DOI: 10.1097/ccm.0000000000007374 ISSN: 0090-3493

Diabetes and Atherosclerotic Cardiovascular Disease Modify the Association Between Soluble Tumor Necrosis Factor Receptor 1 and Sepsis Mortality: A Secondary Cohort Analysis

Timothy G. Gaulton, Rachel Rigsby, Nathan I. Shapiro, Eric P. Schmidt, Maurizio Cereda
Show PDF Cite

Objectives:

To determine whether diabetes mellitus or atherosclerotic cardiovascular disease (ASCVD) modify the association between soluble tumor necrosis factor receptor 1 (sTNFR1) and mortality in sepsis.

Design:

Secondary cohort analysis using the Acetaminophen and Ascorbate in Sepsis: Targeted Therapy to Enhance Recovery Research Materials.

Setting:

Academic hospitals in the United States.

Patients:

A total of 466 adults with sepsis and respiratory failure or vasopressor-dependent shock with available baseline sTNFR1 measurements.

Interventions:

None.

Measurements and Main Results:

We tested the interaction between sTNFR1 and diabetes or ASCVD (defined by Charlson comorbidity categories) on 90-day mortality using Cox regression with restricted cubic splines, adjusted for illness severity, age (in tertiles), and sex. Of 466 patients, 243 patients (52.1%) had diabetes or ASCVD. Despite similar illness severity and mortality between groups, sTNFR1 was higher among patients with diabetes or ASCVD (median, 6547 vs. 4883 pg/mL). Among patients without diabetes or ASCVD, higher sTNFR1 was associated with 90-day mortality (hazard ratio [HR], 2.25 per

sd
; 95% CI, 1.76–2.88). The association was attenuated among those with diabetes or ASCVD, with a plateau in mortality risk above approximately 12,000 pg/mL (overall HR, 1.04; 95% CI, 0.83–1.30; interaction p < 0.001). This interaction was specific to sTNFR1 and was not observed for interleukin-6 or angiopoietin-2. No effect modification was present among patients with obesity or hypertension without diabetes or ASCVD.

Conclusions:

Diabetes or ASCVD attenuated the association between sTNFR1 and sepsis mortality. These findings suggest that comorbid conditions may alter the biological relationship between a biomarker and outcome, with implications for biomarker-guided risk stratification and disease mechanisms in sepsis.

More from our Archive