observed in a small subgroup with concurrent bacteria, fungi, and viruses (57.1% and 85.7%, respectively). Conclusions: In this SICU cohort, increasing pathogen burden and diversity were associated with higher rates of clinically significant adverse events, including major thrombotic events, AKI, and prolonged hospitalization in a graded fashion. A plausible explanation for these patterns is an underlying immunothrombotic process linking infection, thrombosis, and organ injury. These hypothesis-generating findings support larger studies to clarify causal pathways and to test strategies to prevent complications in high-risk surgical ICU patients, such as combinatorial and/or amplifying inflammatory pathways secondary to multimodal infectious stimuli.