Association between delta anion gap/delta bicarbonate and outcome of surgical patients admitted to intensive care unit

Abstract

Background: Patients undergoing high-risk surgeries with acid-based disorders are associated with poor outcomes. The screening of mixed acid-based metabolic disorders by calculating delta anion gap (AG)/delta bicarbonate (Bic) has a clinically relevant role in patients with high AG metabolic acidosis (MA), however its utility in individuals facing high-risk surgical procedures remains unclear.

Objective: Characterize metabolic acidosis using delta-AG/delta-Bic and its associations in patients undergoing high-risk surgeries with possible outcome-related complications.

Design: Prospective observational multicentric study.

Setting: Three tertiary hospitals in Brazil.

Patients: Patients undergoing high-risk surgeries, aged 18 years or older, requiring postoperative critical care.

Main outcome measures: Patients undergoing high-risk surgeries monitored during the postoperative phase across three distinct intensive care units (ICUs), with assessment encompassing laboratory analyses upon admission and 24 h thereafter. Patients with MA and with elevated AG within 24 h were separated into 3 subgroups: [G1 - delta-AG/delta-Bic < 1.0] MA associated with hyperchloremia; [G2 - delta-AG/delta-Bic between 1.0 and 1.6] MA and no mixed disorders; and [G3 - delta-AG/delta-Bic > 1.6] MA associated with alkalosis. Primary endpoint was 30-day mortality. The secondary endpoints were cardiovascular, respiratory, renal, neurological, coagulation and infective complications.

Results: From the 621 surgical patients admitted to ICU, 421 (51.7%) had any type of acidosis. After 24 h, 140 patients remained with MA with elevated AG (G1: 101, G2: 18, and G3: 21). When compared to patients from subgroups 1 and 3, the subgroup with no mixed disorders 2 showed higher 30-day mortality (adjusted HR = 3.72; 95% CI 1.11-12.89, p = 0.001), cardiovascular complications (p = 0.001), ICU mortality (p = 0.03) and sum of all complications during the ICU period (p = 0.021).

Conclusion: In the postoperative time, patients with metabolic acidosis and no mixed disorders present higher ICU-Mortality and higher cardiovascular postoperative complications when compared with patients with combined hyperchloremia or alkalosis. Delta-AG/delta-Bic can be a useful tool to evaluate major clinical outcomes in this population.

Keywords: Anion gap; Critical care; Intensive care; Metabolic acidosis; Surgery.

Fig. 1

Represent the Study setup according to the timeline and the respective collected data. ICU: intensive care unit

Fig. 2

Study population

Fig. 3

A Kaplan-Meier curve was constructed for the 30-day period, illustrating distinct groups: those without acidosis, those with hyperchloremic acidosis, and those with elevated anion gap. The curve’s values were derived from a Cox regression model. Additionally, a separate Kaplan-Meier curve was generated for various types of metabolic acidosis with an elevated anion gap and mixed disorders, also based on the Cox regression model

Fig. 4

(A) -Represents the percentage of patients in each group in the study timepoints 0 h and time 24 h. (B) -Represents patient Subgroups based on the delta-AG/delta-Bic sub-division. (C)- Represents pH, base differences, arterial lactate, and serum albumin levels at admission (0 h) and after 24 h (24 h). (D)-Represents the levels of chloride and sodium in mEq/L at admission and after 24 h, showing little change.(E)- the PCO₂ and bicarbonate levels in mmHg and mmol/L respectively, measured at admission and after 24 h