Abstract
Objective
To investigate the association between the initial serum total calcium and mortality in febrile and non-febrile patients among intensive care unit (ICU).
Methods
A retrospective cohort study with the data from the MIMIC-IV (v2.2) database. The primary outcome was in-hospital mortality, and the design variable was initial serum total calcium.
Results
A total of 39701 patients were enrolled in this study, comprising 38129 survivors and 1572 non-survivors. There was a ‘U’-shaped curve relationship between initial serum total calcium and hospital mortality among non-febrile patients or febrile patients. In unadjusted analysis, hypercalcemia was significantly associated with hospital mortality (p = 0.000) in non-febrile patients and with ICU mortality (p = 0.007) in febrile patients, while no significant correlation between hypocalcemia and mortality was observed. In the logistic regression model, compared to normal serum calcium, hypercalcemia was associated with increased mortality among non-febrile patients (odds ratio (OR) = 1.904, 95% CI: 1.271–2.853, p = 0.002), whereas hypocalcemia does not exhibit a significant correlation. Neither hypercalcemia nor hypocalcemia was associated with increased mortality among febrile patients.
Conclusion
The presence of hypercalcemia in ICU patients, particularly those who are non-febrile, is significantly associated with increased mortality. Therefore, prompt treatment of hypercalcemia should be prioritized for non-febrile patients in ICU.