# Association of serum bicarbonate with 28-day and 90-day mortality in patients with epilepsy and concurrent sepsis: a retrospective cohort study

## TL;DR

This study finds that serum bicarbonate levels have a U-shaped relationship with mortality in ICU patients with epilepsy and sepsis, with the lowest risk at around 25 mEq/L.

## Contribution

The study is the first to identify non-linear associations between serum bicarbonate and mortality in patients with epilepsy and sepsis.

## Key findings

- Lower mortality risk was observed with each unit increase in serum bicarbonate below threshold levels at 28 and 90 days.
- Higher bicarbonate levels above thresholds were linked to increased mortality risk at 90 days.
- Kaplan-Meier analysis showed significant survival differences across bicarbonate levels, with normal levels correlating to better outcomes.

## Abstract

Serum bicarbonate concentration is a predictor of adverse outcomes in various diseases. However, its role in forecasting outcomes specifically for patients with epilepsy and concurrent sepsis remains unclear. This study examines the relationship between serum bicarbonate levels and 28-day and 90-day mortality in patients with epilepsy and sepsis who were admitted to the intensive care unit (ICU).

Clinical data from 1,271 patients with epilepsy and concurrent sepsis were retrieved from the Medical Information Mart for Intensive Care (MIMIC)-IV database (2008–2022) for retrospective analysis. The primary outcomes measured were mortality rates at 28 and 90 days. We used multivariate Cox regression analysis, restricted cubic splines, threshold effect analysis, and survival curves to assess the impact of serum bicarbonate levels on 28-day and 90-day mortality.

Mortality rates for patients with epilepsy and sepsis were 21.4% at 28 days and 28.6% at 90 days. Two distinct non-linear relationships were observed between serum bicarbonate levels and mortality rates at 28 and 90 days. Below their respective threshold points, each unit increase in serum bicarbonate was associated with a decrease in mortality [hazard ratio (HR) 0.941, 95% confidence interval (CI) 0.9–0.985, p = 0.0084 at 28 days, and HR 0.952, 95% CI 0.915–0.99, p = 0.0144 at 90 days]. Above the thresholds, increases in bicarbonate levels were linked with elevated mortality risk (HR 1.1, 95% CI 0.979–1.236, p = 0.109 at 28 days, and HR 1.112, 95% CI 1.002–1.235, p = 0.0464 at 90 days). Kaplan–Meier survival analysis showed statistically significant survival differences at 28 and 90 days across serum bicarbonate levels (p = 0.00015), with normal levels correlating with higher survival rates.

Two unique non-liner U-shaped relationships were identified between serum bicarbonate levels and mortality at 28 and 90 days in patients with epilepsy and concurrent sepsis. The lowest mortality rates were observed at approximately 25.0 and 25.9 mEq/L, respectively.