Diagnostics (Mar 2023)
Initial Serum Levels of Magnesium and Calcium as Predictors of Mortality in Traumatic Brain Injury Patients: A Retrospective Study
Abstract
Background: We sought to evaluate the predictor role of the initial serum level of calcium and magnesium in hospitalized traumatic brain injury (TBI) patients. Materials and methods: A retrospective analysis of all TBI patients admitted to the Hamad Trauma Center (HTC), between June 2016 and May 2021 was conducted. Initial serum electrolyte levels of TBI patients were obtained. A comparative analysis of clinical variables between patients with abnormal and normal serum electrolyte level was performed. Logistic regression analysis with the variables that showed a significant difference (p Results: There was a total of 922 patients with clinical records of serum electrolyte levels at admission. Of these, 757 (82.1%) had hypocalcemia, 158 (17.1%) had normal calcium level, and 7 (0.8%) had hypercalcemia. On the other hand, 616 (66.8%) patients had normal magnesium level, 285 (30.9%) had hypomagnesemia, and 12 (1.3%) had hypermagnesemia. The mortality rate in hypocalcemia group was 24% while in patients with normal calcium level it was 12%, p = 0.001. Proportionate mortality rates in hypomagnesemia and normal magnesium groups were 15% and 23% (p = 0.006), respectively. On the other hand, 7 out of 12 (58%) hypermagnesemia patients died during the index hospitalization. The regression model including GCS, ISS, PT, aPTT, INR, Hemoglobin, Bicarbonate, Lactate, Sodium, Potassium, Calcium, Magnesium, and Phosphate showed that hypocalcemia was not a significant predictor [OR 0.59 (CI 95%: 0.20–1.35)] of mortality after TBI. However, hypermagnesemia was a significant predictor [OR 16 (CI 95%: 2.1–111)] in addition to the GCS, ISS, aPTT, Bicarbonate, and Lactate values on admission. Conclusion: Although hypocalcemia and hypomagnesemia are common in hospitalized TBI patients, hypocalcemia was not a significant predictor of mortality, while hypermagnesemia was an independent predictor. Further studies with larger sample size and with prospective design are required to support these findings and their importance.
Keywords