Abstract
Objective:
To determine the effects of different doses of prophylactic dextrose gel on glycemic stability assessed using continuous interstitial glucose monitoring (CGM) in the first 48 hours when given to babies at risk of neonatal hypoglycemia.
Study Design
CGM was undertaken for the first 48 hours in 133 infants at risk of hypoglycemia who participated in the pre-hPOD randomized dosage trial of dextrose gel prophylaxis.
Results
Low glucose concentrations were detected in 41% of infants by blood glucose monitoring and 68% by CGM. The mean (SD) duration of low interstitial glucose concentrations was 295 (351) minutes in the first 48 hours. Infants who received any dose of dextrose gel appeared less likely than those who received placebo gel to experience low glucose concentrations (<47mg/dl [2.6mmol/l], p=0.08), particularly if they received a single dose of 200mg/kg (RR 0.70, 95% CI 0.50-0.10, p = 0.049). They also spent a larger proportion of time in the central glucose concentration range of 54 to 72 mg/dl [3 to 4 mmol/l] (any dose: mean (SD) 58.2 (20.3)%; placebo: 50.0 (21.9)%, mean difference 8.20%; 95% CI 0.43-15.9%, p = 0.038). Dextrose gel did not increase recurrent or severe episodes of low glucose concentrations and did not increase the peak glucose concentration. These effects were similar for all trial dosages.
Conclusions
Low glucose concentrations were common in infants at risk of hypoglycemia despite blood glucose monitoring and treatment. Prophylactic dextrose gel reduced the risk of hypoglycemia without adverse effects on glucose stability.
Keywords: Hypoglycemia, blood glucose, infant newborn, hyperglycemia, neonatal screening
Neonatal hypoglycemia is common in newborn infants and can cause brain injury. (1) To minimize this, it is recommended that infants known to be at particular risk (usually those born preterm, small, large, or to mothers with diabetes) receive intermittent blood glucose screening for several hours after birth and that those with low glucose concentrations are treated promptly.(2) However, there is some evidence that, even when such screening and treatment protocols are in place, neonatal hypoglycemia may be associated with later impaired executive and visual-motor function (3) and poor school performance.(4) These findings suggest that prophylactic measures should be considered.
Buccal administration of dextrose gel is an effective first-line treatment for hypoglycemia (5) and appears safe and acceptable to families and caregivers.(5,6) In two randomized trials, we have shown that buccal dextrose gel reduces the incidence of hypoglycemia when administered prophylactically to at-risk infants at one hour of age.(7,8) Although no short-term adverse effects have been reported, there are several potential concerns about the use of prophylactic dextrose gel. In particular, in a cohort of infants born at risk of hypoglycemia and treated using a standard protocol which including feeding, treatment dextrose gel, and intravenous dextrose, we previously showed that high or unstable glucose concentrations and a rapid rise after treatment of hypoglycemia with dextrose (especially intravenous dextrose) were associated with later neurodevelopmental impairment.(9,10) Thus, it is important to determine whether prophylactic dextrose gel may cause high or unstable glucose concentrations, particularly as a prophylaxis strategy necessarily means that at-risk infants, including those who would not have developed hypoglycemia, are exposed to the intervention.
Continuous glucose monitoring (CGM) devices, which measure the glucose concentration in the interstitial fluid and report this every 5 minutes, correlate well with intermittent blood glucose concentration measurements in pediatric intensive care,(11) and have been reported to be accurate, reliable, and safe in even the smallest infants.(12-15) We therefore used CGM to determine the effect of prophylactic dextrose gel on neonatal glycemia in the first 48 hours in the subset of participants who consented to CGM monitoring in our previous randomized controlled trial comparing different doses of oral dextrose gel with placebo for prevention of hypoglycemia in infants at risk.(7)
Methods
Participants were infants whose parents consented to continuous glucose monitoring as part of the pre-hPOD double-blinded randomized controlled trial of different doses of prophylactic dextrose gel.(7) Written informed consent was obtained before the birth and confirmed verbally after the birth. The trial was approved by the Northern A Health and Disability Ethics Committee of New Zealand and prospectively registered with the Australian New Zealand Clinical Trials Registry, number ACTRN12613000322730.
In brief, eligible infants were born at Auckland City Hospital, Auckland, New Zealand between 8/2013-11/2014 and were at risk of neonatal hypoglycemia [infant of diabetic, late preterm (35 or 36 weeks of gestation), small (birth weight <10th percentile on population or customized birthweight charts or <2.5 kg) or large (birth weight >90th centile on population or customized birthweight charts or >4.5kg) or other risk factors (e.g. maternal medication)] but were otherwise unlikely to require admission to Neonatal Intensive Care Unit (NICU) (≥35 weeks of gestation, birth weight ≥2.2 kg, no apparent indication for NICU admission) and mother intended to breastfeed. Infants were randomized within the first hour after birth to one of 8 treatment arms: either 40% dextrose gel 0.5 ml/kg once (200 mg/kg), 1 ml/kg once (400 mg/kg), 0.5 ml/kg for 4 doses (800 mg/kg), or 1 ml/kg once followed by 0.5ml/kg for a further 3 doses (1,000 mg/kg), or equivolume 2% hydroxymethylcellulose gel placebo. Study gel was massaged into the buccal mucosa, either once at 1 h of age or an additional 3 times before feeds in the first 12 h, with gel given no more frequently than 3 hourly. Each dose of gel was followed by breastfeeding.
Infants whose parent(s) gave consent to continuous glucose monitoring had a glucose oxidase continuous glucose monitoring sensor (Enlite™, Medtronic MiniMed, Northridge, CA, USA) inserted subcutaneously into the lateral aspect of the thigh using a spring-loaded insertion device (Serter™, Medtronic MiniMed, Northridge, CA, USA) as soon as possible after birth, attached to a continuous glucose monitor (iPRO™2, Medtronic MiniMed, Northridge, CA, USA). Interstitial glucose concentrations could not be viewed in real-time and therefore were not available to parents, clinicians, or research staff and had no influence on clinical management. The sensors remained in situ for at least 48 hours and were calibrated against blood glucose concentrations at least twice daily. Data were downloaded at the end of the study period using the continuous glucose monitoring software package (CareLink iPRO2 software, Medtronic MiniMed, Northridge, CA, USA) and recalibrated using a point-to-point calibration algorithm as previously described.(16) Only data from the first 48 hours after birth were included in the analyses.
Blood glucose concentrations were measured at 2 hours after birth, and then according to hospital protocol, with pre-feed blood glucose measurements every 2-4 hourly for at least the first 12 hours, and until there were 3 consecutive blood glucose concentrations of ≥47 mg/dl [2.6 mmol/l]. All blood glucose concentrations were analyzed by the glucose oxidase method, either with a portable blood glucose analyzer (iSTAT, Abbott Laboratories, Abbott Park, IL USA) or a combined metabolite/blood gas analyzer (e.g. ABL 700, Radiometer Ltd, Copenhagen, Denmark).
Infants who developed hypoglycemia (blood glucose <47 mg/dl [2.6 mmol/l]) were managed by the clinical team according to standard clinical practice, including treatment with 40% dextrose gel, supplementation with formula, and admission to NICU for intravenous dextrose if required.
The primary outcome for this study was the number of episodes of low interstitial glucose concentrations, defined as any interstitial glucose concentration (single data point recorded every 5 minutes) <47 mg/dl in the first 48 hours after birth. Consecutive interstitial glucose concentrations of <47 mg/dl were defined as a single episode. Hypoglycemia was defined as any single blood glucose measurement of <47 mg/dl. Severe episodes were <36 mg/dl [2.0 mmol/l] and prolonged episodes were >30 min. Additional analysis was planned for the duration of interstitial glucose recorded between the central range of 54 to 72 mg/dl [3 to 4 mmol/l] (9,10) and for the incidence of hyperglycemia defined as interstitial or blood glucose concentrations > 144 mg/dl [8 mM].
Data were analyzed using SAS (v 9.4, SAS Institute Inc, Cary, NC) and are described as mean (standard deviation) or median (range) and measures of effect with 95% confidence intervals. A two-tailed probability of <0.05 was considered statistically significant. No adjustment for multiple comparisons or imputation of missing data was performed. All analyses were performed on the intention to treat principle. Generalized linear models were used to calculate mean difference between groups (normal distribution and identity link function) for continuous variables, relative risk (binomial distribution and log link function) for categorical comparisons, and rate ratios (Poisson distribution log link function) for comparison of counts of numbers of episodes. Median difference (95% confidence intervals) were calculated based on the Hodges-Lehmann estimator for continuous outcomes that could not be rendered normal by transformation.
The interstitial glucose concentration response to study gel was estimated for four hours after any administration of 0.5ml/kg study gel or, for those infants randomized to multiple doses, up until the time of the next dose. Between-group comparisons were made using appropriate general linear models with a clustering term for each study participant. Group allocation and dose order main effects and their interactions were fitted to data from each epoch. The rate of rise in glucose concentration after a dose of study gel was estimated as the maximum increase in glucose concentration over the time to reach the maximum glucose concentration. The rate of change in glucose concentration was estimated from the slope of the ordinary least squares regression line for absolute glucose concentrations. Areas around the continuous blood glucose trace were calculated using the trapezoidal method. Firstly, the area under the curve (AUC) of interstitial glucose concentration over time was calculated for each infant. Secondly, the baseline value of interstitial glucose concentration at the time of gel administration was extrapolated and the area above the extrapolated baseline to the observed interstitial glucose concentrate was calculated as positive AUC. Thirdly, the area below the extrapolated baseline to the observed interstitial glucose over time was calculated as negative AUC.
Results
Of the 415 infants recruited to the pre-hPOD trial, consent was obtained for continuous glucose monitoring in 138. Monitors were inserted in all infants, but for five, no continuous glucose data were recorded, leaving 133 included in this analysis. Of these, 123 (92%) had at least 48 h of satisfactory recordings (Figure 1 online). Demographic and baseline characteristics were similar in this group and in all those who took part in the pre-hPOD trial, and similar in all randomization groups (Table 1 online). The median (range) birth weight was 3190 (2200, 5255) g and gestational age was 38 (35, 42) weeks; 87/133 infants (65%) were infants of diabetics, and 64/133 (48%) were born by cesarean delivery. Overall, 10/133 (7.5%) of infants were admitted to NICU; 6 for hypoglycemia, 3 for respiratory distress, and 1 for seizures not related to hypoglycemia. Formula was given to 14/133 (11%) infants during admission, with the median (range) number of formula feeds being 7 (1, 13) for those infants given formula. At discharge from the hospital, 89/133 (67%) of infants were fully or exclusively breastfeeding.
Figure 1.
CONSORT Diagram
Table 1:
Baseline characteristics of mothers and infants randomized to different doses of prophylactic dextrose gel
| Single Dose | Multiple Dose | For CGM study |
For overall pre-hPOD trial |
|||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Placebo | Dextrose 0.5ml/kg |
Dextrose 1ml/kg |
Placebo | Dextrose 0.5ml/kg x 4 |
Dextrose 1ml/kg x 1 0.5ml/kg x 3 |
Any dose of placebo gel |
Any dose of dextrose gel |
Total | Total | |
| Mothers (n = 129) | 18 | 23 | 19 | 23 | 18 | 28 | 41 | 88 | 129 | 403 |
| Number | ||||||||||
| Maternal age (years) | 36 (20, 45) | 35 (22, 41) | 32 (21, 51) | 31 (18, 44) | 31 (21, 42) | 34 (19, 41) | 32 (18, 45) | 32 (19, 51) | 32 (18, 51) | 32 (18, 51) |
| Parity | 2 (1, 5) | 2 (1, 5) | 1 (1, 5) | 2 (1, 3) | 2 (1, 10) | 2 (1, 8) | 2 (1, 5) | 2 (1, 10) | 2 (1,10) | 2 (0,10) |
| Weight at booking (kg) median (range) | 73 (45, 146) | 83 (49, 140) | 72 (48, 200) | 63 (42, 148) | 69 (53, 156) | 71 (46, 164) | 70 (42, 148) | 74 (46, 200) | 72 (42, 200) | 71 (42, 200) |
| Diabetic | ||||||||||
| Type 1 Diabetes | 1 (6) | 1 (4) | 0 | 1 (4) | 1 (6) | 2 (7) | 2 (5) | 54 (5) | 6 (5) | 18 (4) |
| Type 2 Diabetes | 2 (11) | 1 (4) | 5 (26) | 0 | 2 (11) | 2 (7) | 2 (5) | 10 (11) | 12 (9) | 24 (6) |
| Gestational Diabetes | 10 (56) | 12 (52) | 7 (37) | 13 (57) | 11 (61) | 15 (54) | 23 (56) | 45 (51) | 68 (53) | 256 (64) |
| Insulin therapy | 9 (50) | 11 (48) | 11 (58) | 9 (39) | 11 (61) | 11 (39) | 18 (44) | 44 (50) | 62 (48) | 212 (53) |
| Pre-eclampsia | 3 (17) | 3 (13) | 0 | 0 | 0 | 2 (7) | 3 (7) | 5 (6) | 8 (6) | 15 (4) |
| Hypertension | 1 (6) | 5 (22) | 1 (5) | 2 (9) | 1 (6) | 6 (21) | 3 (7) | 13 (15) | 16 (12) | 41 (10) |
| Prioritized Ethnicity | ||||||||||
| Māori | 1 (6) | 3 (13) | 3 (16) | 4 (17) | 0 | 3 (11) | 5 (12) | 9 (10) | 14 (11) | 40 (10) |
| Pacific | 3 (17) | 7 (30) | 5 (26) | 3 (13) | 5 (28) | 6 (21) | 6 (15) | 23 (26) | 29 (22) | 64 (16) |
| Chinese | 1 (6) | 0 | 3 (16) | 3 (13) | 2 (11) | 2 (7) | 4 (10) | 7 (8) | 11 (9) | 44 (11) |
| Indian | 3 (17) | 1 (4) | 1 (5) | 3 (13) | 1 (6) | 1 (4) | 6 (15) | 4 (5) | 10 (8) | 62 (15) |
| Other | 5 (28) | 7 (30) | 2 (11) | 7 (30) | 4 (22) | 5 (18) | 12 (29) | 18 (20) | 30 (23)# | 83 (21)## |
| New Zealand European | 5 (28) | 5 (22) | 5 (26) | 3 (13) | 6 (33) | 11 (39) | 8 (20) | 27 (31) | 35 (27) | 110 (27) |
| Infants (n=133) | 18 | 24 | 22 | 23 | 18 | 28 | 41 | 92 | 133 | 415 |
| Female | 8 (44) | 13 (54) | 10 (45) | 13 (57) | 10 (56) | 11 (39) | 21 (51) | 44 (48) | 65 (49) | 202 (49) |
| Birth weight (g) median (range) | 3220 (2490, 4730) | 2933 (2290, 4580) | 2968 (2375, 4430) | 3390 (2290, 4410) | 3305 (2320, 4290) | 3100 (2380, 5255) | 3285 (2290, 4730) | 3140 (2290, 5255) | 3140 (2290, 5255) | 3190 (2200, 5255) |
| Birth weight Z score | 0.30 (1.50) | 0.01 (1.44) | 0.17 (1.29) | 0.40 (1.46) | 0.40 (1.33) | −0.00 (1.19) | 0.36 (1.46) | 0.12 (1.30) | 0.19 (1.35) | 0.14 (1.29) |
| Gestation (weeks) | 38 (37, 40) | 38 (35, 40) | 38 (35, 39) | 38 (35, 40) | 38 (35, 40) | 38 (36, 41) | 38 (35, 40) | 38 (35, 41) | 38 (35, 41) | 38 (35, 42) |
| Singleton birth | 17 (94) | 21 (88) | 16 (73) | 22 (96) | 17 (94) | 29 (93) | 39 (95) | 80 (87) | 119 (89) | 383 (93) |
| Cesarean birth | 10 (56) | 11 (46) | 9 (41) | 14 (61) | 6 (33) | 14 (50) | 24 (59) | 40 (43) | 64 (48) | 199 (48) |
| Apgar score of <5 at 5 minutes | 2 (11) | 0 | 0 | 0 | 0 | 0 | 2 (5) | 0 | 2 (1.5) | 2 (0.5) |
| Primary risk factor for hypoglycemia | ||||||||||
| Infant of diabetic mother | 13 (72) | 14 (58) | 13 (59) | 14 (61) | 14 (78) | 19 (68) | 27 (66) | 60 (65) | 87 (65) | 301 (73) |
| Late preterm | 0 | 4 (17) | 2 (9) | 1 (4) | 1 (6) | 3 (11) | 1 (2) | 10 (11) | 11 (8) | 27 (7) |
| Small | 2 (11) | 5 (21) | 5 (23) | 3 (13) | 2 (11) | 4 (14) | 5 (12) | 16 (17) | 21 (16) | 49 (12) |
| Large | 3 (17) | 1 (4) | 2 (9) | 5 (22) | 1 (6) | 2 (7) | 8 (20) | 6 (7) | 14 (11) | 38 (9) |
| Blood glucose monitoring | ||||||||||
| Age at first sample (h) | 2.13 (0.20) | 2.22 (0.54) | 2.08 (0.14) | 2.31 (0.73) | 2.07 (0.28) | 2.16 (0.14) | 2.23 (0.56) | 2.14 (0.32) | 2.17 (0.41) | |
| Duration of monitoring (h) | 36.25 (10.71) | 40.09 (6.61) | 34.91 (13.13) | 39.39 (9.83) | 38.70 (13.61) | 35.44 (11.86) | 38.35 (10.27) | 37.17 (11.49) | 37.53 (11.10) | |
| Number of blood samples* median (range) | 7.5 (5, 14) | 6.5 (4, 11) | 6 (4, 11) | 7 (4, 15) | 6 (4, 14) | 6 (4, 9) | 7 (4, 15) | 6 (4, 14) | 7 (4, 15) | |
| Interstitial glucose monitoring | ||||||||||
| Age recording commenced (h) | 0.62 (0.29) | 0.70 (0.29) | 0.73 (0.33) | 0.59 (0.32) | 0.87 (0.24) | 0.78 (0.28) | 0.60 (0.30) | 0.76 (0.29) | 0.71 (0.30) | |
| Duration of monitoring (h) | 51.35 (8.96) | 52.05 (5.18) | 49.69 (9.77) | 52.56 (11.93) | 50.50 (10.60) | 50.00 (8.38) | 52.03 (10.62) | 50.56 (8.46) | 51.01 (9.16) | |
| Number of glucose readings* median (range) | 551 (269, 559) | 551 (523, 558) | 551.5 (28, 554) | 550 (85, 564) | 550.5 (255, 564) | 551 (142, 553) | 550 (85, 559) | 551 (28, 564) | 551 (28, 564) | |
| Volume of formula received* (ml) | 174.8 (73.4) | 96.0 (69.7) | 258.0 (120.0) | 83.1 (98.7) | 159.6 (118.9) | 174.6 (126.1) | 123.4 (98.4) | 179.9 (124.4) | 159.7 (118.3) | |
| Volume of open label 40% dextrose gel received* (ml) | 1.77 (0.60) | 3.52 (2.15) | 3.74 (2.55) | 3.18 (1.24) | 2.68 (1.09) | 1.74 (0.70) | 2.41 (1.16) | 2.77 (1.79) | 2.65 (1.60) | |
Data are mean (SD), median (range) or number (%)
First 48 h only
Afghani (1), Asian (8), Ethiopian (1), European (2), Fijian (3), Fijian Indian (1), Filipino (5), Middle Eastern (2), North American (3), South African (1), Latin American (3)
Afghani (1), Asian (20), Ethiopian (1), European (14), Fijian (4), Fijian Indian (3), Filipino (17), Middle Eastern (4), Nauruan (2), North American (5), South African (5), Latin American (6), Not further specified (1)
There were no complications related to CGM use. Specifically, there was no edema, bruising, or infection at any insertion site. All CGM remained in place for the monitoring period until removed by staff, except for one infant with a thick vernix layer, for whom the adherent dressing failed to secure the CGM sensor and monitor. No infant had clinical signs of hypoglycemia.
The proportion of infants who experienced at least one low glucose concentration (<47 mg/dl [2.6mmol/l]) detected by blood glucose monitoring was 41% and detected by interstitial glucose monitoring was 68% (p < 0.0001), with 36/133 infants (27%) experiencing a low interstitial glucose concentration that was not detected by blood glucose monitoring (Table 2). The majority of low glucose concentrations occurred in the first 12 hours (51/55, 93% of blood and 74/91, 81% of interstitial), and of these 9.6% occurred before 2.5 hours of age, but more infants experienced low interstitial concentrations than blood glucose concentrations during all time periods. Interstitial glucose monitoring detected more infants with ≥3 episodes (29% vs 5%, p <0.0001) but similar numbers with 1 or 2 episodes, and more infants with long episodes (>30 min, 57% vs 16%, p <0.0001) but not more with severe episodes (<36 mg/dl [2.0mmol/l], Table 2). The mean (SD) duration of low interstitial glucose concentrations was 295 (351) min, and infants spent 55.6 (21.1)% of the time with a glucose concentration of 54 to 72 mg/dl [3 to 4 mmol/l] in the first 48 h after birth. There were no infants with high glucose concentrations (144 mg/dl [>8 mmol/l]) detected using either blood or interstitial glucose monitoring.
Table 2:
Comparison of intermittent blood with continuous interstitial glucose monitoring
| Blood glucose N=133 |
Interstitial glucose N=133 |
P | |
|---|---|---|---|
| Age at which low glucose concentration first detected (h) | 5.0 (1.9, 46.2) | 4.5 (1.9, 47.6) | <0.0001 |
| Infants with any episode of low glucose concentrations | |||
| 0 - 12h | 51 (38.3) | 74 (55.6) | <0.0001 |
| 12 - 24h | 6 (4.5) | 26 (19.5) | <0.0001 |
| 24- 36h | 3 (2.3) | 27 (20.3) | <0.0001 |
| 36- 48h | 8 (6.0) | 37 (27.8) | <0.0001 |
| 0 - 48h | 55 (41.4) | 91 (68.4) | <0.0001 |
| Infants with: | |||
| 0 episodes | 78 (58.6) | 42 (31.6) | <0.0001 |
| 1 episode | 39 (29.3) | 38 (28.6) | 0.87 |
| 2 episodes | 10 (7.5) | 15 (11.3) | 0.28 |
| ≥ 3 episodes | 6 (4.5) | 38 (28.6) | <0.0001 |
| Episodes of low glucose concentrations per infant* | |||
| 0 - 12h | 1 (1, 2) | 1 (2, 3) | 0.095 |
| 12 - 24h | 1 (1,2) | 1 (1, 3) | 0.17 |
| 24- 36h | 1 (1, 1) | 1 (1, 4) | 0.52 |
| 36- 48h | 2 (1, 2) | 1 (1, 4) | 0.58 |
| 0 - 48h | 1 (1, 4) | 2 (1, 11) | 0.004 |
| Infants with long episodes of low glucose concentrations | 21 (16) | 76 (57) | <0.0001 |
| Long episodes per infant | 0 (0, 3) | 1 (0, 8) | <0.0001 |
| Infants with episodes of severe low glucose concentrations: | |||
| 0 - 12h | 10 (7.5) | 8 (6.0) | 0.48 |
| 12 - 24h | 1 (0. 8) | 2 (0.5) | 0.57 |
| 24- 36h | 0 | 2 (1.5) | DNC |
| 36- 48h | 1 (0. 8) | 3 (2.2) | 0.18 |
| 0 - 48h | 12 (9.0) | 12 (9.0) | 0.99 |
| Infants with: | |||
| 1 severe episode | 10 (7.5) | 10 (7.5) | 0.99 |
| 2 severe episodes | 2 (1.5) | 1 (0.8) | 0.57 |
| ≥ 3 severe episodes | 0 | 1 (0.8) | DNC |
| Severe episodes of low glucose concentrations per infant: | |||
| 0 - 12h | 0 (0, 2) | 0 (0, 1) | 0.26 |
| 12 - 24h | 0 (0, 1) | 0 (0, 1) | 0.57 |
| 24- 36h | 0 (0, 0) | 0 (0, 2) | DNC |
| 36- 48h | 0 (0, 1) | 0 (0, 1) | 0.18 |
| 0 - 48h | 0 (0, 2) | 0 (0, 4) | 0.73 |
| Infants with long episodes of severe low glucose concentrations | 6 (4.5) | 7 (5.2) | 0.71 |
| Long episodes of severe low glucose concentrations per infant | 0 (0, 1) | 0 (0, 4) | 0.34 |
| Total duration of low glucose concentration per infant(min) | 116 (283) | 295 (351) | 0.0002 |
| Proportion of time with glucose concentration in defined range (% of total time measured per infant) | |||
| < 36 mg/dl [2 mol/l] | 0.3 (1.5) | 1.3 (6.0) | 0.06 |
| < 47 mg/dl [2.6mmol/l] | 6.9 (16.0) | 10.0 (13.5) | 0.09 |
| 54-72 mg/dl [3-4 mmol/l] | 62.1 (25.4) | 55.6 (21.1) | 0.024 |
| > 72 mg/dl [4 mmol/l] | 11.8 (19.6) | 14.3 (18.0) | 0.28 |
Data are median (range), mean (SD) or number (%). DNC = model did not converge
For infants with at least one episode
Low glucose = < 47 mg/dl [2.6 mmol/l], severe low glucose = < 36 mg/dl [2.0 mmol/l], long low glucose = > 30 min.
Compared with infants allocated to any dose of placebo gel, those allocated to any dose of dextrose gel appeared less likely to experience low glucose concentrations (RR 0.82, 95%CI 0.66-1.03, p = 0.08), especially in the first 12 h after birth, but none of these differences reached statistical significance (Table 3). The age at which low glucose concentrations were first detected was not affected by dextrose gel. Infants allocated to any dose of dextrose gel spent an additional 8% of the time with a glucose concentration of 3-4 mmol/l in the first 48 h.
Table 3:
Measures of glycemia in infants randomized to different doses of prophylactic dextrose gel
| Single Dose | Multiple Dose | Total | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Placebo | Dextrose 200mg/kg |
RR, or mean/median difference, 95% CI, P |
Dextrose 400mg/kg |
RR, or mean/median difference, 95% CI, P |
Placebo | Dextrose 800mg/kg |
RR, or mean/median difference, 95% CI, P |
Dextrose 1000mg/kg |
RR, or mean/median difference, 95% CI, P |
Any dose of placebo gel |
Any dose of dextrose gel |
RR, or mean/median difference, 95% CI, P |
||
| Infants | 18 | 24 | - | 22 | - | 23 | 18 | - | 28 | - | 41 | 92 | - | 133 |
| Age at which low glucose concentration first detected (h) | 3.8 (2.1, 31) | 4.6 (1.9, 38.3) |
0.41
*
(−1.40, 9.33) P=0.742 |
3.2 (2.0, 44.3) | −0.49* (−3.02, 1.26) P=0.29 | 7.9 (1.9, 47.6) | 4.8 (2.4, 44.9) | −1.72* (−5.45, 1.93) P=0.59 |
3.93 (2.1, 45.3) | −1.76* (−5.45, 1.93) | 5.4 (1.9, 47.6) | 4.1 (1.9, 45.3) | −0.18* (−2.2, 1.03) P=0.63 |
4.5 (1.9, 47.6) |
| Infants with any episode of low glucose concentrations: | ||||||||||||||
| 0 - 12h | 13 (72) | 11 (45) | 0.63 (0.38, 1.07) P=0.087 |
13 (59) | 0.82 (0.52, 1.28) P=0.38 |
14 (61) | 8 (44) | 0.73 (0.40, 1.35) P=0.73 |
15 (54) | 0.88 (0.55, 1.42) P=0.60 |
27 (66) | 47 (51) | 0.78 (0.58, 1.04) P=0.095 |
74 (56%) |
| 12 - 24h | 4 (22) | 3 (13) | 0.56 (0.14, 2.21) P=0.41 |
6 (27) | 1.23 (0.41, 3.69) P=0.72 |
6 (26) | 2 (11) | 0.43 (0.10, 1.86) P=0.26 |
5 (18) | 0.68 (0.24, 1.96) P=0.48 |
10 (24) | 16 (17) | 0.71 (0.35, 1.43) P=0.34 |
26 (20) |
| 24- 36h | 5 (28) | 5 (21) | 0.75 (0.26, 2.21) P=0.60 |
2 (9) | 0.34 (0.08, 1.56) P=0.17 |
6 (26) | 2 (11) | 0.42 (0.10, 1.97) P=0.29 |
7 (25) | 0.96 (0.37, 2.45) P=0.93 |
11 (27) | 16 (17) | 0.66 (0.34, 0.130) P=0.23 |
27 (20) |
| 36- 48h | 5 (28) | 9 (38) | 1.35 (0.55, 3.34) P=0.52 |
4 (18) | 0.69 (0.22, 2.17) P=0.52 |
6 (26) | 4 (22) | 0.90 (0.30, 2.71) P=0.85 |
9 (32) | 1.23 (0.51, 2.95) P=0.64 |
11 (27) | 26 (55) | 1.08 (0.59, 1.96) P=0.81 |
37 (28) |
| 0 - 48h | 16 (89) | 15 (63) |
0.70 (0.50, 0.998)
P=0.049 |
14 (64) | 0.72 (0.50. 1.02) P=0.066 |
16 (70) | 10 (56) | 0.80 (0.49, 1.31) P=0.37 |
20 (71) | 1.03 (0.72, 1.47) P=0.88 |
32 (78) | 59 (64) | 0.82 (0.66, 1.03) P=0.084 |
91 (68) |
| Infants with: | ||||||||||||||
| 1 episode | 9 (50) | 5 (21%) | 0.42 (0.17, 1.03) P=0.058 |
5 (23) | 0.45 (0.19, 1.12) P=0.085 |
5 (23) | 4 (22) | 1.02 (0.32, 3.26) P=0.97 |
10 (36%) | 1.64 (0.65, 4.13) P=0.29 |
14 (37) | 24 (26) | 0.76 (0.44, 1.32) P=0.33 |
38 (29) |
| 2 episodes | 1 (5.6) | 2 (8.3%) | 1.50 (0.15, 15.28) P=0.73 |
3 (14) | 2.45 (00.28, 21.6) P=0.42 |
4 (17) | 4 (22) | 1.28 (0.37, 4.42) P=0.70 |
1 (3.6) | 0.21 (0.025, 1.71) P=0.14 |
5 (12) | 10 (11) | 0.89 (0.33, 2.44) P=0.82 |
15 (11) |
| ≥ 3 episodes | 6 (33) | 8 (33) | 1 (0.42, 2.37) P=0.99 |
6 (27) | 0.82 (0.32, 2.10) P=0.68 |
7 (30) | 2 (11) | 0.37 (0.09, 1.55) P=0.17 |
9 (32) | 1.06 (0.47, 2.40) P=0.90 |
13 (32) | 25 (27) | 0.86 (0.49, 1.50) P=0.59 |
38 (29) |
| Episodes of low glucose concentrations per infant: | ||||||||||||||
| 0 - 12h | 1 (0, 3) | 1 (0, 3) | −0.5 (−1, 1) P=0.24 |
1 (0, 2) | −0.5 (−1, 0) P=0.59 |
1 (0, 3) | 0 (0, 2) | −0.5 (−1,0) P=0.17 |
1 (0, 2) | −0.5 (−1, 0) P=0.38 |
1 (0, 3) | 1 (0, 3) | −0.5 (−1,0) P=0.11 |
1 (0,3) |
| 12 - 24h | 0 (0, 2) | 0 (0, 1) | 0 (0, 0) P=0.39 |
0 (0, 3) | 0 (0, 0) P=0.69 |
0 (0, 3) | 0 (0, 1) | 0 (0, 0) P=0.20 |
0 (0, 3) | 0 (0, 0) P=0.48 |
0 (0, 3) | 0 (0, 3) | 0 (0,0) P=0.32 |
0 (0,3) |
| 24- 36h | 0 (0, 3) | 0 (0, 2) | 0 (0, 0) P=0.59 |
0 (0, 1) | 0 (0, 0) P=0.12 |
0 (0, 2) | 0 (0, 2) | 0 (0, 0) P=0.30 |
0 (0, 4) | 0 (0, 0) P=0.96 |
0 (0, 3) | 0 (0, 4) | 0 (0,0) P=0.25 |
0 (0,4) |
| 36- 48h | 0 (0, 4) | 0 (0, 3) | 0 (0, 0) P=0.77 |
0 (0, 1) | 0 (0, 0) P=0.32 |
0 (0, 2) | 0 (0, 3) | 0 (0, 0) P=0.99 |
0 (0, 4) | 0 (0, 0) P=0.48 |
0 (0, 4) | 0 (0, 4) | 0 (0,0) P=0.87 |
0 (0,4) |
| 0 - 48h | 1 (0, 10) | 1 (0, 6) | −0.50 (−1, 0) P=0.33 |
1 (0, 5) | −0.5 (−1, 00) P=023 |
1 (0, 7) | 1 (0, 6) | −1 (−2, 0) P=0.23 |
1 (0, 11) | 0 (−1, 0) P=0.79 |
1 (0, 10) | 1 (0, 11) | −0.5 (−1, 0) P=0.26 |
1 (0,11) |
| Infants with long episodes of low glucose concentrations | 14 (78) | 13 (54) | 0.70 (0.45, 1.09) P=0.13 |
12 (55) | 0.70 (0.45, 1.11) P=0.14 |
13 (57) | 8 (44) | 0.79 (0.42, 1.47) P=0.47 |
16 (57) | 1.01 (0.63, 1.64) P=0.97 |
27 (66) | 49 (53) | 0.97 (0.70, 1.33) P=0.84 |
76 (57) |
| Long episodes of low glucose concentrations per infant | 1 (0, 5) | 1 (0, 4) | 1 (0,4) | 1 (0, 3) | 0 (0, 4) | 1 (0,8) | 1 (0, 5) | 1 (0, 8) | 1 (0, 8) | |||||
| Age at which severe low glucose concentrations first detected (h) | 3.1 | - | 3.7 (2.0, 15.7) | 11.6 (2.0, 21.2) | 1.7 (1.3, 2.1) | 22.5 (2.1, 41.4) | 3.1 (2.0, 21.2) | 3.5 (1.3, 41.4) | 3.3 (1.3, 41.4) | |||||
| Infants with episodes of severe low glucose concentrations: | ||||||||||||||
| 0 - 12h | 1 (5.6) | 0 | - | 3 (14) | 2.5 (0.3, 21.6) P=0.46 |
1 (4.3) | 2 (11) | 2.6 (0.3, 26) P=0.48 |
1 (3.6) | 0.82 (0.05, 12.4) P=0.90 |
2 (4.9) | 6 (6.5) | 1.34 (0.28, 6.35) P=0.76 |
8 (6.0) |
| 12 - 24h | 0 | 0 | - | 0 | - | 1 (4.3) | 0 | - | 1 (3.6) | - | 1 (2.4) | 1 (1.1) | 0.46 (0.03, 6.95) P=0.62 |
2 (1.5) |
| 24- 36h | 0 | 0 | - | 1 (4.5) | - | 0 | 0 | - | 1 (3.6) | - | 0 | 2 (2.2) | - | 2 (1.5) |
| 36- 48h | 0 | 0 | - | 1 (4.5) | - | 0 | 0 | - | 2 (7.1) | - | 0 | 3 (3.3) | - | 3 2.3) |
| 0 - 48h | 1 (5.6) | 0 | - | 4 (18) | 3.27 (0.40, 26.6) P=0.28 |
2 (8.7) | 2 (11) | 1.28 (0.20, 8.21) P=0.81 |
3 (11) | 1.23 (0.22, 6.76) P=0.84 |
3 (7.3) | 9 (10) | 1.34 (0.38, 4.68) P=0.68 |
12 (9.0) |
| Infants with: | ||||||||||||||
| 1 severe episode | 1 (5.6) | 0 | - | 3 (14) | 2.45 (0.28, 21.6) 0.46 | 2 (8.7) | 2 (11) | 1.28 (0.20, 8.21) P=0.81 |
2 (7.1) | 0.82 (0.13, 5.39) P=0.85 |
3 (7.3) | 7 (7.6) | 1.04 (0.28. 3.82) P=0.98 |
10 (7.5) |
| 2 severe episodes | 0 | 0 | - | 1 (4.5) | - | 0 | 0 | - | 0 | - | 0 | 1 (1.1) | - | 1 (0.8) |
| ≥ 3 severe episodes | 0 | 0 | - | 1 (4.5) | - | 0 | 0 | - | 0 | - | 0 | 1 (1.1) | - | 1 (0.8) |
| Severe episodes of low glucose concentrations per infant: | ||||||||||||||
| 0 - 12h | 1 (0, 1) | 0 (0, 0) | 0 (0, 1) | 0 (0, 1) | 0 (0, 1) | 0 (0, 1) | 0 (0, 1) | 0 (0, 1) | 0 (0, 1) | |||||
| 12 - 24h | 0 (0, 0) | 0 (0, 0) | 0 (0, 0) | 0 (0, 1) | 0 (0, 0) | 0 (0, 1) | 0 (0, 1) | 0 (0, 1) | 0 (0, 1) | |||||
| 24- 36h | 0 (0, 0) | 0 (0, 0) | 0 (0, 1) | 0 (0, 0) | 0 (0, 0) | 0 (0, 2) | 0 (0, 0) | 0 (0, 2) | 0 (0, 2) | |||||
| 36- 48h | 0 (0, 0) | 0 (0, 0) | 0 (0, 1) | 0 (0, 0) | 0 (0, 0) | 0 (0, 1) | 0 (0, 0) | 0 (0, 1) | 0 (0, 1) | |||||
| 0 - 48h | 0 (0, 1) | 0 (0, 0) | 0 (0, 2) | 0 (0, 1) | 0 (0, 1) | 0 (0, 4) | 0 (0, 1) | 0 (0, 4) | 0 (0, 4) | |||||
| Infants with long episodes of severe low glucose concentrations | 0 (5.6) | 0 | 1 (4.5) | 2 (8.7) | 1 (5.6) | 2 (7.1) | 3 (7.3) | 4 (4.3) | 7 (5.3) | |||||
| Long episodes of severe low glucose concentrations per infant | 0 (0, 1) | 0 (0, 0) | 0 (0, 1) | 0 (0, 1) | 0 (0, 1) | 0 (0, 4) | 0 (0, 4) | 0 (0, 1) | 0 (0, 4) | |||||
| Duration of low glucose concentration per infant (min) | 329.4 (336.2) | 301.2 (282.7) | 28.2 (−191.9, 248.3) P=0.80 |
193.8 (154.6) | 135.6 (−53.3, 324.6) P=0.15 |
332.8 (355.6) | 220.8 (268.8) | 111.9 (−135.6, 359.5) P=0.37 |
346.4 (520.5) | −13.6 (−305.8, 278.6) P=0.93 |
331.2 (341.4) | 276.0 (356.5) | 52.2 (−91.6, 202.1) P=0.46 |
294.6 (350.7) |
| Proportion of time with glucose concentration in defined range (% of total time measured per infant) | ||||||||||||||
| < 36 mg/dl [2 mol/l] | 0.46 (1.11) | 0.34 (0.99) | 0.12 (−0.54, 0.78) P=0.62 |
1.93 (6.86) | −1.47 (−4.78, 1.84) P=0.38 |
1.33 (3.46) | 0.63 (1.28) | 0.70 (−1.04, 2.44) P=0.42 |
2.60 (11.2) | −1.27 (−6.16, 3.61) P=0.60 |
0.95 (2.70) | 1.45 (7.04) | −0.52 (−2.77, 1.73) P=0.65 |
1.31 (0.64) |
| < 47 mg/dl [2.6mmol/l] | 10.4 (11.0) | 8.7 (10.5) | 1.7 (−5.0, 8.5) P=0.61 |
8.5 (13.2) | 1.87 (−6.01, 9.75) P=0.63 |
12.7 (14.5) | 7.26 (9.4) | 5.44 (−2.54, 13.43) P=0.17 |
11.6 (18.3) | 1.16 (−8.29. 10.6) P=0.81 |
11.7 (13.0) | 9.2 (13.7) | 2.46 (−2.54, 7.46) P=0.33 |
10.0 (13.5) |
| 54-72 mg/dl [3-4 mmol/l] | 51.6 (20.3) | 55.3 (22.1) | −3.7 (−17.1, 9.8) P=0.58 |
56.5 (22.7) | −4.8 (−18.8, 9.1) P=0.49 |
48.7 (23.5) | 65.8 (13.4) | −17.1 (−29.6, −4.5) P=0.009 |
65.7 (13.4) | −8.41 (−20.7, 3.88) P=0.18 |
50.0 (21.9) | 58.2 (20.3) | −8.20 (−15.9, −0.43) P=0.038 |
55.6 (21.1) |
| > 72 mg/dl [4 mmol/l] | 17.2 (17.7) | 14.7 (23.9) | 2.5 (−11.1, 16.0) P=0.70 |
18.1 (19.1) | −0.93 (−12.8, 10.9) P=0.75 |
14.8 (17.3) | 11.6 (11.6) | 3.2 (−6.4, 12.8) P=0.47 |
10.4 (16.0) | 4.4 (−5.0, 13.8) P=0.35 |
15.9 (17.3) | 13.6 (18.4) | 2.2 (−4.5, 9.0) P=0.51 |
14.3 (18.0) |
Data are mean (SD)
median (95% CI Hodges-Lehmann estimator), or number (%).
Low glucose: < 47 mg/dl [2.6 mmol/l], severe low glucose: < 36 mg/dl [2.0 mmol/l], long low glucose: > 30 min.
Compared with infants allocated to any single dose of placebo gel, those allocated to a single dose of dextrose gel 200mg/kg were less likely to experience any low glucose concentrations in the first 48 hours (RR 0.70, 95%CI 0.50-1.00, p = 0.049), and were less likely to experience a single, but not multiple, episodes (Table 3). The relative risks for these outcomes were similar for those allocated to a single dose of 400 mg/kg (RR 0.72, 95%CI 0.50-1.02, p = 0.066). Compared with infants allocated to multiple doses of placebo gel, infants allocated to multiple doses of dextrose gel (800mg/kg or 1,000mg/kg groups) were not less likely to experience single or multiple episodes of low interstitial glucose concentrations. Infants allocated to the 800 mg/kg group, but not those allocated to other dextrose dose groups, spent a larger proportion of time with a glucose concentration of 54 to 72 mg/dl [3 to 4 mmol/l] than those in the placebo groups (mean difference 17.1%, 95%CI 4.5-29.6%).
There were no differences between infants allocated to placebo gel and those allocated to dextrose gel in the time taken to reach the peak glucose after each 0.5 ml/kg dose of gel, the change in glucose concentration, the peak glucose concentration, or the area above or below the interstitial glucose curve (Table 4). The effects were similar regardless of the number of previous doses. Infants were less likely to experience a single episode of low glucose concentrations in the 4 h after each dose of 200 mg/kg dextrose gel, but there was no significant effect on the proportion of time spent in any specific glycemia range (Table 5).
Table 4:
Effect of a single dose of dextrose gel or placebo on interstitial glucose concentrations over the following 4 hours
| Placebo N=29 /82 epochs |
Dextrose 200mg/kg N=69 /175 epochs |
Mean difference, (95% CI), P |
|
|---|---|---|---|
| Change in glucose concentration from baseline to peak (mg/dl) | 15.12 (12.42, 17.82) | 13.14 (11.16, 14.94) | 1.98 (−1.31, 5.40) P=0.23 |
| Time from baseline to peak glucose concentration (h) | 0.43 (0.31, 0.55) | 0.39 (0.31, 0.48) | 0.036 (−0.11, 0.18) P=0.63 |
| Peak glucose concentration (mg/dl) | 62.4 (62.6, 70.0) | 66.1 (63.5, 68.4) | 0.342 (−4.10, 4.79) P=0.88 |
| Rate of rise in glucose concentration from baseline to peak (mg.dl−1.h−1) | 17.28 (9.54, 25.2) | 15.12 (10.08, 20.16) | 2.21 (−7.07, 11.50) P=0.64 |
| Area under the curve of interstitial glucose concentration from extrapolated baseline (mg.dl−1.h) | 3.42 (−2.34, 9.00) | 4.50 (0.72, 8.46) | 1.26 (−5.58, 8.10) PP=0.72 |
| Area above the curve of interstitial glucose concentration from extrapolated baseline (mg.dl−1.h) | 10.26 (5.22, 15.12) | 11.88 (8.82, 15.12) | 1.80 (−4.14, 7.56) P=0.55 |
| Area below curve of interstitial glucose concentration from extrapolated baseline (mg.dl−1.h) | −9.72 (−12.96, −6.48) | −8.82 (−11.34, −6.48) | 0.72 (−3.42, 4.86) P=0.72 |
Data are mean (95% CI). See Methods for details of analyses.
Table 5:
The effect of a single dose of dextrose gel or placebo on measures of glycemia over the following 4 hours
| Placebo | Dextrose gel 200mg/kg |
RR or median difference, 95% CI, P |
Dextrose gel 400mg/kg |
RR or median difference, 95% CI, P |
|
|---|---|---|---|---|---|
| n=41 | n=42 | n=50 | |||
| Age at which low glucose concentration first detected (min) | 162.9 (56.0) | 160.2 (56.5) | 2.8 (−36.6, 42.1) P=0.89 |
166.0 (53.5) | −3.1 (−38.6, 32.5) P=0.84 |
| Infants with any episode of low glucose concentrations | 18 (41.9) | 16 (38.1) | 21 (42.0) | P=0.99 | |
| Infants with: | |||||
| 1 episode | 15 (36.5) | 13 (31.0) | P=0.04 | 19 (38.0) | P=0.62 |
| 2 episodes | 3 (7.1) | 3 (7.1) | 2 (4.0) | ||
| ≥3 episodes | 0 | 0 | 0 | ||
| Age at which severe low glucose concentrations first detected (min) | 133.0 (16.5) | 135.3 (58.4) | −2.3 (−83.0, 88.9) P=0.94 |
185.0 (52.0) | −52.0 (−129.7, 25.7) P=0.15 |
| Infants with episodes of severe low glucose concentrations | 3 (7.3) | 4 (9.5) | P=0.99 | 5 (10.0) | P=0.92 |
| Infants with: | |||||
| 1 severe episode | 2 (4.9) | 0 | P=0.24 | 0 | P=0.53 |
| 2 severe episodes | 0 | 1 (2.3) | 0 | ||
| ≥ 3 severe episodes | 1 (2.4) | 3 (7.1) | 5 (10.0) | ||
| Severe episodes of low glucose concentration per infant | 0 (0,3) | 0 (0, 1) | 0 (0, 4) | ||
| Total duration of low glucose concentration per infant (min) | 33.1 (49.5) | 36.4 (57.0) | −3.2 (−26.8, 20.4) P=0.79 |
24.1 (40.6) | 9.0 (−10.0, 28.0) P=0.35 |
| Proportion of time with glucose concentrations in defined range (% of total time measured per infant) | |||||
| <36 mg/dl [2 mol/l] | 2.5 (10.9) | 3.0 (10.9) | −0.52 (−5.3, 4.3) P=0.83 |
1.89 (6.4) | 0.60 (−3.1, 4.3) P=0.75 |
| < 47 mg/dl [2.6mmol/l] | 20.2 (9.9) | 23.2 (36.3) | −3.0 (−17.7, 11.7) P=0.69 |
16.3 (27.0) | 3.92 (−8.1, 15.9) P=0.52 |
| 54-72 mg/dl [3-4 mmol/l] | 49.6 (39.8) | 51.4 (41.8) | −1.8 (19.9, 16.3) P=0.84 |
51.2 (36.7) | −1.56 (−17.7, 14.6) P=0.85 |
| > 72 mg/dl [4 mmol/l] | 7.4 (17.7) | 6.8 (22.7) | 0.6 (−8.4, 9.6) P=0.90 |
6.1 (19.5) | 1.3 (−5.9, 8.5) P=0.72 |
Data are median (range), mean (SD) or number (%).
Low glucose = < 47 mg/dl [2.6 mmol/l], severe low glucose = < 36 mg/dl [2.0 mmol/l], long low glucose = > 30 min.
Discussion
The glycemic response to dextrose gel treatment for neonatal hypoglycemia assessed using CGM has been reported previously,(10) but we are not aware of any previous reports of the effects of dextrose gel prophylaxis and of different doses on glycemic regulation in the first 48 hours.
Our findings of the effect of prophylactic dextrose gel using CGM were consistent with those previously reported using blood glucose monitoring, with any dose of prophylactic dextrose gel reducing the chance of low glucose concentrations, particularly in the first 12 hours after birth.(7,8) However, the CGM data provide valuable additional reassurance about the likely safety of this approach. Amongst other concerns have been the possibility that prophylactic dextrose gel may result in inappropriate stimulation of insulin secretion, potentially resulting in recurrent or more severe hypoglycemia in infants at risk. Also, high or rapidly changing glucose concentrations and a greater proportion of time spent outside the central glucose concentration range of 54 to 72 mg/dl [3 to 4 mmol/l], particularly in the period after dextrose administration to treat hypoglycemia, have been associated with a greater risk of adverse neurodevelopmental outcomes.(9,10) However, in that cohort, when different treatments for hypoglycemia were examined separately, glycemic instability was greater after intravenous dextrose and after dextrose plus formula feeding but not after treatment with dextrose gel plus breastmilk. (10) Consistent with those findings, these CGM data show that dextrose gel prophylaxis followed by breastfeeding resulted in no increase in recurrent or severe episodes, no increase in the peak blood glucose concentration or the area above or below the interstitial glucose curve in the 4 hours after each dose, no high glucose concentrations (>144 mg/dl [8 mmol/L]) and a greater proportion of time spent with glucose concentrations in the central range. Taken together, these findings suggest that prophylactic dextrose results in more rather than less glucose stability while also reducing the incidence of low glucose concentrations.
It is perhaps surprising that larger and repeated doses of prophylactic dextrose gel did not appear to result in additional effects on measures of glycemia, although the 800mg/kg dose group did spend more time in the central 54 to 72 mg/dl [3 to 4 mmol/l] range. There was also no evidence of a cumulative response to repeat doses. This may reflect the physiological gradual increase in glucose concentrations in the hours after birth,(17) so that additional doses of gel after the initial glucose nadir do not add detectable benefit. However, the numbers of infants with CGM data available in each dose group was small, limited by the relatively small number of families who consented to CGM monitoring within the original pre-hPOD trial, which was powered to detect differences between dosage groups in the incidence of low blood glucose concentrations. This meant that we had very limited power to detect anything other than large differences between dosage groups in the CGM outcomes reported in this study.
We found that CGM is safe and clinically manageable, even in the postnatal ward setting, and was acceptable to families and clinical staff as a research tool in this setting. CGM detected low glucose concentrations in approximately one-third more infants than did routine clinical blood glucose monitoring; differences similar to those reported previously.(14) Importantly, CGM detected an additional 32% of infants with three or more episodes of low glucose concentration, and an additional 55% of infants who experienced episodes lasting longer than 30 minutes, but not more infants with glucose concentrations <36 mg/dl [2.0mmol/L]. These additional detected episodes are of uncertain clinical significance, and it may be reassuring that more severe hypoglycemia was not missed using blood glucose monitoring. However, repeated and prolonged hypoglycemia may be associated with later adverse outcomes(18-21) and undetected low glucose concentrations have been associated with poor executive function at 4.5 years of age.(3) Thus further follow-up of this trial cohort will be important to further assess the predictive value of CGM in infants at risk of neonatal hypoglycemia.
We defined an episode as one or more consecutive low interstitial glucose concentrations, to be able to compare the incidence in this cohort with previous reports (14), and to maximize the chances of detecting any possible differences between dosage groups. Because the CGM provides a reading every 5 minutes, this potentially means that even very brief episodes are included, and that sensor artifact cannot be excluded as contributing to some of these. Defining an episode as two (3,9) or even three consecutive low glucose concentrations would potentially reduce the risk of including artifacts and increase clinical relevance. Similarly, the proportion of time spent in a specified glucose range may be a more clinically relevant outcome measure for future CGM studies.
Although all infants were included in a research trial and were routinely screened and treated to maintain blood glucose concentrations at 47 mg/dl [2.6 mmol/L] or greater, CGM revealed that on average infants spent almost 4 hours with glucose concentrations below that level in the first 48 hours. This appears longer than the median of 15 minutes (range 0-145 min) previously reported in a cohort who were similarly screened and treated, but who were at higher risk because it included more preterm infants and those requiring NICU admission.(9) Longer exposure to low glucose concentrations in our cohort may be a reflection of infants who received less treatment. For example, only 7.5% of this cohort were admitted to NICU, compared with 39% in the previous cohort. Many infants admitted to the NICU receive intravenous dextrose, which may have helped maintain higher glucose concentrations. However, the proportion of time spent in the middle range of 54 to 72 mg/dl [3 to 4 mmol/l] (56%) was similar in the two cohorts. Taken together, these findings further emphasize the need for a margin of safety between the glucose concentration threshold selected for intervention in infants with low glucose concentrations and that thought likely to result in brain injury, as current screening and treatment approaches, at least in these cohorts, still result in infants spending considerable periods of time below this threshold.
We conclude that prophylactic dextrose gel given to infants at risk of neonatal hypoglycemia who are otherwise unlikely to require NICU admission reduces the risk of low glucose concentrations while increasing rather than decreasing glucose stability in the first 48 hours after birth.
Acknowledgments
We thank all the babies and their families who participated in the study; the staff at National Women’s Hospital, Auckland and Birthcare, Auckland for their support, patient referrals, undertaking of trial interventions and help with point of care testing; and all members of the hPOD study group.
Funding Sources:
The pre-hPOD trial was supported by grants from the A+ Trust (A+5696) to JoH; Auckland Medical Research Foundation (1113012) to JA; Cure Kids (3537) to JA; Lottery Health Research (326844) to JA, and philanthropic donations to the University of Auckland Foundation to JaH. Further analysis of the CGM data was funded in part by grants from the Health Research Council of New Zealand (13/131, 15/216) and the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of Health (R01HD091075) to JaH.
Abbreviation:
- NICU
neonatal intensive care unit
Footnotes
Conflicts of Interest:
JaH has in the past received an undirected research grant from Biomed NZ who manufactures dextrose gel. Neither Biomed nor the funders of this study had any role in study design, data collection, and analysis, preparation of the manuscript, or decision to publish. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Eunice Kennedy Shriver National Institute of Child Health and Human Development or the National Institutes of Health.
References
- (1).Hay WW Jr, Raju TNK, Higgins RD, Kalhan SC, Devaskar SU. Knowledge gaps and research needs for understanding and treating neonatal hypoglycemia: workshop report from Eunice Kennedy Shriver National Institute of Child Health and Human Development. J Pediatr 2009;155:612–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (2).Kallem VR, Pandita A, Gupta G. Hypoglycemia: When to treat? Clin Med Insights Pediatr 2017;11:1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (3).McKinlay CJD, Alsweiler JM, Anstice NS, Burakevych N, Chakraborty A, Chase JG, et al. Association of neonatal glycemia with neurodevelopmental outcomes at 4.5 years. JAMA Pediatr 2017;171:972–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (4).Kaiser JR, Bai S, Gibson N, Holland G, Lin TM, Swearingen CJ, et al. Association between transient newborn hypoglycemia and fourth-grade achievement test proficiency: a population-based study. JAMA Pediatr 2015;169:913–21. [DOI] [PubMed] [Google Scholar]
- (5).Harris DL, Weston PJ, Signal M, Chase JG, Harding JE. Dextrose gel for neonatal hypoglycaemia (the Sugar Babies Study): a randomised, double-blind, placebo-controlled trial. Lancet 2013;382:2077–83. [DOI] [PubMed] [Google Scholar]
- (6).Harris DL, Alsweiler JM, Ansell JM, Gamble GD, Thompson B, Wouldes TA, et al. Outcome at 2 years after dextrose gel treatment for neonatal hypoglycemia: follow-up of a randomized trial. J Pediatr 2016;170:54–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (7).Hegarty JE, Harding JE, Gamble GD, Crowther CA, Edlin R, Alsweiler JM. Prophylactic oral dextrose gel for newborn babies at risk of neonatal hypoglycaemia: a randomised controlled dose-finding trial (the pre-hPOD study). PLoS Med 2016;13:e1002155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (8).Harding JE, Hegarty JE, Crowther CA, Edlin RP, Gamble GD, Alsweiler JM, et al. Evaluation of oral dextrose gel for prevention of neonatal hypoglycemia (hPOD): A multicenter, double blind, randomized controlled trial. PLoS Med 2021; 18: e1003411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (9).McKinlay CJD, Alsweiler JM, Ansell JM, Anstice NS, Chase JG, Gamble GD, et al. Neonatal glycemia and neurodevelopmental outcomes at 2 years. N Engl J Med 2015;373:1507–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (10).Burakevych N, McKinlay CJD, Harris DL, Alsweiler JM, Harding JE. Factors influencing glycaemic stability after neonatal hypoglycaemia and relationship to neurodevelopmental outcome. Sci Rep 2019;9:8132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (11).Bridges BC, Preissig CM, Maher KO, Rigby MR. Continuous glucose monitors prove highly accurate in critically ill children. Crit Care 2010;14:R176. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (12).Beardsall K, Ogilvy-Stuart AL, Ahluwalia J, Thompson M, Dunger DB. The continuous glucose monitoring sensor in neonatal intensive care. Arch Dis Child Fetal Neonatal Ed 2005;90:F307–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (13).Beardsall K, Vanhaesebrouck S, Ogilvy-Stuart AL, Van hole C, van Weissenbruch M, Midgley P, et al. Validation of the continuous glucose monitoring sensor in preterm infants. Arch Dis Child Fetal Neonatal Ed 2013;98:F136–40. [DOI] [PubMed] [Google Scholar]
- (14).Harris DL, Battin MR, Weston PJ, Harding JE. Continuous glucose monitoring in newborn babies at risk of hypoglycemia. J Pediatr 2010;157:198–202. [DOI] [PubMed] [Google Scholar]
- (15).Platas II, Lluch MT, Almiñana NP, Palomo AM, Sanz MI, Vidal XK. Continuous glucose monitoring in infants of very low birth weight. Neonatology 2009;95:217–23. [DOI] [PubMed] [Google Scholar]
- (16).Signal M, Le Compte A, Harris DL, Weston PJ, Harding JE, Chase JG, et al. Impact of retrospective calibration algorithms on hypoglycemia detection in newborn infants using continuous glucose monitoring. Diabetes Technol Ther 2012;14:883–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (17).Harris DL, Weston PJ, Gamble GD, Harding JE. Glucose profiles in healthy term infants in the first 5 days: The glucose in well babies (GLOW) study. J Pediatr 2020;223:34–41. [DOI] [PubMed] [Google Scholar]
- (18).Duvanel CB, Fawer CL, Cotting J, Hohlfeld P, Matthieu JM. Long-term effects of neonatal hypoglycemia on brain growth and psychomotor development in small-for-gestational-age preterm infants. J Pediatr 1999;134:492–8. [DOI] [PubMed] [Google Scholar]
- (19).Lucas A, Morley R, Cole TJ. Adverse neurodevelopmental outcome of moderate neonatal hypoglycaemia. BMJ 1988;297:1304–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (20).Qiao L, Wang J, Yan J, Xu S, Wang H, Zhu W, et al. Follow-up study of neurodevelopment in 2-year-old infants who had suffered from neonatal hypoglycemia. BMC Pediatr 2019;19:133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- (21).Melana N, Ahmed N, Soni RK, Goyal M. Neurodevelopmental outcome in neonates with hypoglycaemia and associated risk factors: A follow up study. J Pregnancy Child Health 2017;4:323. [Google Scholar]