Hypalbuminemia independently predicts survival after acute myocardial infarction

BACKGROUND Acute myocardial infarction (AMI) is a leading cause of morbidity and mortality, posing a significant public health challenge. Despite interventional and pharmacotherapeutic advancements, the multiplicity of short- and long-term complications underscores the need for improved treatment strategies beyond the immediate coronary intervention. Sporadic reports have associated a higher lactate/albumin ratio or uric acid/albumin ratio with enhanced risk of mortality. Literature suggests that low serum albumin associates with increased inflammation, endothelial dysfunction, increased vascular permeability, edema, lower antioxidant ability, higher blood viscosity, and higher risk of thrombosis, all of which may exacerbate heart tissue damage. Thus, we hypothesized that low albumin serum levels independently predict all-cause mortality risk in patients with AMI.

METHODS In this single-center retrospective analysis (Ethics Committee approval #20230618-01), serum albumin and routine laboratory markers was quantified in AMI patients admitted to the University Hospital of Würzburg between 2018 and 2023. Information on death was obtained from resident´s registration office for all participants between May and August 2024.

RESULTS The median age of 926 included AMI patients was 64 years (quartiles 55, 74 years) and 216 patients (23%) were women. The majority of 525 patients (57%) had suffered an ST elevation AMI. Peak creatine kinase levels were 771 U/l (285, 1660). The left anterior descending artery was the culprit vessel in the majority of patients (n=404, 44%), followed by the right coronary artery (n=322, 35%), and the circumflex artery (n=163, 18%). Coronary intervention was performed in 757 patients (82%), with drug-eluting stents used in 635 patients (84%), bare metal stents in 114 patients (15%), and drug-eluting balloons in 8 patients (1%).

Laboratory analysis revealed hypoalbuminemia (serum albumin <3.5g/dl) in 139 patients (15%) at index event. While sex (p=0.099) and body mass index (p=0.178) did not differ between groups, patients with hypalbuminemia were older (p=0.040), and had higher maximal creatine kinase (p<0.001), lactate (p<0.001), C-reactive protein (p<0.001), alanine aminotransferase (p<0.001), as well as aspartate aminotransferase (p<0.001) levels. In turn, estimated glomerular filtration rate (p<0.001), total protein levels (p<0.001) and hematocrit (p<0.001) were lower in patients with hypalbuminemia.

Within a median follow-up time of 27 months (14, 77), 134 patients (14%) had died. A multivariable Cox proportional hazards model was built adjusting for age, lactate level, sex, body mass index, peak creatinine kinase, total protein, C-reactive protein, estimated glomerular filtration rate, alanine aminotransferase, aspartate aminotransferase, hematocrit, culprit vessel, number of vessels in coronary artery disease, indication for coronary artery bypass grafting, stenting, and type of AMI. In this model, hypalbuminemia independently predicted death: hazard ratio 3.36, 95% CI 2.18-5.17; p<0.001. Predictive discrimination was good.

CONCLUSIONS Our data indicate that hypalbuminemia is an independent predictor of all-cause death in patients with AMI. In addition, hypalbuminemia may serve as a strong, previously unnoticed indicator of underlying risk, offering the advantage of being easy to assess in clinical settings.