Authors: Meng-Qin Wang, Ya-Ning Zheng, Ying Zhuang
Categories: 6200, Meta-analysis, neonatal hypoglycemia, oral glucose gel, systematic review, Research Article, Systematic Review and Meta-Analysis
Source: Medicine
Neonatal hypoglycemia (NH) is the most prevalent metabolic disorder in neonates and glucose gel in oral solution is a relatively new treatment option for NH. We aimed to determine whether oral glucose gel can prevent NH.
We conducted an open literature search using PubMed, Embase, Cochrane Library, and Web of Science. We used relative risk as the statistical data, expressed each outcome effect as a 95% confidence interval, and conducted a heterogeneity test. If heterogeneity statistics indicated that I2 was ≥ 50%, the random effects model analysis was used; otherwise, the fixed effects model analysis was conducted, and sensitivity analyses were conducted for all outcomes.
In this review, we included a total of 10 studies involving 4801 neonates. Meta-analysis revealed that there were no significant differences between the preventive oral glucose gel group and the control group in terms of blood glucose concentration, glucose concentration 30 minutes after the first breastfeeding, length of stay, Bayley-III composite score, subsequent need for intravenous injection of glucose, 24-hour glucose > 50 mg/dL, separation from mother for treatment of hypoglycemia/admitted to neonatal intensive care unit for hypoglycemia, normoglycemia after 1 to 2 treatments, or normoglycemia after more than 2 treatments, breastfeeding at discharge, delayed feeding, neurosensory impairment, parental satisfaction, developmental delay, and seizure. The subsequent intake was significantly lower in the glucose gel group compared to the control group.
The use of oral glucose gel as a preventative measure may not reduce the incidence of NH. In order to assess the efficacy of glucose gel in preventing NH, a more high-quality, large-sample, and rigorously designed randomized controlled trial is required.
Keywords: Meta-analysis, neonatal hypoglycemia, oral glucose gel, systematic review
Neonatal hypoglycemia (NH) is the most prevalent metabolic disorder in neonates, with a prevalence ranging from 5% to 15%.^[1,2]^ The primary source of energy for the brain is glucose, and NH can cause brain damage and death.^[3–5]^ Infants born with a risk for NH may have an increased risk of developmental retardation as they grow older. Although the impact or duration of hypoglycemia prior to morbidity is unknown, it is well-known that even asymptomatic infants can develop negative outcomes.^[6–8]^ Consequently, it is essential to take necessary steps to reduce the incidence of NH.
Currently, hypoglycemia can be treated with complementary feeding (expressed breast milk or infant formula), but intravenous glucose injection cannot be administered.^[1,9]^ Options for treatment depend on birth weight and gestational age. For late preterm and full-term infants, attention should be paid to feeding and heightened monitoring, and require repeated, painful blood tests.^[10]^ If the blood glucose level of the infant remains low, the infant must be transferred to the neonatal intensive care unit (NICU) for intravenous glucose administration.^[10,11]^ However, both strategies may have a negative impact on breastfeeding initiation.^[12]^ Glucose gel in oral solution is a relatively new treatment option for NH, with the potential advantage of keeping the mother and baby together during treatment; it is also simple to administer and inexpensive.^[10,13]^ Based on studies, glucose gel is effective in controlling NH during postpartum ward management, thereby reducing admissions to the NICU.^[1,10]^ Glucose gel has been widely used in clinical practice for preventative purposes. In a randomized controlled trial (RCT),^[14]^ 20 infants were randomly assigned to the breastfeeding group or the breastfeeding + 400 mg/kg glucose gel group on the first day after birth, however, the results demonstrated that there was no difference in blood glucose concentration between 15 and 30 minutes of treatment.
In this study, a systematic review and meta-analysis were conducted to evaluate the efficacy of glucose gel in preventing NH.
PubMed, Embase, Cochrane Library, and Web of Science were queried from the date of database establishment to December 26, 2020, to retrieve published articles. “Glucose gel” or “glucose” or “d-glucose” or “d glucose” or “glucose” or “glucose (alpha-d)-isomer” or “anhydrous glucose” or “glucose, ‘glucose’ isomer” or “L-glucose” or “glucose” or “glucose” (isomer); “infants” or “newborns” or “neonate”; “hypoglycemia” or “postprandial hypoglycemia” or “reactive hypoglycemia” or “hypoglycemia, reactive” or “fasting hypoglycemia” or “hypoglycemia, fasting” or “hypoglycemia after absorption” or “hypoglycemia, after absorption.”
Inclusion NH; neonates who received oral glucose gel for preventive therapy were included in the experimental group, while those not administered oral glucose gel or who received delayed treatment were included in the control group; RCTs, quasi-experimental studies, or cohort studies; English literature.
Exclusion Animal experiments; materials unrelated to the topic; reviews, meta-analyses, protocols, case reports, and conference abstracts; studies from which it was impossible to extract valid data.
For RCTs and quasi-experimental studies, the literature quality was evaluated using a modified Jadad scale (total 7 points; low 1–3 points; high 4–7 points). The modified Newcastle-Ottawa Scale (total 10 points, low < 5 points, high ≥ 5 points) was used to evaluate the quality of the cohort study literature.^[15]^
The following information was compiled for each author, year, country, design, patient, group, intervention, pregnancy, gender, birth weight, quality, and outcome.
The prevention of NH with oral glucose gel was evaluated based on changes in neonatal blood glucose concentration, the next intake, glucose concentration 30 minutes after the first feeding, length of stay, Bayley-III composite score, the need for intravenous injection of glucose, blood glucose > 50 mg/dL for 24 hours, separation from mother for treatment of hypoglycemia/admission to NICU for hypoglycemia, normoglycemia after 1 or 2 treatment attempts, and normoglycemia after more than 2 treatments, breastfeeding at discharge, delayed feeding, neurosensory impairment, parental satisfaction, developmental delay, and seizure.
For statistical analysis, Stata 15.1 software (Stata Company, College Station, TX, USA) was utilized. Counting data were analyzed using relative risk (RR), each outcome effect was expressed as a 95% CI, and the heterogeneity test was conducted. If I2 was ≥ 50%, the random effects model analysis was conducted based on heterogeneity statistics; otherwise, the fixed effects model analysis was conducted. Sensitivity analyses were conducted for all outcomes. P < .05 indicated a statistically significant difference.
There were no participants in this study. Ethical approval and consent were not required as this study was based on publicly available data.
A total of 3349 articles were initially retrieved using the retrieval strategy. After removing duplicates, 2100 articles remained; 68 articles were initially screened based on titles and abstracts, and 10 were finalized and selected for this study.^[1,10,14,16–22]^ The flowchart is depicted in Figure 1. This study included 4801 neonates, of which 2551 were assigned to the experimental group and 2250 were assigned to the control group. Table 1 displays the baseline characteristics of the included studies.
Figure 1. The flow diagram of the search strategy.
Troughton et al compared the changes in blood glucose concentration before and after treatment in the experimental group and the control group, and found no statistically significant difference between the 2 groups after treatment (P = .2)^.[14]^ The results of their study also indicated that the subsequent feeding was significantly lower in the glucose gel group than in the control group (P = .001).
Coors et al measured glucose concentrations 30 minutes after the initial feeding (control group) or after administering glucose gel (preventive participants)^.[16]^ The bivariate analysis revealed no difference between the prevention group and the control group in terms of the initial glucose concentration (52.1 ± 17.1 vs 50.5 ± 15.3 mg/dL, P = .69).
In a study evaluating the use of glucose gel in the control of NH, it was determined that the length of stay in the postnatal ward were comparable for all neonates receiving treatment for hypoglycemia (P = .55).^[1]^
Mild developmental retardation (Bayley-III cognitive or language score is 1–2 SDs below the mean), moderate (Bayley-III cognitive or language score is 2–3 SDs below the mean), or severe (Bayley-III cognitive or language score is 3–5 SDs below the mean) Harris et al discovered that the mean Bayley-III cognitive or language score among all children was 0.25 to 0.5 SD below the mean, neurosensory impairment was found in 66 children (36%), and impairment rates were comparable (38% in the glucose group and 34% in the placebo group, RR = 1.11, 95% CI: 0.75–1.50, P = .60).^[17]^ In addition, developmental retardation was found in 61 children (33%) (severe: 1, 6, 54), and incidence rates were comparable in both treatment groups (34% in the glucose group and 32% in the placebo group, RR = 1.07, 95% CI: 0.71–1.61, P = .75).
Two studies were included on the need for subsequent intravenous glucose administration.^[10,14]^ All of the heterogeneity test results were statistically significant (I^2^ = 95.9%), so the random effects model was used for analysis. The results of the combined study revealed no statistically significant difference (RR = 0.808, 95% CI: 0.291–2.246, P = .682). After being administered oral glucose gel, there was no difference between the experimental group and the control group regarding the need for intravenous glucose injection (Fig. 2).
Figure 2. Forest plot of subsequent need for intravenous dextrose between the experimental group and control group.
In a study evaluating the effect of glucose gel on neonates with the risk of NH in a pediatric hospital,^[22]^ the proportion of neonates with 24-hour blood glucose > 50 mg/dL before and after the intervention did not differ significantly (67.3% vs 70.4%, P = .27).
Seven studies were included to evaluate separation from mother for hypoglycemia treatment/admission to the NICU for hypoglycemia.^[1,10,16,19–22]^ The results of the heterogeneity test indicated that the difference was statistically significant (I^2^ = 81.9%); thus, the random effects model was utilized for analysis. The combined study results indicated that there was no statistically significant difference (RR = 0.592, 95% CI: 0.334–1.051, P = .073) (Fig. 3).
Figure 3. Forest plot of separation from mother for treatment of hypoglycemia/admitted to NICU for hypoglycemia between the experimental group and control group. NICU = neonatal intensive care unit.
Ter et al found no difference in the proportion of neonates achieving normoglycemia after 1 or 2 treatments in the postnatal ward; the majority of neonates received only 1 dose of glucose gel (75%)^.[1]^ To achieve normoglycemia, the remaining 11 infants in the post-implementation group received more than 2 treatments with glucose gel and supplement feed.
Two studies were included for analysis of breastfeeding at discharge.^[19,22]^ The heterogeneity test yielded no statistically significant results (I^2^ = 0.0%), so the fixed effects model was utilized for analysis. There was no difference between the experimental and control groups in terms of breastfeeding at discharge (RR = 1.018, 95% CI: 0.913–1.134, P = .751) (Fig. 4).
Figure 4. Forest plot of breastfeeding at discharge between the experimental group and control group.
Hegarty et al conducted a randomized, double-blind, placebo-controlled study in 2 New Zealand hospitals to determine the effectiveness of glucose gel in preventing NH^.[19]^ In terms of the incidence of delayed feeding, there was no statistical difference between the experimental group and the control group (P > .05). In addition, there was no difference in parental satisfaction between infants receiving a single dose or multiple doses of glucose gel (RR = 1.06, 95% CI: 1.00–1.13, P = .06). In terms of infant satisfaction, there was no difference between infants receiving glucose gel or placebo gel (RR = 0.95, 95% CI: 0.90–1.01, P = .12).
Two studies evaluating the efficacy of glucose gel involved seizures.^[17,22]^ According to the fixed effects model, there was no difference in the incidence of seizures between the experimental group and the control group [RR = 0.809, 95% CI: (0.224–2.915), P = .746] (Fig. 5).
Figure 5. Forest plot of seizures between the experimental group and control group.
Glucose gel is increasingly used to treat NH.^[2]^ To investigate the efficacy of oral glucose gel in the treatment of NH, a systematic review and meta-analysis were conducted utilizing an integrated search of multiple databases. A total of 10 studies involving 4801 patients were included (2551 in the experimental group, and 2250 in the control group). Results suggested that oral glucose gel may not be effective in preventing NH. In contrast, the subsequent intake in the glucose gel group was significantly lower than that in the control group. Consequently, glucose gel should not be employed to prevent NH. Preventive therapy should be conducted based on the view that the identification and treatment of hypoglycemia after the first feeding may not be sufficient to prevent adverse sequelae. In addition, the preventive use of glucose gel may reduce the need for laboratory tests and other interventions that would separate mothers from neonates and reduce the next feeding.
Previous trials included a multi-center randomized trial regarding the prevention of NH in high-risk infants with oral glucose gel, involving 2149 cases with NH (maternal diabetes, large or small gestational age, late preterm birth), and a trial involving 416 high-risk newborns.^[19,23]^ In the meta-analysis of both trials, early preventive administration of oral glucose gel (within 1 hour of birth) reduced the incidence of hypoglycemia (38% vs 43%, RR 0.87, 95% CI 0.79–0.95).^[24]^ However, the rates of intravenous glucose administration were the same in both groups (37%), as were the percentages of neonates who were separated from their parents or caregivers due to hypoglycemia (5.6% vs 5%).
As demonstrated in a previous study, glucose gel may improve the blood glucose level in neonates with asymptomatic hypoglycemia in comparison to breastfeeding.^[25]^ Similar conclusions were reached in other studies regarding the effect of glucose gel on blood glucose levels. The results of this study found that glucose gel could be used to treat infant hypoglycemia and improve blood glucose levels.^[18]^ However, the sample sizes of these studies were still inadequate, and additional research is required.
The blood glucose levels of some neonates could not be restored through individual feeding; therefore, intravenous glucose injection must be administered; as a result, the corresponding neonates must be transferred from the newborn baby-caring nursery (NBN) to the NICU. The neonates would need to be separated from their mothers, which would have a negative impact on intimacy, cause excessive anxiety about breastfeeding, and raise hospitalization expenses.^[25]^ Few infants receiving glucose gel would need to be separated from their mothers, and they would be more likely to be exclusively breastfed after discharge, according to a review study.^[12]^ The results of a quality improvement project and a retrospective study revealed that glucose gel decreased NICU admissions, increased breastfeeding, and decreased the need for intravenous glucose injections.^[25,26]^ The reduced NICU admission rate was a consistent result of reversing NH with glucose gel (the NICU admission rate could be reduced by 73% in the first year). Also, the rate of exclusive breastfeeding could be increased to 49%.^[21]^ Similar findings demonstrated that the use of 200 mg/kg glucose gel in infants with NH did not inhibit subsequent feeding, but rather enhanced the quality of breastfeeding.^[27]^
Given that the primary objective of NH treatment is to prevent cerebral injury, it is crucial to determine whether glucose gel can have positive or negative effects on later development.^[17]^ In a region birth cohort, transient NH was associated with school performance at age 10, and there was a graded relationship between the severity of hypoglycemia and the rate of literacy and numeracy.^[28]^ However, Harris et al failed to find evidence that glucose gel improved brain injury in infants.^[17]^ To solve this issue, randomized treatment comparison trials need to be conducted.
In this systematic review and meta-analysis, we collected data from various comprehensive databases and included additional studies evaluating the efficacy of oral glucose gel. The main limitation of this study is the small sample size. Large, well-designed trials need to be conducted in future.
Based on the integrated retrieval from multiple databases, a systematic review and meta-analysis were conducted to examine the efficacy of oral glucose gel in the prevention of NH. The outcomes suggest that preventative oral glucose gel may not be effective in reducing the incidence of NH. However, due to the benefits of simple administration and low cost, physicians should place greater emphasis on the clinical efficacy of oral glucose gel. This study may offer some insights and recommendations for the prevention and reduction of NH.
Conceptualization: Meng-Qin Wang.
**Data ** Meng-Qin Wang, Ya-Ning Zheng, Ying Zhuang.
**Formal ** Meng-Qin Wang, Ya-Ning Zheng.
**Funding ** Ying Zhuang.
Investigation: Ya-Ning Zheng.
**Project ** Meng-Qin Wang.
Resources: Ya-Ning Zheng.
Software: Meng-Qin Wang.
Validation: Ying Zhuang.
Visualization: Ying Zhuang.
**Writing – original ** Meng-Qin Wang, Ying Zhuang.
**Writing – review & ** Meng-Qin Wang, Ya-Ning Zheng, Ying Zhuang.
Meng-Qin Wang, Email: wang82266@outlook.com.
Ya-Ning Zheng, Email: zhengyaningzyn8@126.com.