SCIENCE@home

ABSTRACT

The difference between serum sodium and chloride concentrations (SCD) is a simple parameter reflecting the interplay between major extracellular electrolytes. We investigated whether SCD assessed upon intensive care unit (ICU) admission holds predictive value for short-term outcomes in critically ill adults. We retrospectively analyzed 1,726 consecutive patients admitted to a mixed (medical-surgical) ICU (median age 67 years; 70.7% male). The relationship between admission SCD and 30-day mortality was first assessed using restricted cubic spline (RCS) regression to explore non-linearity. To define potential thresholds, cut-off values were evaluated using ROC analysis with the Youden index. Cox proportional hazards models were then applied. Internal validation was performed using bootstrap resampling with 1,000 iterations. RCS analysis revealed a non-linear association between SCD and mortality, with risk increasing markedly at lower values. Both spline modeling and receiver operating characteristic (ROC) analysis consistently indicated 30 mmol/L as the threshold of prognostic significance. Patients with SCD < 30 mmol/L (n = 343; 19.9%) had significantly higher 30-day mortality compared with those with SCD ≥ 30 mmol/L (55.7% vs. 44.3%, p < 0.001). In Cox regression, SCD < 30 mmol/L was associated with increased mortality (unadjusted HR 1.41, 95% CI 1.20–1.66; p < 0.001). This relationship remained robust after adjustment for age, SOFA score, lactate, and major comorbidities (adjusted HR 1.48, 95% CI 1.24–1.77; p < 0.001) and was confirmed by bootstrap validation. Low SCD at ICU admission was independently associated with higher 30-day mortality. This routinely available parameter, reflecting the sodium–chloride interplay, may support early risk stratification in this group of patients.

PUBLICATION RECORD

CITATION MAP

EXTRACTION MAP

CLAIMS

  • No claims are published for this paper.

CONCEPTS

  • No concepts are published for this paper.

REFERENCES

Showing 1-37 of 37 references · Page 1 of 1

CITED BY