Glycemic Variability in Critically Ill Patients: Risk Factors and Association with Mortality

0
0
0

Glycemic Variability in Critically Ill Patients: Risk Factors and Association with Mortality

Glycemic variability (GV), defined as fluctuations in blood glucose levels over time, has emerged as a critical prognostic indicator in critically ill patients. This study investigates the relationship between GV and mortality in intensive care unit (ICU) populations, explores how diabetes mellitus (DM) status modifies this association, and identifies independent risk factors for high GV.

GV and Mortality in Critically Ill Patients

The study analyzed 1,234 ICU patients categorized into four groups based on the coefficient of variation (CV) of blood glucose levels: CV <15.0%, 15.0%–30.0%, 30.0%–45.0%, and ≥45.0%. Mortality rates increased progressively with higher GV. ICU mortality rates were 8.3%, 14.3%, 21.9%, and 36.7% across the four groups, while hospital mortality rates were 11.7%, 21.9%, 29.0%, and 45.9%, respectively. Both ICU and hospital mortality differed significantly among the groups (P <0.01 for both comparisons), establishing a clear dose-response relationship between GV severity and adverse outcomes.

Impact of Diabetes Mellitus Status

Stratification by DM status revealed striking differences in outcomes. Patients were divided into four cohorts: low GV with DM, low GV without DM, high GV with DM, and high GV without DM (Figure 1A). ICU mortality rates for these groups were 8.0%, 16.0%, 19.9%, and 33.7%, while hospital mortality rates were 12.9%, 23.5%, 26.1%, and 43.2%, respectively. Non-diabetic patients exhibited significantly higher mortality than their diabetic counterparts within both low-GV and high-GV categories (P <0.01 for all comparisons). Notably, non-diabetic patients with high GV had the worst outcomes, with ICU mortality 4.2 times greater than diabetic patients with low GV.

The interaction between DM status and GV persisted after adjusting for illness severity. APACHE II scores, a measure of disease acuity, were significantly lower in diabetic patients across GV categories (Figure 1B). Despite this, non-diabetic patients consistently demonstrated higher mortality at equivalent GV levels, suggesting intrinsic differences in physiological responses to glucose fluctuations between these populations.

Independent Risk Factors for High GV

Multivariate logistic regression identified five independent predictors of high GV:

  1. APACHE II Score: Each point increase in APACHE II score raised the odds of high GV by 7.4% (OR 1.074, 95% CI 1.047–1.102; P <0.001).
  2. Female Gender: Women had 69.8% higher odds of high GV compared to men (OR 1.698, 95% CI 1.288–2.239; P <0.001).
  3. Mechanical Ventilation: Ventilated patients showed 65.8% greater odds of high GV (OR 1.658, 95% CI 1.157–2.375; P =0.006).
  4. Diabetes Mellitus: Diabetic status increased high GV risk by 42.9% (OR 1.429, 95% CI 1.078–1.898; P =0.013).
  5. Serum Creatinine: Each unit increase in creatinine level elevated high GV odds by 11.9% (OR 1.119, 95% CI 1.015–1.233; P =0.024).

These findings highlight that both acute physiological stress (reflected by APACHE II scores and mechanical ventilation) and chronic metabolic conditions (DM and renal dysfunction) contribute to glucose instability in critical illness.

Clinical Implications and Mechanistic Considerations

The differential impact of GV on diabetic versus non-diabetic patients may stem from distinct adaptive mechanisms. Diabetic individuals often develop physiological tolerance to glucose fluctuations through chronic exposure, whereas acute GV in non-diabetics likely represents a maladaptive response to severe illness. This is supported by the higher APACHE II scores in non-diabetic patients with high GV (Figure 1B), suggesting that GV in this population serves as a marker of systemic dysregulation.

The study’s identification of female gender as a GV risk factor warrants further investigation. Potential explanations include hormonal influences on glucose metabolism, sex-specific differences in immune responses, or variations in body composition affecting insulin sensitivity. The association between elevated creatinine and GV emphasizes the interplay between renal dysfunction and glucose homeostasis, possibly mediated through reduced insulin clearance or uremia-induced insulin resistance.

From a clinical perspective, these results argue for tailored glycemic management strategies. While tight glucose control remains controversial in critical care, minimizing GV may represent a safer therapeutic target, particularly in non-diabetic patients. The data suggest that GV monitoring could enhance risk stratification, with CV ≥45% serving as a critical threshold associated with near 40% ICU mortality.

Methodological Strengths and Limitations

With 1,234 patients, this study provides robust estimates of GV-related mortality risks across DM subgroups. The use of CV as a GV metric accounts for proportional glucose fluctuations, making it particularly suitable for comparing patients with different baseline glucose levels. However, the retrospective design limits causal inferences, and the single-center nature may affect generalizability. Additionally, the study did not examine specific interventions affecting GV, leaving open questions about optimal management approaches.

Conclusion

This comprehensive analysis establishes glycemic variability as an independent predictor of mortality in critically ill patients, with particularly strong prognostic value in non-diabetic individuals. The identified risk factors—APACHE II score, female gender, mechanical ventilation, DM status, and renal function—provide a framework for identifying high-risk patients. These findings underscore the need for GV-focused monitoring and intervention strategies in critical care settings, while highlighting the importance of considering diabetic status in prognostic assessments.

doi: 10.1097/CM9.0000000000000686

Was this helpful?

0 / 0