An exploratory study evaluating the predictive capacity of the CALLY index and SOFA-2 for 28-day all-cause mortality...

Early risk stratification of septic shock remains challenging. Although the SOFA score is widely used, its prognostic accuracy is limited in contemporary ICU practice. The C-reactive protein-albumin-lymphocyte (CALLY) index reflects inflammation, nutritional reserve, and immune status, but its independent and incremental prognostic value beyond organ dysfunction scores remains uncertain.

In this retrospective cohort study, 204 patients with septic shock were enrolled and stratified according to 28-day mortality. Baseline clinical characteristics, laboratory parameters, and severity scores were collected. The prognostic value of CALLY, SOFA, and SOFA-2 was evaluated using univariable and multivariable logistic regression analyses. Discriminative ability was assessed using receiver operating characteristic (ROC) curves, and pairwise AUC comparisons were planned using the DeLong test. In addition, subgroup analyses were performed to explore the stability of associations between CALLY and mortality across predefined clinical subgroups.

Among 204 patients, 104 (51.0%) died within 28 days. Non-survivors presented with older age, higher lactate levels, increased inflammatory markers, worse organ dysfunction, and higher severity scores. In multivariable analysis, SOFA-2 remained independently associated with 28-day mortality (adjusted OR 1.26, 95% CI 1.08-1.50; p = 0.005), whereas CALLY was not an independent predictor. ROC analysis showed numerically better discrimination for SOFA-2 than SOFA (AUC 0.679 vs. 0.624), while the addition of CALLY provided only minimal improvement (SOFA-2 + CALLY AUC 0.686 vs. SOFA-2 AUC 0.679). Using individual-level predicted probabilities for reanalysis, the pairwise DeLong test results are shown: SOFA-2 vs. SOFA, p = 0.0086; SOFA-2 + CALLY vs. SOFA-2, p = 0.3755. Subgroup analysis demonstrated that higher CALLY levels were associated with reduced mortality risk in unadjusted subgroup comparisons, particularly among patients aged > = 65 years (P for interaction = 0.042), but this association was not consistent after adjustment for confounders. Overall, SOFA-2 showed the strongest prognostic signal in this cohort, while the incremental prognostic value of CALLY was limited.

SOFA-2 showed numerically better prognostic performance than conventional SOFA for 28-day mortality in patients with septic shock. Although CALLY was associated with outcomes in descriptive and subgroup analyses, it was not an independent predictor and did not provide a clinically meaningful incremental improvement when added to SOFA-2 in this cohort.