Risk Potential for Organ Dysfunction Associated With Sodium Bicarbonate Therapy (SBT) in Critically Ill Patients

Study Population We obtained data from the MIMIC-III. The newest version of MIMIC-III, v. 1.4, integrates 61,532 patients admitted to the ICUs of Beth Israel Deaconess Medical Center in Boston, Massachusetts.

Our study included patients with severe metabolic acidosis within 48 hours after ICU admission who fulfilled all of the following criteria: (1) pH≤7.3; (2) BE≤-8, or HCO3- < 20 mmol/L if BE was missing; and (3) PaCO2< 50 mmHg, or without chronic obstructive pulmonary disease (COPD) in discharge diagnoses if PaCO2 was unavailable. The lowest values of pH, BE, and HCO3- and the highest value of PaCO2 were selected when there were multiple measurements. We defined a composite of newly "developed or exacerbated organ dysfunction" (d/eOD) as a net increase in Sequential Organ Failure Assessment (SOFA) score ≥2 of at least one organ within 7 days of ICU admission over the baseline (SOFA score in the first 24 hours). Hemodynamic status during the initial resuscitation was defined as hemodynamic improvement, hemodynamic worsening and unchanged hemodynamics.

Variables Characteristic data of patients were extracted, including age and sex; category of diseases (surgical vs medical), history of comorbidity and whether cardiopulmonary resuscitation (CPR) was experienced; and diagnosis of sepsis, shock and AKI. Supportive interventions such as mechanical ventilation (MV) and RRT were recorded. Fluid balances at 24 hours and 48 hours after ICU admission and the total SB dosage administered within the first 48 hours were extracted. The SOFA score including each component was calculated every 24 hours if data were available. The Simplified Acute Physiology Score II (SAPSII) was calculated within the first 24 hours after ICU admission.

Laboratory variables included pH, actual bicarbonate (AB), base excess (BE), PaO2, PaCO2, hematocrit (Hct), hemoglobin (Hb), lactate concentration and electrolytes such as [Na+], [K+], and [Cl-]. We also calculated the anion gap (AG) by the formula AG=[Na+] + [K+] - [Cl-] - [HCO3-] .

A composite of newly d/eOD within 7 days of ICU admission was used as the primary outcome of this study. The secondary outcome was 28-day mortality.

Statistical Analysis Propensity score matching (PSM) was applied to minimize either selection biases or potential confounders for SBT. Continuous variables are expressed as the mean (standard deviation) or median (upper quantile, lower quantile), and a t-test or the Wilcoxon-test was used to compare the differences between groups. Categorical variables are depicted as the number of groups (the number of percentages), and the chi-squared test or Fisher's exact test was applied.Univariate and multivariat logistic regression and Cox regression were used to evaluate the effectiveness of SBT.

All statistical analyses were performed by R (version 3.6.1). A p value less than 0.05 was indicative of statistical significance.

Inclusion criteria

• (1) pH≤7.3; (2) BE≤-8, or HCO3- < 20 mmol/L if BE was missing; and (3) PaCO2< 50 mmHg, or without chronic obstructive pulmonary disease (COPD) in discharge diagnoses if PaCO2 was unavailable.

Exclusion criteria

• (1) age 16 or under; (2) diabetes insipidus ; (3) comorbidity of peripheral paralysis; (4) ICU stay over 100 days; (5) SBT was initiated before ICU admission , later than 48 hours after ICU admission or after a record of order for renal replacement therapy (RRT); and (6) death within the first 24 hours after ICU admission.