Blood Glucose and Lactate Levels and Cerebral Oxygenation in Preterm and Term Neonates—A Systematic Qualitative Review of the Literature

Background: Cerebral oxygenation monitored non-invasively by near-infrared spectroscopy (NIRS) is of increasing interest in neonatal care. Cerebral oxygenation is determined by cerebral oxygen delivery and cerebral oxygen consumption. Oxygen delivery as well as oxygen consumption might be influenced by metabolic parameters like blood glucose and lactate. Objective: The aim of the present systematic qualitative review is therefore to identify and summarize all studies, which describe cerebral oxygenation measured with NIRS and blood glucose and/or blood lactate levels in neonates. Data sources: A systematic search of Ovid Embase and PubMed was performed. Search terms included near-infrared spectroscopy, fractional tissue oxygen extraction, cerebral tissue oxygen saturation, regional cerebral tissue oxygen saturation, oxygenation, term, and preterm neonates, cesarean delivery, transition, after-birth, newborn, vaginal delivery, cesarean delivery, baby, neonatal transition, metabolism, lactate, glucose, and blood glucose level. Study selection/data synthesis: Studies analyzing cerebral oxygenation and blood glucose and/or blood lactate levels in neonates were included. Animal studies, duplicates, or studies in non-English language were excluded. Results: Twenty-five studies were identified that describe blood glucose and/or blood lactate levels as primary or secondary outcome parameters with additional measured cerebral oxygenation by NIRS in neonates. Twelve studies were included with blood glucose measurements: four described an association between blood glucose levels and cerebral oxygenation, two show no association, and six do not report on possible associations. Eighteen studies were included with lactate measurements: one describe an association between lactate levels and cerebral oxygenation, while three show no association and 14 do not report on possible associations. Discussion: The influence of blood glucose and blood lactate levels on the cerebral oxygenation in neonates is still controversial. However, there seems to be an association between cerebral oxygenation and the metabolic parameter blood glucose and lactate, which need further investigation.

Discussion: The influence of blood glucose and blood lactate levels on the cerebral oxygenation in neonates is still controversial. However, there seems to be an association between cerebral oxygenation and the metabolic parameter blood glucose and lactate, which need further investigation.

INTRODUCTION
Irreversible cerebral injury due to impaired cerebral oxygenation is a persisting problem in the neonatal period despite improved monitoring and intervention options. Standard noninvasive monitoring in neonatal care does not yet assess cerebral oxygenation, oxygen delivery to the brain, or cerebral oxygen consumption (1)(2)(3). However, cerebral near-infrared spectroscopy (NIRS) monitoring has the potential to detect impaired cerebral oxygenation in neonates while other vital parameters such as arterial oxygen saturation or heart rate remain within their normal range (4). NIRS is a continuous, non-invasive monitoring technique to measure the cerebral oxygenation in neonates and measures the cerebral regional oxygen saturation and fractional tissue oxygen extraction. A recently published multicenter trial using cerebral NIRS monitoring to reduce the burden of cerebral hypoxia in preterm neonates described beside cardiovascular and respiratory interventions also interventions based on blood glucose levels (5). Another recently published study describe an association between blood glucose level and cerebral oxygenation in preterm and term neonates immediately after birth (6). Further, lactate as a product of anaerobic metabolism might be associated with hypoxic conditions in the tissue. An association between the blood lactate level and the cerebral oxygenation has been described in extremely preterm neonates during the 1st days after birth (7).
The aim of the present systematic qualitative review is therefore to identify and summarize all studies, which describe cerebral oxygenation measured with NIRS and blood glucose and/or blood lactate levels in neonates.

Search Strategy and Selection Criteria
Studies were identified using the stepwise approach specified in the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) Statement (8).

Eligibility Criteria
Studies had to address cerebral oxygenation measurements with NIRS as well as the metabolic parameters blood glucose and/or lactate in neonates.

Search Strategy
A systematic search of Ovid Embase and PubMed NCBI was performed to identify studies in English language published between 1974 and November 2019. Search terms included near-infrared spectroscopy, fractional tissue oxygen extraction, cerebral tissue oxygen saturation, regional cerebral tissue oxygen saturation, oxygenation, term, and preterm neonates, cesarean delivery, neonatal transition, after-birth, newborns, vaginal delivery, baby, after cesarean delivery, metabolism, lactate, glucose, and blood glucose level.

Inclusion and Exclusion Criteria-Population
To be eligible, studies had to investigate human neonates. Neonates were defined as infants with a postnatal age of <28 days. Studies that included neonates and infants or children were also included in our analysis, when the results were not separately analyzed for neonates. Animal studies were excluded.

Inclusion and Exclusion Criteria-Measurements (Exposure)
We included studies with different NIRS devices, if any additional measurements of either capillary, venous, or arterial blood glucose levels and/or lactate levels were included.

Inclusion and Exclusion Criteria-Types of Publication
We included clinical or observational studies published in English language. Non-original articles, such as comments, book chapters, editorials, reviews, and methods papers, were excluded. Duplications and publications in non-English languages were also excluded.

Study Selection
The articles identified in the literature review were evaluated independently by two authors (CM and GP) for inclusion using the titles and abstracts. Then, full texts were retrieved and were included based on the eligibility criteria. Any disagreement was resolved through discussion and consensus between two authors. If there was uncertainty regarding eligibility for inclusion on the basis of the abstract, the full text was assessed too. Data were analyzed qualitatively. Data extraction included the study design, characterization of type (preterm/term) and number of neonates included in the study, applied device, NIRS and metabolic values, age of neonates during NIRS and metabolic measurements, and the presence or absence of any association.

Risk of Bias in Individual Studies
A longer period between cerebral measurements and blood samples may originate a bias. Therefore, we included the exact time between cerebral measurements and blood samples in our qualitative analysis.

RESULTS
After the initial search, 978 abstracts were identified, which were assessed for eligibility. After full text search, 25 studies remained to be included in the present review (Figure 1) (6,7,.
None of the studies reported on possible simultaneous associations between both metabolic parameters (glucose and lactate) and cerebral oxygenation.

DISCUSSION
In the last few years, interest into research of cerebral oxygenation and metabolic parameters during the neonatal period increased significantly. There are several studies describing results of possible or missing association between metabolic parameters and cerebral oxygenation measured with NIRS. These results are controversial.
Frontiers in Pediatrics | www.frontiersin.org of cerebral oxygenation with increasing blood glucose levels. However, two studies described no association (9,11). Naulears et al. (9) described an increase of cerebral oxygenation from day 1 to 3 after birth in neonates with postmenstrual age of 28 weeks. In this cohort, the multiple regression analysis showed no correlation between tissue oxygenation index and glycemia.
In the largest cohort of neonates described by Weiss et al. (11) no association between blood glucose and cerebral oxygenation was observed. Interestingly, there was a negative association of blood glucose level with cerebral oxygenation observed in neonates after a Norwood procedure (14). Jia et al. (16) described a negative association between hyperglycemia and oxygen delivery. Further, she described a positive association between hyperglycemia and oxygen extraction ratio in neonates 72 h after Norwood procedure. Mattersberger et al. (6) demonstrated that blood glucose levels have a negative correlation to the cerebral oxygen saturation and a positive correlation to the cerebral fractional tissue oxygen extraction in preterm and term neonates 15 min after birth. Cerebral hemoglobin concentration that influences cerebral oxygenation, measured with NIRS, was investigated by Von Siebental K in neonates in the first 6 h of life. (12) He described different parameters influencing the cerebral hemoglobin concentration of neonates, whereby blood glucose had a negative correlation with cerebral hemoglobin concentration. The changes in cerebral hemoglobin concentration are in accordance with the above-described negative association between cerebral oxygenation and blood glucose levels when taking into account an auto-regulatory mechanism to maintain glucose supply to the brain. With decreasing blood glucose levels, there might be an increase in cerebral hemoglobin volume/concentration by increase of cerebral blood flow due to vasodilatation. This causes an increase in oxygen delivery with increase in cerebral oxygenation in case of a consistent cerebral oxygen consumption.

Lactate Level and Cerebral Oxygenation
High lactate levels might be associated with an adverse neurologic outcome and can be a predictor for short-term neonatal adverse outcomes with similar predictive value as the pH value (32). Since lactate is a product of anaerobic metabolism, an increased level of lactate might represent hypoxic conditions in the tissue. Therefore, the interest in lactate in relation to the cerebral oxygenation in the neonatal period increased in the last years. Eighteen studies were identified, which investigated cerebral oxygenation and blood lactate level in neonates (7,11,13,15,16,18,(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31). However, only one of these publications demonstrated a negative association between cerebral oxygenation and lactate (7), and three studies found no association (11,23,29) between these factors.
Weiss et al. (11) described, in the largest cohort of critically ill neonates, no significant correlation between cerebral oxygenation and lactate. Amigoni et al. (23) also did not find an association between serum lactate and cerebral oxygenation. However, they described a correlation between pH value and cerebral oxygenation. Mebius et al. assessed the course of cerebral regional oxygen saturation and clinical factors in neonates born with duct-dependent congenital heart disease and found no correlation during the first 72 h after birth (29). In extremely preterm infants, it has been demonstrated that the crSO2 and preductal perfusion index were weakly correlated with lactate and blood gas (7).

LIMITATION
The identified publications show many differences in methods: (e.g., study population, number of included neonates, NIRS devices, time point, and frequency of NIRS measurements). Important limitations are also the differences in frequencies of blood samples and differences in time periods between taking blood samples and NIRS measurements, ranging from 5 min (6) to 24 h (29). Several studies even provide no or inaccurate information on frequencies and time points of taking blood samples (11,12,16,19,27). This review identified only observational studies, where associations between cerebral oxygenation and blood glucose and/or lactate levels are described. No interventional study was identified elucidating any causality. Furthermore, there were several studies just describing cerebral oxygenation and blood glucose or lactate in neonates without analyzing any possible associations between these parameters.

CONCLUSION
The influence of blood glucose level and blood lactate level on the cerebral oxygenation in neonates is still controversial. However, there is some evidence that there is an association between cerebral oxygenation and the metabolic parameters, blood glucose, and blood lactate, whereby causal relationship needs further investigation.

AUTHOR CONTRIBUTIONS
CM, GP, and BU: conception and design. CM and GP: literature search and drafting of the article. CM, GS, BU, and GP: analyses and interpretation of data, critical revision, editing, and final approval of the article. All authors contributed to the article and approved the submitted version.