Anion gap is a significant predictor of 28-day all-cause mortality in patients with chronic total coronary occlusion admitted to the ICU, with the highest quartile showing markedly elevated risk (HR 21.22) compared to the lowest quartile.
Key Findings
Results
Anion gap was significantly associated with 28-day all-cause mortality in CTO ICU patients in univariate analysis.
Univariate Cox regression: HR 1.09, 95% CI 1.05–1.13, P < .001
Study population: 830 patients with CTO admitted to the ICU
Mean age was 69.4 years; 76.5% were male
Data sourced from the MIMIC-IV database (version 3.1), spanning 2008 to 2019
Results
After adjustment for clinical and laboratory parameters, anion gap remained a significant independent predictor of 28-day all-cause mortality.
Multivariate Cox regression: HR 1.15, 95% CI 1.10–1.20, P < .001
Adjustments were made for various clinical and laboratory parameters
The association strengthened slightly after multivariate adjustment compared to univariate analysis (HR 1.09 vs 1.15)
Results
Mortality risk increased across anion gap quartiles, with the highest risk observed in the fourth (highest) quartile.
Q4 (highest anion gap quartile): HR 21.22, 95% CI 2.78–161.93, P = .003 compared to the lowest quartile
Risk increased progressively across quartiles Q1 through Q4
The wide confidence interval in Q4 suggests limited sample size in that subgroup
Results
Subgroup analyses revealed significant interactions between anion gap and both gender and heart failure status on 28-day mortality.
Significant interactions were found specifically with gender and heart failure
These interactions suggest that the prognostic value of anion gap may differ by sex and presence of heart failure comorbidity
No further quantitative breakdown of subgroup HRs was provided in the abstract
Methods
The study used a retrospective cohort design drawing from a large critical care database, excluding patients with missing anion gap data or non-first-time ICU admissions.
Data source: Medical Information Mart for Intensive Care, version IV (MIMIC-IV, version 3.1)
Study period: 2008 to 2019
Final analytical sample: 830 patients with CTO admitted to the ICU
Exclusion criteria included missing anion gap data and non-first-time ICU admissions
What This Means
This research suggests that a simple blood test measurement called the anion gap — which reflects the balance of electrically charged particles in the blood and can indicate metabolic stress or acid-base disturbances — is strongly linked to the risk of dying within 28 days among patients with a condition called chronic total coronary occlusion (CTO) who are admitted to intensive care. CTO means that one or more coronary arteries are completely blocked, representing a severe form of heart disease. The study analyzed records from 830 such patients admitted to ICUs between 2008 and 2019, finding that for every unit increase in anion gap, mortality risk rose by about 15% after accounting for other health factors.
The association was particularly striking when patients were divided into four groups based on their anion gap levels: those in the highest group had roughly 21 times the mortality risk of those in the lowest group. The study also found that the relationship between anion gap and mortality differed depending on whether patients were male or female and whether they had heart failure, suggesting these factors may modify how anion gap reflects disease severity in CTO patients.
This research suggests that routinely measuring anion gap — an inexpensive and widely available test — could help clinicians identify which critically ill CTO patients are at the greatest risk of short-term death, potentially enabling earlier or more intensive intervention. However, because this was a retrospective observational study, it cannot prove that anion gap causes worse outcomes, and the authors call for further research to confirm these findings and understand the underlying biological mechanisms.
Yang J, Huang M, Zhu G, Cai Z, Tu G. (2026). Relation between anion gap and 28-day all-cause mortality in patients with chronic total coronary occlusion: A retrospective cohort analysis.. Medicine. https://doi.org/10.1097/MD.0000000000050537