Liberalization of the Systemic Glucose Management is Associated with a Reduced Frequency of Neuroglucopenia in Subarachnoid Hemorrhage Patients: An Observational Cohort Study

Abstract

Background: Pathologically low brain glucose levels, referred to as neuroglucopenia, are associated with unfavorable outcomes in neurocritical care patients. We sought to investigate whether an increase in serum glucose levels would be associated with a reduction of neuroglucopenia.

Methods: In this retrospective analysis of prospectively collected data, we included 55 consecutive patients with spontaneous subarachnoid hemorrhage who underwent cerebral microdialysis (CMD) monitoring. Neuroglucopenia was defined as CMD-glucose levels < 0.7 mmol/l. We identified systemic glucose liberalization events, defined as a day with median serum glucose levels < 150 mg/dl, followed by a day with median serum glucose levels > 150 mg/dl, and compared concentrations of cerebral metabolites between these days. Unfavorable outcome was defined as modified Rankin Scale score ≥ 3 at 3 months after the bleeding.

Results: Episodes of neuroglucopenia were more frequent in patients with unfavorable outcome (19.8% [19.3-20.3%] vs. 10.9% [10.4-11.5%], p = 0.007). Sixty-nine systemic glucose liberalization events were identified in 40 patients. Blood glucose levels increased from 141.2 (138.7-143.6) mg/dl to 159.5 (157.0-162.2) mg/dl (p < 0.001), CMD-glucose levels increased from 1.44 (1.39-1.50) mmol/l to 1.68 (1.62-1.75) mmol/l (p = 0.001), and the frequency of neuroglucopenia decreased from 24.7% (22.9-26.5%) to 20.2% (18.7-21.8%) (p = 0.002) during these events. Liberalization was not associated with changes in CMD-lactate, CMD-pyruvate, CMD-lactate-to-pyruvate ratio, CMD-glutamate, or CMD-glycerol.

Conclusions: In conclusion, the liberalization of serum glucose concentrations to levels between 150 and 180 mg/dl was associated with a significant reduction of neuroglucopenia.

Keywords: Cerebral metabolism; Cerebral microdialysis; Glucose; Neurocritical care; Subarachnoid hemorrhage.

Fig. 1

Temporal dynamics of interstitial brain glucose levels (A) and the daily frequency of pathologically low brain glucose levels (< 0.7 mmol/l), referred to as neuroglucopenia, (B) over 2 weeks after the ictus. Error bars indicate the 95% confidence interval of mean. N = 8659 CMD-samples; CMD = cerebral microdialysis; SAH = subarachnoid hemorrhage.

Fig. 2

Liberalization was associated with a significant increase of serum glucose levels (A, p < 0.001) and brain glucose levels (B, p = 0.001), as well as with a significant reduction of the frequency of pathologically low brain glucose levels (< 0.7 mmol/l), referred to as neuroglucopenia (C, p = 0.002). Error bars indicate the 95% confidence interval of mean. N = 2534 CMD-samples; CMD = cerebral microdialysis.

Fig. 3

Liberalization was not associated with statistically significant changes of the levels of CMD-lactate (A), CMD-pyruvate (B), CMD-glutamate (C), CMD-glycerol (D), the frequency of cerebral metabolic distress (E) or the frequency of the cerebral metabolic profile of mitochondrial dysfunction (F). Error bars indicate the 95% confidence interval of mean. N = 2534 CMD-samples; CMD = cerebral microdialysis.

Fig. 4

A scatter plot depicting the correlation between serum and CMD-glucose levels. N = 3784 pairs of CMD and serum glucose values; CMD = cerebral microdialysis.