Tip: Try author name, DOI (10.xxxx/…), or keywords.

ISSN (Online): 1694-4674
  1. Home
  2. Vol. 05, No. 07, (2026)
  3. Prognostic Value of Serum Magnesium and Glycemic Gap in Critically ill
Original Article Open Access

Prognostic Value of Serum Magnesium and Glycemic Gap in Critically ill Adults: A Prospective Observational Study

,,,
Annals of Medicine and Medical SciencesVol. 05, No. 07, (2026) July 3, 2026pp. 938 - 942

Abstract

Background: Stress hyperglycemia and magnesium disturbances are common in critically ill patients and have been associated with adverse clinical outcomes. The glycemic gap, which reflects acute hyperglycemia relative to chronic glycemic status, may provide additional prognostic information in critically ill patients. Objective: To evaluate the association of admission serum magnesium levels and glycemic gap with mortality, mechanical ventilation requirements, and duration of intensive care and hospital stay among critically ill adults. Methods: This prospective observational study included 106 critically ill adults admitted to a tertiary-care intensive care unit with APACHE II scores >10. Admission serum magnesium, HbA1c, and random blood glucose levels were measured. Estimated average glucose (eAG) was calculated from HbA1c, and glycemic gap was defined as the difference between admission blood glucose and eAG. Associations between metabolic parameters and clinical outcomes were analyzed using Chi-square and correlation tests, with p<0.05 considered statistically significant. Results: The mortality rate was 57.5%, and 69.8% of patients required mechanical ventilation. Abnormal serum magnesium levels were observed in 59.4% of patients. Serum magnesium demonstrated significant associations with mortality (p=0.001), requirement for mechanical ventilation (p=0.017), and duration of ventilatory support (p=0.043). No significant association was observed between serum magnesium and ICU stay duration (p=0.420). Glycemic gap showed no statistically significant association with mortality (p=0.116), duration of ventilation (p=0.331), ICU stay (p=0.729), or hospital stay (p=0.439). Conclusion: Admission serum magnesium levels were significantly associated with mortality and ventilatory outcomes in critically ill patients, whereas glycemic gap was not significantly associated with major clinical outcomes. Serum magnesium may represent a useful biomarker for early risk stratification in critically ill adults.

Keywords

Magnesium Hyperglycemia Critical Illness Intensive Care Units Respiration Artificial Mortality.

Introduction

Metabolic derangement remains a central hallmark of critical illness, where the collapse of homeostatic regulation frequently precipitates multi-organ dysfunction syndrome. Among these, acute dysglycemia stands as a potent surrogate for illness severity, reflecting a complex interplay between systemic stress and pre-existing metabolic reserves [1]. While historical management prioritized intensive insulin therapy for all hyperglycemic patients, contemporary paradigms recognize that the prognostic significance of elevated glucose is context-dependent, necessitating an assessment of a patient’s glycemic baseline. Consequently, clinical focus has shifted toward derived metrics like the glycemic gap (or Stress Hyperglycemia Ratio), which quantifies the deviation of admission blood glucose from the estimated average glucose (eAG) derived from glycated hemoglobin (HbA1c) [1-2].

A widened glycemic gap signifies an intense neuroendocrine and inflammatory surge, characterized by the hypersecretion of counter-regulatory hormones such as cortisol and catecholamines and pro-inflammatory cytokines. This acute metabolic burden triggers a cascade of oxidative stress, endothelial damage, and immune dysfunction. Accumulating evidence suggests that the glycemic gap is a superior, independent predictor of adverse outcomes, including acute kidney injury, prolonged mechanical ventilation, and increased mortality, often outperforming absolute glucose or HbA1c levels in predictive modelling [3].

Parallel to glycemic volatility, electrolyte disturbances profoundly influence the pathophysiological trajectory of ICU patients. Magnesium, a critical intracellular cation, serves as an essential cofactor for over 600 enzymatic reactions governing energy metabolism, membrane stability, and cardiovascular integrity. Its role in glucose regulation is particularly vital, as it modulates insulin receptor tyrosine kinase activity and key glycolytic enzymes [4]. Despite its importance, hypomagnesemia affects up to 60% of the critically ill, driven by gastrointestinal losses, renal wasting often exacerbated by hyperglycemia and polypharmacy and intracellular shifts [5].

Notwithstanding these individual insights, a significant knowledge gap remains regarding the synergistic relationship between admission magnesium levels and glycemic variation. Few study has yet integrated these parameters to evaluate their collective prognostic value for hard clinical endpoints in a heterogeneous ICU population. Addressing this deficit is imperative for transitioning toward multidimensional metabolic risk profiling. Accordingly, the current study investigates the correlation between the glycemic gap and serum magnesium levels at admission and their combined association with pivotal clinical outcomes.

Materials and Methods

Study Design and Setting

The prospective observational study was conducted at the Department of Medicine at tertiary healthcare institute. The study protocol was designed to evaluate the metabolic and prognostic profiles of critically ill patients admitted to the Intensive Care Unit (ICU) between July 2024 to December 2025. Ethical clearance was obtained from the Institutional Review Board (SGRD/IEC/2024-367), and informed consent was secured from all participants or their legal representatives.

Participants and Eligibility

A total of 106 adult patients were consecutively enrolled based on a high-acuity threshold, defined by an APACHE II score>10 and an age range of 18–80 years. To ensure a clean metabolic baseline and minimize confounding variables affecting magnesium or glucose levels- Metabolic/Endocrine Factors: Diagnosis of diabetic ketoacidosis or hyperosmolar hyperglycemic state, and pregnancy-related eclampsia; Pharmacological Interferences: Prior administration of magnesium supplements, diuretics, aminoglycosides, cisplatin, corticosteroids, or amphotericin; Clinical Confounders: Pre-admission blood transfusions or mortality within 24 hours of ICU admission were excluded from the present study.

Data Collection and Biochemical Analysis

Upon admission, comprehensive clinical evaluations were performed, and baseline physiological parameters were recorded to calculate APACHE II scores. Venous blood samples were collected within 24 hours of admission into EDTA and clot-activating tubes. Analysis was performed via automated laboratory systems for routine hematological and biochemical profiles. Specifically, HbA1c was quantified using High-Performance Liquid Chromatography (HPLC). The Estimated Average Glucose (eAG), or Average Daily Glucose (ADAG), was derived from HbA1c values to represent the chronic glycemic baseline. The glycemic gap was subsequently calculated as the difference between the initial emergency department blood glucose and the calculated ADAG.

eAG = (28.7 × HbA1c) − 46.7

Then: Glycemic Gap = Admission Blood Glucose − eAG

Serum magnesium levels were measured concurrently to assess electrolyte status at the point of acute insult.

Outcome Measures

The primary outcomes included: Mortality: In-hospital survival status; Respiratory Support: Requirement for and duration of mechanical ventilation; Resource Utilization: Length of stay in the ICU and total hospital duration.

Statistical Analysis

Statistical analysis was performed using Microsoft Excel 2019 (Microsoft Corp, Redmond, WA, USA). Continuous variables were expressed as mean±SD or median (range), and categorical variables as counts (%). Associations between categorical variables were assessed by Chi-square test, and correlations between continuous variables by Pearson’s r. A two-sided p<0.05 was considered significant.

Results

The demographic and clinical profile of the study cohort (n=106) reveals a predominantly middle-aged to elderly population (Table 1), with nearly three-quarters of participants aged 50 years or older. While the vast majority of cases were non-infective in origin (87.7%), a significant metabolic burden was evident: over 60% of the subjects exhibited abnormal glycemic status comprising both prediabetics and diabetics and 34.9% presented with hyperglycemia at admission. Notably, serum magnesium derangements were highly prevalent, affecting nearly 60% of the cohort, with hypomagnesemia (34.9%) and hypermagnesemia (24.5%) occurring at comparable frequencies. These findings underscore a substantial intersection between dysglycemia and magnesium homeostasis within the clinical presentation of the sampled population.

Table 1 Baseline Demographics and Clinical Characteristics (n=106)
Parameter Category Frequency (n) Percentage (%)
Age Group 18–49 years 28 26.41
50–69 years 51 48.11
≥70 years 27 25.5
Diagnosis Non-infective 93 87.7
Infective 13 12.3
Glycemic Status Normal (HbA1c <5.7) 42 39.6
Prediabetic (5.7–6.4) 18 17.0
Diabetic (≥6.5) 46 43.4
RBS at Admission Euglycemia (75–200 mg/dL) 64 60.4
Hyperglycemia (>200 mg/dL) 37 34.9
Hypoglycemia (<75 mg/dL) 5 4.7
Serum Magnesium Normal (1.7–2.3 mg/dL) 43 40.6
Hypomagnesemia (<1.7) 37 34.9
Hypermagnesemia (>2.3) 26 24.5

The study population exhibited a high level of physiological instability and clinical complexity, characterized by a mean APACHE II score of 18.41±5.92 and significant neurological impairment, with GCS scores averaging 10.93±4.28 (Table 2). Metabolic and haemodynamic derangements were evident across the cohort, featuring a broad glycemic range (RBS up to 540mg/dL) and substantial leucocyte variability (median WBC 48,270/µL), alongside a mean arterial pressure of 80.18±21.46 mmHg. Despite the severity of illness indicated by these parameters, arterial pH remained relatively compensated at 7.33±0.13, while the median respiratory rate of 22 breaths per minute reflected a prevailing tachypnoeic trend consistent with the critical nature of the patient profiles.

Table 2 Physiological Severity and Vital Parameters
Parameter Central Tendency (Mean±SD/Median) Range (Min–Max)
APACHE II Score 18.41±5.92 10-33
GCS Score 10.93± 4.28 3-15
Arterial pH 7.33±0.13 6.89-7.62
RBS (mg/dL) 171.51± 100.75 45-540
MAP (mmHg) 80.18±21.46 70-124
WBC Count (/µL) 14,270 (Median) 4,000-96,000
Respiratory Rate 22 (Median) 12-44

Table 3 represents the clinical outcome analysis of the 106-patient cohort reveals a high-acuity profile, characterized by a 57.5% mortality rate (n=61) and a significant reliance on mechanical ventilation (69.8%). Resource utilization was predominantly concentrated within the first week of admission, as 68.9% of patients required an ICU stay of ≤7 days and 50.9% concluded their total hospital course within the same timeframe. Notably, the near-even distribution in hospital stay duration with 49.1% of patients remaining hospitalized for over a week—underscores a polarized recovery or stabilization pattern among survivors versus those with rapid clinical deterioration.

Table 3 Clinical Outcomes and Resource Utilization
Outcome Category Frequency (n) Percentage (%)
Mortality Died 61 57.5
Survived 45 42.5
Ventilation Required 74 69.8
Not Required 32 30.2
Duration of ICU Stay ≤7 Days 73 68.9
>7 Days 33 31.1
Duration of Hospital Stay ≤7 Days 54 50.9
>7 Days 52 49.1

Table 4 represents statistical analysis identifies serum magnesium as a primary prognostic indicator, demonstrating a highly significant association with mortality (p=0.001) and the subsequent requirement for mechanical ventilation (p=0.017). Furthermore, magnesium levels significantly influenced the duration of ventilatory support (p=0.043), whereas no such correlation was observed regarding the length of ICU residency (p=0.420). Conversely, the glycemic gap failed to reach statistical significance across all measured clinical endpoints, including mortality (p=0.116) and hospital stay duration (p=0.439), suggesting that while electrolyte imbalances are critical drivers of acute outcomes in this cohort, glycemic variance did not independently dictate the trajectory of recovery or survival.

Table 4 Statistical Associations (Bivariate Analysis)
Independent Variable Dependent Variable (Outcome) p-Value Significance
Serum Magnesium Mortality 0.001 Significant
Need for Ventilation 0.017 Significant
Duration of Ventilation 0.043 Significant
Duration of ICU Stay 0.420 Non-Significant
Glycemic Gap Mortality 0.116 Non-Significant
Duration of Ventilation 0.331 Non-Significant
Duration of ICU Stay 0.729 Non-Significant
Duration of Hospital Stay 0.439 Non-Significant

Discussion

In this prospective observational study involving 106 critically ill adults, admission serum magnesium levels demonstrated significant associations with mortality, requirement for mechanical ventilation, and duration of ventilatory support. In contrast, glycemic gap did not show significant associations with major clinical outcomes. These findings suggest that disturbances in magnesium homeostasis may have greater prognostic relevance than acute glycemic variability in this heterogeneous ICU population. Our findings indicate a high prevalence of magnesium derangement (59.4%), with hypomagnesemia (34.9%) and hypermagnesemia (24.5%) occurring frequently alongside a significant metabolic burden of dysglycemia (60.4%). This high-acuity profile, marked by a mean APACHE II score of 18.41±5.92, resulted in a 57.5% mortality rate and a substantial 69.8% reliance on mechanical ventilation. Notably, our statistical analysis identified serum magnesium as a significantly associated with adverse clinical outcomes, demonstrating a highly significant association with mortality (p=0.001) and the subsequent requirement for ventilatory support (p=0.017).

The adverse impact of magnesium deviation particularly hypermagnesemia aligns with recent large-scale evidence linking elevated levels to increased 28-day mortality in septic and critically ill populations [6-8]. While literature such as the multicenter study by Li et al.,[6] validates hypermagnesemia as an independent risk factor for death, our study extends this relevance to a heterogeneous ICU population where admission magnesium levels significantly influenced the duration of ventilatory support (p=0.043) [6,9]. This suggests that magnesium derangement is not merely a bystander but may aid risk stratification for physiological decompensation. Interestingly, while magnesium dictated survival and respiratory needs, it did not correlate with the length of ICU residency (p=0.420), suggesting that in the most severe cases, clinical outcomes are determined rapidly within the first week of admission.

Mechanistically, the association between elevated magnesium and poor outcomes likely reflects an intersection of intense inflammatory responses and acute organ failure. In sepsis and critical illness, hypermagnesemia often coincides with impaired renal clearance and elevated inflammatory markers such as CRP and the neutrophil–lymphocyte ratio (NLR) [8-10]. In our cohort, the presence of severe neurological impairment (Mean GCS 10.93±4.28) and hemodynamic instability (Mean MAP 80.18±21.46 mmHg) further a magnesium abnormalities may reflect greater physiological derangement and illness severity. Paradoxically, while the glycemic gap a traditional marker of acute stress was prevalent in our study, it failed to reach statistical significance regarding mortality (p=0.116) or hospital stay (p=0.439). This suggests that in high-acuity settings, electrolyte homeostasis provides more robust independent prognostic information than conventional glycemic indices [11,12].

Integrating serum magnesium with inflammatory markers significantly enhances predictive performance. Recent data indicate that combining magnesium with CRP or NLR achieves higher discriminative ability (AUC) than individual indices alone [12,13]. In our study, magnesium's highly significant correlation with hard endpoints (p<0.05) reinforces the necessity of moving toward a multi-biomarker risk assessment strategy. Given that magnesium testing is rapid, inexpensive, and routine, its incorporation into initial triage especially for elderly patients with diminished physiological reserves could identify high-risk individuals who require aggressive monitoring or early respiratory intervention [14,15]. Ultimately, serum magnesium should join the clinical "toolbox" alongside traditional inflammatory markers to refine early risk stratification and improve management in the vulnerable critically ill population.

This study evaluated two clinically relevant metabolic parameters—serum magnesium and glycemic gap—in a heterogeneous cohort of critically ill adults. The prospective design, inclusion of objective biochemical measurements obtained at admission, and assessment of clinically meaningful outcomes including mortality and mechanical ventilation enhance the clinical relevance of the findings. Furthermore, serum magnesium estimation is inexpensive, readily available, and easily applicable in resource-limited settings.

Several limitations should be acknowledged. First, this was a single-center study with a relatively modest sample size, which may limit the generalizability of the findings. Second, multivariable regression analysis was not performed; therefore, potential confounding by disease severity and comorbidities cannot be excluded. Third, serum magnesium was measured only at admission, and serial measurements were not available. Finally, the heterogeneous nature of ICU diagnoses may have influenced outcome assessment.

Conclusion

Admission serum magnesium abnormalities were significantly associated with mortality and ventilatory outcomes among critically ill adults, whereas glycemic gap was not significantly associated with major clinical outcomes. These findings support the potential role of serum magnesium as an accessible biomarker for early risk stratification in critically ill patients. Larger multicenter studies incorporating multivariable analyses and serial magnesium measurements are warranted to validate these observations and clarify their clinical implications.

Declarations

Ethical Approval and Consent to Participate

All procedures performed in this study were carried out in accordance with the ethical standards of the Institutional Ethics Committee and the principles of the 1964 Declaration of Helsinki and its subsequent amendments. Ethical clearance was obtained from the Institutional Ethics Committee (Approval No. SGRD/IEC/2024-367). Written informed consent was obtained from all participants or their legally authorized representatives prior to inclusion in the study.

Consent for Publication

Written informed consent for publication of clinical information and relevant images was obtained from the patients or their legal representatives. All identifying information was removed to maintain patient confidentiality and privacy.

Availability of Data and Materials

The data generated and analysed during the present study are available from the corresponding author upon reasonable request, subject to institutional and ethical guidelines.

Competing Interests

The authors declare that there are no competing financial or non-financial interests related to this study.

Funding

No financial support or specific funding was received from any public, commercial, or non-profit funding agency for the conduct of this study.

Authors’ Contributions

All authors contributed to the conception and design of the study, data collection, analysis, and interpretation of results. All authors were involved in drafting and revising the manuscript critically for important intellectual content and approved the final version prior to submission.

References

  1. Marik PE, Bellomo R. Stress hyperglycemia: an essential survival response! Crit Care. 2013;17(2):305. Google Scholar ↗
  2. Dungan KM, Braithwaite SS, Preiser JC. Stress hyperglycaemia. Lancet. 2009;373(9677):1798-807. Google Scholar ↗
  3. Bai H, Niu D, Li B, Zong Y, Wang Y. Elevated stress hyperglycemia ratio associated with higher hospital mortality in patients with respiratory failure. Sci Rep. 2025;15(1):27972. Google Scholar ↗
  4. Roopa RT, Sujeesh HS. Serum magnesium as a prognostic factor in critically ill patients and its correlation with APACHE 2 score. Int J Acad Med Pharm. 2024;6(1):1639-43. Google Scholar ↗
  5. Şen İA. Prognostic value of admission serum magnesium and systemic inflammation indices in septic ICU patients aged 85 years and older: A retrospective observational study. Medicine. 2025 Oct 17;104(42):e44979. Google Scholar ↗
  6. Li L, Li L, Zhao Q, Liu X, Liu Y, Guo K, Zhang D, Hu C, Hu B. High serum magnesium level is associated with increased mortality in patients with sepsis: an international, multicenter retrospective study. MedComm. 2024;5(10):e713. Google Scholar ↗
  7. Yi X, Yang X, Li D, Zheng M, Yang J, Liu B, Huang S, Shen X, Jin Y. Association between serum magnesium and all-cause mortality in patients with sepsis-associated liver injury: a cohort study. 2025. Google Scholar ↗
  8. Li L, Zhou S, Huang Q, Hu H, Wu J, Chen Z, An S, Zeng Z. Active supplementation with magnesium sulfate may be associated with worse prognosis in sepsis: a retrospective analysis of the MIMIC-IV database. 2021. Google Scholar ↗
  9. Saglietti F, Girombelli A, Marelli S, Vetrone F, Balzanelli MG, Tabaee Damavandi P. Role of magnesium in the intensive care unit and immunomodulation: a literature review. Vaccines (Basel). 2023;11:1122. Google Scholar ↗
  10. Raza D, Bzowej N, Rini J, Rhinehart R, Samant H. Navigating diagnostic challenges: a case of acquired hepatocerebral degeneration presenting as worsening hepatic encephalopathy in chronic liver disease. Eur J Case Rep Internal Med. 2024;11:004629. Google Scholar ↗
  11. Tan L, Xu Q, Li C, Liu J, Shi R. High-normal serum magnesium and hypermagnesemia are associated with increased 30-day in-hospital mortality: a retrospective cohort study. Front Cardiovasc Med. 2021;8:625133. Google Scholar ↗
  12. Liang L, Su Q. Systemic immune-inflammation index and the short-term mortality of patients with sepsis: a meta-analysis. Biomol Biomed. 2025;25:1. Google Scholar ↗
  13. Luka S, Golea A, Tat RM, Lupan Mureșan EM, Voicescu GT, Vesa SC, et al. Biomarkers as predictors of mortality in sepsis and septic shock for patients admitted to emergency department: who is the winner? A prospective study. J Clin Med. 2024;13:5678. Google Scholar ↗
  14. Ibarz M, Haas LEM, Ceccato A, Artigas A. The critically ill older patient with sepsis: a narrative review. Ann Intensive Care. 2024;14:6. Google Scholar ↗
  15. Gu W-J, Duan X-J, Yin H-Y. Association of magnesium sulfate use with mortality in critically ill patients with sepsis. Response to Br J Anaesth 2024; 132: 187. Br J Anaesth. 2024;132:442–3. Google Scholar ↗