INTRODUCTION — During the normal transition to extrauterine life, blood glucose concentration in the healthy term newborn falls during the first one to two hours after delivery, reaching a nadir with a median concentration of approximately 55 mg/dL. It is important to differentiate this normal physiologic transitional response from disorders that result in persistent or recurrent hypoglycemia, which if left untreated may lead to significant neurologic and developmental sequelae.
This topic will discuss the management and outcome of neonatal hypoglycemia, including evaluation of persistent hypoglycemia. The physiology of normal transient neonatal low blood glucose levels, causes of persistent or pathologic neonatal hypoglycemia, and the clinical manifestations and diagnosis of neonatal hypoglycemia are discussed separately. (See "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia".)
GOALS — The goals of managing neonatal hypoglycemia are:
●To correct blood glucose levels in symptomatic patients (see "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia", section on 'Clinical presentation')
●To prevent symptomatic hypoglycemia in at-risk patients
●To avoid unnecessary treatment of infants with physiologic low blood glucose during the transition to extrauterine life, which will self-resolve without intervention
●To identify newborns with a serious underlying hypoglycemic disorder
●Prevent long-term neurologic sequelae of neonatal hypoglycemia
Clinical guidelines have been developed by the American Academy of Pediatrics (AAP) and the Pediatric Endocrine Society (PES) [1,2]. The goals of these guidelines are to reduce neurologic impairment due to neonatal hypoglycemia (see 'Neurodevelopmental outcome' below), while minimizing overtreatment of neonates with normal transitional low glucose concentrations. In addition, the PES guidelines provide guidance on how and when to identify infants with a serious underlying persistent hypoglycemic disorder. (See 'Society guideline links' below.)
However, it is important to recognize that a precise blood glucose level and/or duration of hypoglycemia that accurately predict poor neurodevelopmental outcome have not been established (see 'Neurodevelopmental outcome' below). An optimal management strategy based on specific blood glucose thresholds to reduce adverse long-term neurologic outcomes has not been identified . (See "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia", section on 'Challenge of defining neonatal hypoglycemia'.)
THRESHOLDS FOR TREATMENT — Threshold plasma glucose levels consistent with limited published data and the American Academy of Pediatrics (AAP) and Pediatric Endocrine Society (PES) guidelines are used to provide a margin of safety for infants who are at risk for neonatal hypoglycemia [1,2,4].
The thresholds used for intervention are as follows (table 1):
●Symptomatic patients – This includes patients with jitteriness/tremors, pathological hypotonia, changes in level of consciousness, apnea/bradycardia, cyanosis, tachypnea, pathological poor feeding, sustained hypothermia, and/or seizures). (See "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia", section on 'Clinical presentation'.)
Plasma glucose thresholds used for treatment in these patients are as follows:
•<48 hours after birth: <50 mg/dL (2.8 mmol/L)
•≥48 hours after birth: <60 mg/dL (3.3 mmol/L)
●Asymptomatic patients – This includes patients who undergo glucose screening because they are at risk for hypoglycemia (eg, preterm infant, maternal diabetes, large or small for gestational age) and patients who have low glucose concentration identified as an incidental laboratory finding. (See "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia", section on 'Who should be screened?'.)
Plasma glucose thresholds used for treatment in these patients are as follows [1,4]:
•<4 hours after birth: <25 (1.4 mmol/L)
•4 to <24 hours after birth: <35 mg/dL (1.9 mmol/L)
•24 to <48 hours after birth: <50 mg/dL (2.8 mmol/L)
•≥48 hours after birth: <60 mg/dL (3.3 mmol/L)
●In newborns with a suspected or confirmed genetic hypoglycemia disorder (such as a family history of a hypoglycemia disorder or physical exam features consistent with Beckwith-Wiedemann syndrome), the threshold for plasma glucose concentrations is <70 mg/dL (3.9 mmol/L). This treatment goal is higher because the risks of harm from repetitive low glucose concentrations in this population are significant [5,6]. In addition, consultation with a specialist should be considered for further diagnostic testing to diagnose the underlying disorder .
TREATMENT TARGET — For all neonates receiving intervention, regardless of symptoms or underlying etiology, we target glucose concentrations between the lower limit (ie, the thresholds defined above) and an upper limit of 90 to 100 mg/dL (5 to 5.5 mmol/L) (table 1).
MANAGEMENT APPROACH — The treatment of hypoglycemia is a stepwise process depending on the presence or absence of symptoms and signs, and the response of the infant at each step. The following sections describe the approach in term and late preterm infants. Our approach is generally consistent with guidelines published by the American Academy of Pediatrics (AAP) and the Pediatric Endocrine Society (PES) [1,2,7].
Severely symptomatic patients
Overview — Newborns with hypoglycemia who have severe signs and symptoms (lethargy, coma, seizures) require urgent intervention. These signs are an indicator of possible brain injury. For this reason, intravenous (IV) dextrose is used to increase blood glucose (BG) levels in these patients [1,2]. (See "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia", section on 'Symptomatic infants' and 'Symptomatic hypoglycemia' below.)
Therapy should be initiated while awaiting laboratory confirmation of low plasma glucose levels. Although these severe symptoms most commonly occur when BG is <25 mg/dL (1.4 mmol/L), there is great variability in the clinical response in neonates to low BG ; some newborns become symptomatic at the same or even higher BG levels than those observed in asymptomatic infants. As a result, there is not a precise numerical BG level that accurately predicts when and if a neonate will present with severe symptoms. (See "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia", section on 'Challenge of defining neonatal hypoglycemia'.)
IV dextrose infusion — Hypoglycemic neonates with severe symptoms (lethargy, coma, seizures) require urgent intervention with IV dextrose.
●Initial treatment – Initial treatment is as follows:
•An initial bolus of IV dextrose (0.2 g/kg) given over 5 to 15 minutes (2 mL/kg of 10% dextrose in water [D10W])
•Followed by a continuous IV dextrose infusion at a rate of 5 to 8 mg/kg of dextrose per minute
●Subsequent monitoring and titration – Plasma glucose concentration should be measured 30 to 45 minutes after the initiation of IV therapy, and the infusion rate or dextrose concentration adjusted as needed to maintain BG >50 mg/dL (2.8 mmol/L) during the first 48 hours after birth and >60 mg/dL (3.3 mmol/L) beyond 48 hours  with an upper limit of 90 to 100 mg/dL (5 to 5.5 mmol/L). (See 'Treatment target' above.)
●Maximum infusion rate – The maximum rate of glucose infusion for treatment is limited by the maximum amount of fluids that can be administered to a patient (this is variable for each patient, but we have used rates as high as 200 mL/kg per day while monitoring for evidence of hyponatremia and fluid overload) and the maximum concentration of dextrose for the type of vascular access. In general, if the glucose infusion rate exceeds 12 mg/kg per minute or infusion rates exceed 160 mL/kg per day, other interventions should be considered. (See 'Other therapeutic options' below.)
The maximum dextrose concentrations for fluid administered through a peripheral IV catheter or a low lying umbilical venous catheter is 12.5 percent, and through a central venous catheter (including a centrally positioned umbilical venous catheter) is 25 percent. In severe cases, rates of fluid administration required to deliver sufficient glucose to treat hypoglycemia may be greater than the rate of maintenance fluids. In these cases, the patient's fluid balance and clinical status should be monitored closely for volume overload, looking for evidence of pulmonary edema, heart failure, and hyponatremia. Infants who depend upon high infusion rates or a dextrose concentration >12.5% require placement of a central venous catheter. In some cases, diuretics may be warranted to manage fluid overload.
●Weaning and transition to oral feeds – When the BG is stabilized and maintained within the target range, the glucose infusion rate can be tapered slowly, and the infant can be transitioned to oral feeds.
The weaning protocol used at our center is as follows:
•Weaning is typically started when the infant's BG levels have been in the target range for at least six to nine hours. (See 'Treatment target' above.)
•If BG levels are above the target range, the glucose infusion rate should be decreased sooner. In such cases, we decrease the infusion by 0.4 mg/kg min and recheck the BG 30 to 45 minutes later.
•For infants who remain asymptomatic while in the target range for six to nine hours, BG levels are obtained before each feed (every three to four hours) unless otherwise noted. We use a taper schedule based on mg/kg per min instead of mL/min to differentiate between glucose management and fluid management:
-BG ≥90 mg/dL (5.5 mmol/L): Decrease by approximately 1.6 mg/kg per min
-BG ≥70 mg/dL (3.9 mmol/L) to <90 mg/dL (5.5 mmol/L): Decrease by approximately 1.2 mg/kg per min
-BG ≥60 mg/dL (3.3 mmol/L) to <70 mg/dL (3.9 mmol/L): Decrease by approximately 0.8 mg/kg per min
-BG ≥50 mg/dL (2.8 mmol/L) to <60 mg/dL (3.3 mmol/L): Decrease by approximately 0.4 mg/kg per min
-BG ≥40 mg/dL (2.2 mmol/L) to <50 mg/dL (2.8 mmol/L) and the previous BG was ≥50 mg/dL (2.8 mmol/L): No change
-BG ≥40 mg/dL (2.2 mmol/L) to <50 mg/dL (2.8 mmol/L) and the previous BG was <50 mg/dL (2.8 mmol/L): Increase by approximately 0.4 mg/kg per min and recheck in 30 to 45 minutes
-BG <40 mg/dL (2.2 mmol/L): Increase by approximately 0.8 mg/kg per min and recheck in 30 to 45 minutes.
•If severe signs or symptoms (eg, lethargy, coma, seizures) develop during the wean, give an IV bolus of dextrose (0.2 g/kg) over 5 to 15 minutes (2 mL/kg of 10 percent D10W), increase the continuous dextrose infusion rate by approximately 1.6 mg/kg/min, and recheck the BG in 30 to 45 minutes. For these infants, we wait for them to have BG levels in the target range for six to nine hours before attempting another wean.
This protocol is not appropriate for infants with a known or suspected genetic hypoglycemic disorder. Such infants should be managed in consultation with an endocrinologist.
Other therapeutic options
Glucagon — Administration of glucagon should be considered in the rare patients with persistent BG <50 mg/dL (2.8 mmol/L) despite continuous maximum IV dextrose infusion. We suggest starting glucagon at an initial dose of 20 to 30 mcg/kg, which can be given as an intramuscular or subcutaneous injection, or a slow IV push over one minute . However, a wide range of glucagon doses (20 to 200 mcg/kg, maximum dose of 1 mg) have been used to treat infants with acute severe hypoglycemia [9-13].
BG levels typically rise by approximately 30 to 50 mg/dL within 15 to 30 minutes of administration. The effect lasts approximately two hours, though BG levels may drop sooner. Due to glucagon's short duration of action, BG levels should be checked frequently and repeat doses of glucagon may be needed. In refractory cases, a continuous IV glucagon infusion can be used (1 mg glucagon total infused over 24 hours, which is equivalent to approximately 10 to 20 mcg/kg per hour).
If the BG does not rise within 20 minutes of glucagon administration, then a repeat dose of glucagon (200 mcg/kg) is given. Failure to respond to glucagon should raise suspicion for an underlying metabolic disorder, particularly glycogen storage disorder, a defect in glycogen synthesis, or fatty acid oxidation disorder. These patients, require further evaluation. (See 'Evaluation of infants with persistent hypoglycemia' below and "Overview of inherited disorders of glucose and glycogen metabolism".)
Diazoxide — Diazoxide therapy is becoming more common for management of neonatal hyperinsulinemic hypoglycemia . If this is a concern, consultation with a pediatric endocrinologist is warranted to assist in identifying the cause of hypoglycemia, determining whether diazoxide is appropriate, and guiding the initial dosing and monitoring of diazoxide therapy. In addition, the endocrinologist is typically the clinician who will manage these infants in the outpatient setting.
In our center, the starting dose of diazoxide is typically between 5 to 15 mg/kg/day . We use lower doses (5 to 10 mg/kg/day) for suspected cases of nongenetic hyperinsulinism, such as infants of mothers with diabetes or perinatal stress hyperinsulinism. For suspected genetic forms of hyperinsulinism, higher doses (15 mg/kg/day) are typically needed. Diuretic therapy is also started to counteract the fluid retention occasionally observed in neonates on diazoxide. In patients at risk for heart failure, we obtain an echocardiogram prior to initiating diazoxide.
The exact timing of when to start diazoxide for hypoglycemia associated with hyperinsulinism depends on several factors, including the pregnancy and delivery history, gestational and postconceptual age, physical examination, and the type and amount of supplemental dextrose being used and whether this is increasing or decreasing.
Glucocorticoids — Administration of glucocorticoid therapy (hydrocortisone 2 to 6 mg/kg per day divided in two to three doses orally or intravenously) has been used to treat infants requiring a glucose infusion rate ≥12 mg/kg per minute. However, due to the potential side effects of glucocorticoid administration, its use should be restricted to a short course (one to two days), unless a patient has documented adrenal insufficiency. The proposed mechanism of action of glucocorticoids is stimulation of gluconeogenesis and reduction in peripheral glucose utilization. Serum cortisol and insulin concentrations during an episode of hypoglycemia should be measured before beginning glucocorticoid treatment, if possible.
Need for further evaluation — For infants with severe hypoglycemia that require prolonged and/or high rates of IV dextrose infusion to maintain glucose threshold levels, further laboratory evaluation is warranted, especially if no underlying cause has been identified by either history or physical examination. In these uncommon situations such as persistent hyperinsulinemic hypoglycemia, consultation with a pediatric endocrinologist or a clinician with expertise in managing neonatal hypoglycemia is recommended. (See 'Evaluation of infants with persistent hypoglycemia' below.)
Asymptomatic and mildly symptomatic infants — Asymptomatic patients are typically identified through glucose screening performed in infants who are at risk for hypoglycemia (eg, maternal diabetes, large or small for gestational age, late preterm), or it may be identified as an incidental laboratory finding. (See "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia", section on 'Who should be screened?'.)
Mild symptoms of hypoglycemia include jitteriness or tremors.
Oral feeds — Oral feeding is the initial intervention for newborns who are at risk for hypoglycemia and those with documented hypoglycemia if they have no associated symptoms or have only mild symptoms (eg, jitteriness). Patients with documented hypoglycemia are also treated with buccal dextrose gel, as discussed below. (See 'Dextrose gel' below.)
●Initial feeding – Infants who are at risk for hypoglycemia should be fed within the first hour of life . The BG should be measured within 30 minutes after the initial feeding, or within 90 to 120 minutes of age if the first feeding is delayed. (See "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia", section on 'Timing of glucose screening'.)
While breast milk is strongly preferred, formula feeding may be provided for infants when breast milk is not available.
●Subsequent feedings – The frequency of subsequent oral feedings and/or need for additional interventions depend upon the age of the neonate, whether there are symptoms of hypoglycemia, and the plasma glucose level. Thresholds for intervention vary over the first 48 hours after birth (table 1), as described above. (See 'Thresholds for treatment' above and 'Treatment target' above.)
Our suggested approach is as follows:
•Initial BG within target range – Infants whose initial postprandial BG is >25 mg/dL (1.4 mmol/L) should be offered feedings at two- to three-hour intervals and should be monitored for signs of hypoglycemia. Prefeeding BG levels should be checked every three to six hours for the first 24 to 48 hours. (See "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia", section on 'Timing of glucose screening'.)
•Infants less <4 hours of age with BG below threshold – For asymptomatic infants the threshold for intervention is a BG <25 mg/dL (1.4 mmol/L); for patients with symptoms (eg, jitteriness), the threshold for intervention is BG <50 mg/dL (2.8 mmol/L) (table 1). For these infants, we suggest buccal dextrose gel followed by an oral feed (preferably with breast milk). BG should be measured 30 to 45 minutes after completion of the feed. If the subsequent BG increases to within the target range (table 1), oral feedings should continue every two to three hours with preprandial BG measurements. However, if the BG remains below threshold after the additional oral feed, IV dextrose is administered. (See 'Use of IV dextrose in asymptomatic patients' below.)
•Infants between 4 and 24 hours of age with BG below threshold – For asymptomatic infants ≥4 and ≤24 hours old, the threshold for intervention is a BG <35 mg/dL (1.9 mmol/L); for patients with symptoms (eg, jitteriness), the threshold for intervention is a BG <50 mg/dL (2.8 mmol/L) (table 1). For these infants, we suggest buccal dextrose gel followed by an oral feed (preferably with breast milk). BG should be measured 30 to 45 minutes after completion of the feed. If the subsequent BG increases within the target range (table 1), oral feedings should continue every two to three hours with preprandial BG measurements. However, if the BG remains below threshold after the additional oral feed, IV dextrose is administered. (See 'Use of IV dextrose in asymptomatic patients' below.)
•Infants who develop concerning symptoms – If a neonate develops severe symptoms of hypoglycemia (lethargy, coma, seizures), IV dextrose therapy should be initiated while awaiting laboratory confirmation of low BG. (See 'IV dextrose infusion' above.)
●Therapeutic use in infants with documented hypoglycemia – We suggest dextrose gel in conjunction with milk feeding for newborn infants with asymptomatic or mildly symptomatic hypoglycemia. For these neonates, dextrose gel is an effective intervention to increase BG that is safe and easy to administer [16-20]. It also appears to be cost effective .
Buccal dextrose gel is given at a dose of 0.2 g/kg (0.5 mL/kg of 40% dextrose gel) which does not impair subsequent feeding . It is administered before each feed if the preprandial BG is below the threshold for intervention (table 1). If the BG remains below threshold after treatment with dextrose gel and oral feeding, IV dextrose is administered. In addition, IV dextrose is generally warranted if BG remains <45 mg/dL (2.5 mmol/L) after three feedings or if the newborn requires >5 doses of dextrose gel during the course of their birth hospitalization. (See 'Use of IV dextrose in asymptomatic patients' below.)
A neonatal 40% dextrose gel is commercially available for this purpose; however, it may not be available in all settings. If a neonatal product is not available, over-the-counter (OTC) gels that are commonly used for treatment of hypoglycemia in patients with diabetes can be used. These OTC products have varying glucose concentrations, hence it is important to appropriately adjust the amount given based on the concentration to ensure that the recommended dose of 200 mg/kg is administered . These formulations may also contain artificial colorants, flavors, and preservatives .
The efficacy of dextrose gel for treating neonatal hypoglycemia is supported by randomized trials and meta-analyses [16,18,20,24]. In a meta-analysis of two trials, involving 312 at-risk term and late preterm infants, dextrose gel was more effective in raising BG levels compared with placebo (mean difference 4.3 mg/dL [0.24 mmol/L]); though the need for IV dextrose treatment was similar in both groups (13 versus 17 percent; RR 0.78, 95% CI 0.46-1.32) . In the larger of the two trials, the Sugar Babies trial (n=237), neonates assigned to dextrose gel were less likely to have treatment failure (defined as BG <45 mg/dL [2.5 mmol/L] after two treatment attempts) compared with those assigned to placebo (14 versus 24 percent; RR 0.57, 95% CI 0.33-0.98 . Dextrose gel was effective in raising BG levels both in infants who received breast milk and those who received formula, though the absolute increase was greater among formula-fed infants. Three infants in the trial experienced severe hypoglycemia (ie, BG <18 mg/dL [<1 mmol/L]), all three were in the placebo group. There were no other serious adverse events reported in the trial. In a follow-up report of the Sugar Babies trial, rates of neurodevelopmental impairment (NDI) at age two years were similar in both groups (38 versus 34 percent: RR 1.11, 95% CI 0.75-1.63), as were rates of processing difficulty (10 versus 18 percent; RR 0.52, 95% CI 0.23-1.15) .
There are theoretical concerns that dextrose gel may be detrimental to early establishment of breastfeeding, but this was not observed in the Sugar Babies trial.
The results of the Sugar Babies trial support our preferred approach for most asymptomatic or mildly symptomatic neonates with hypoglycemia, which consists of administering buccal dextrose gel followed by breastfeeding. This regimen is preferable to providing formula as it promotes breastfeeding while also reducing the risk of recurrent hypoglycemia. (See 'Oral feeds' above.)
●Prophylactic early use in at-risk infants – We suggest not using dextrose gel to prevent hypoglycemia in at-risk newborns in the absence of documented hypoglycemia. The rationale for prophylactic therapy is based upon the notion that identifying and treating hypoglycemia after the first feeding may not be sufficient to prevent adverse sequalae (ie, NDI). In addition, it was thought that prophylactic use of dextrose gel might reduce the need for laboratory testing and other interventions that often separate mother and newborn and interfere with breastfeeding. However, the available evidence does not support this approach. The clinical trial data suggest prophylactic dextrose gel modestly reduces the incidence of hypoglycemia, but it does not appear to reduce the risk of NDI [25-27].
The two trials investigating this question were the prevention of neonatal hypoglycemia (hPOD) trial, which was a multicenter randomized trial involving 2149 newborn infants at risk for hypoglycemia (maternal diabetes, large or small for gestational age, late preterm) and the pre-hPOD trial, which was a similar trial that enrolled 416 at-risk newborns [24,28]. In a meta-analysis of both trials, early (ie, within one hour after birth) prophylactic administration of oral dextrose gel reduced the incidence of hypoglycemia compared with placebo (38 versus 43 percent, relative risk [RR] 0.87, 95% CI 0.79-0.95) . However, rates of IV dextrose use were the same in both groups (37 percent) and the percentage of newborns who were separated from parents/caregivers for treatment of hypoglycemia was also similar in both groups (5.6 versus 5 percent).
Both trials reported outcomes at two years [26,27], which were not pooled in the meta-analysis because the two-year outcomes for hPOD were not yet available. In hPOD, two-year outcomes were reported only for a subset of the full trial cohort because many participants could not travel to the study centers due to COVID restrictions. Among the 1197 children who underwent follow-up evaluation, overall rates of NDI were similar between those who received dextrose gel compared with placebo (21 versus 19 percent; adjusted RR 1.13, 95% CI 0.90-1.41) . However, children in the dextrose gel group had a higher rate of motor delay (2.5 versus 0.7 percent; adjusted RR, 3.79, 95% CI 1.27-11.32) and slightly lower composite scores for cognitive, language, and motor performance on standardized testing (with mean differences ranging from one to two points on a 100-point scale). The reasons why children in the dextrose gel group would have adverse neurodevelopmental outcomes is unclear. It could represent a chance finding given that >30 secondary outcomes were evaluated. The earlier pre-hPOD trial did not detect any differences in neurodevelopment or executive function at age two years between children who received dextrose gel versus placebo . Taken together, these trials suggest that early prophylactic use of dextrose gel modestly reduces the incidence of neonatal hypoglycemia, but it does not appear to reduce the need for IV therapy, nor separation from the parents/caregivers, nor the risk of NDI. As such, we suggest not using this approach.
Use of IV dextrose in asymptomatic patients — In asymptomatic patients and those with mild symptoms (eg, jitteriness), use of IV dextrose is limited to the following patients:
●Newborns with persistent hypoglycemia that fails to increase above the treatment threshold after buccal dextrose and oral feeding (table 1)
●Newborns with persistent hypoglycemia that remains <45 mg/dL (2.5 mmol/L) after three oral feedings
●Newborns requiring more than five doses of dextrose gel during the course of their birth hospitalization
For asymptomatic patients, parenteral dextrose infusion is initiated as a continuous infusion. A bolus of dextrose is not administered because of concerns that a bolus dextrose leads to too rapid correction with subsequent neurologic sequela in asymptomatic infants . The rate of infusion for asymptomatic neonates is dependent on the clinical setting:
●For infants who are small for gestational age and/or with intrauterine growth restriction (IUGR), we start with a rate of 5 to 7 mg/kg per min
●For infants born to mothers with diabetes (in the absence of IUGR), we start with a rate of 3 to 5 mg/kg per min
●For infants who are large for gestational age (LGA), we start with a rate of 3 to 5 mg/kg per min
●For infants not in the above risk groups, we start with a rate of 4 to 6 mg/kg per min
An alternative approach is to base the initial intravenous dextrose infusion rate on the degree of hypoglycemia at the time the infusion is started (ie, using a lower infusion rate for less severe hypoglycemia). In a retrospective study, this approach resulted in decreased BG variability and decreased length and cost of the neonatal intensive care unit (NICU) stay .
BG levels should be measured 30 to 45 minutes after initiation of IV dextrose, and the rate of infusion increased if hypoglycemia persists. BG should be measured 30 to 45 minutes after any increase in the IV dextrose infusion rate.
Discharge criteria — It is important to ensure that neonates are able to maintain plasma glucose concentrations in a normal range through cycles of feeding and fasting prior to discharge. However, data are lacking to determine the optimal discharge criteria, particularly the minimal required threshold glucose level.
●General criteria – In our practice, for infants who have been identified as having hypoglycemia, preprandial BG through three feed-fast cycles should be >50 mg/dL (2.8 mmol/L) in infants <48 hours of age, and >60 mg/dL (3.3 mmol/L) in those who are ≥48 hours of life [2,7]. In general, if a neonate can maintain BG >60 mg/dL (3.3 mmol/L) after a 6- to 8-hour fast, it is likely that the infant is safe for discharge .
Other experts suggest a higher threshold of maintaining a preprandial BG >70 mg/dL (3.9 mmol/L) through several feed-fast cycles . Although this higher threshold would be more sensitive for identifying newborns with persistent hypoglycemia, it lowers specificity so that the number of unnecessary evaluations may increase. It also prolongs hospitalization, which may have an inadvertent negative effect on maternal-infant bonding for infants without persistent hypoglycemia.
●Safety fast – In some patients, we perform a safety fast prior to discharge. This safety fast may also be diagnostic if the etiology of the patient's hypoglycemia is unclear and appropriate laboratory studies are performed while the patient is hypoglycemic. In our practice we consider a safety fast prior to discharge in the following patients:
•Hypoglycemia associated with any of the following:
-Severe signs (lethargy, coma, seizures)
-No known risk factors, but required intravenous dextrose to treat hypoglycemia
-Family history of sudden infant death of unknown cause in a sibling
-Physical exam consistent with a congenital disorder associated with hypoglycemia (Beckwith-Wiedemann, hypopituitarism)
-Inability to consistently maintain plasma glucose above age-appropriate normal
•Family history of a chronic hypoglycemia disorder (in consultation with an endocrinologist)
For infants with a known or suspected genetic hypoglycemia disorder, discharge criteria should be made on a patient-specific basis in consultation with an endocrinologist.
PRETERM INFANTS — Preterm infants ≤34 weeks are at risk for low hypoglycemia, likely due to immaturity of counter-regulatory hormone systems and poor nutrient reserves . In addition, infants managed with parenteral nutrition are at risk of having hypoglycemia when transitioning to bolus enteral feeds . Even on full bolus enteral feeds, preterm infants have episodes of both low and high blood glucose (BG) levels [33,34]. It appears that infants who are at risk for growth failure are also at risk for recurrent and persistent episodes of hypoglycemia .
Although a safe plasma glucose concentration for these infants has not been established, experts in the field, including the author of this topic, suggest maintaining BG >50 to 60 mg/dL (2.8 to 3.3 mmol/L). This is likely to be a safe strategy to avoid long-term neurologic sequelae [2,35,36]. However, setting the targeted threshold levels must also take into account the patient's overall clinical and nutritional status. (See 'Preterm infants' below.)
The management for asymptomatic preterm infants who are able to receive sufficient nutrition through enteral feeds entails early feeds and monitoring BG levels . For those who are not expected to be able to receive enough enteral nutrition due to prematurity, parenteral nutrition, which includes dextrose, should be started promptly. These issues are discussed separately. (See "Approach to enteral nutrition in the premature infant" and "Parenteral nutrition in premature infants".)
Acute episodes of symptomatic hypoglycemia in preterm neonates are generally managed with intravenous dextrose (0.2 g/kg given over 5 to 15 minutes [2 mL/kg of 10% dextrose in water]).
EVALUATION OF INFANTS WITH PERSISTENT HYPOGLYCEMIA
Definition and timing of evaluation — Persistent hypoglycemia is defined as low plasma glucose concentrations that persist beyond the first 48 hours after birth or the requirement of intravenous dextrose infusion to treat hypoglycemia beyond 48 hours after birth . The normal physiologic transitional drop in blood glucose generally resolves within this timeframe. Thus, infants with persistent hypoglycemia beyond 48 hours after birth are more likely to have a pathological hypoglycemia disorder (transient or permanent). (See "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia", section on 'Normal transitional low glucose levels'.)
However, evaluation for hypoglycemic disorders should generally be deferred until 96 hours to a week or more after birth to ensure that the transitional period of adjusting the source of glucose from a continuous supply provided by the mother through the placenta to an intermittent supply from oral feeds has passed. In our practice, we recommend waiting at least seven days unless more intensive therapies (eg, glucagon, diazoxide) are being considered. We wait even longer for evaluation if the patient has a known risk factor for hypoglycemia or if the results of the evaluation are not expected to change management.
Deferring the evaluation is necessary because the biochemical features of normal physiologic transitional hypoglycemia (mild hyperinsulinism) can be difficult to distinguish from a pathologic process. For example, the constellation of findings consistent with hyperinsulinemic hypoglycemia in an older child (ie, inappropriately elevated serum insulin levels, low ketone body [beta-hydroxybutyrate] and free fatty acid concentrations, and a brisk rise in glucose concentrations following glucagon administration) is also seen in an asymptomatic and otherwise healthy term infant <48 hours old with normal low blood glucose levels. It remains uncertain exactly when this group of biochemical findings changes from normal physiology to representing a pathological hypoglycemic condition. However, for most healthy term newborns, this transition is believed to take place by 48 to 96 hours of age [2,37].
Evaluation — Once persistent hypoglycemia is established, further evaluation is warranted to determine the underlying cause. Most cases of persistent hypoglycemia in term infants will have biochemical features of hyperinsulinism and typically resolve in the first few weeks of life. However, some cases may persist and require ongoing medical management. This is true even in the absence of a suspected or defined genetic hypoglycemia disorder [38,39]. Therefore, the clinician needs to determine the appropriate and safe point at which glucose monitoring can be discontinued and when hospital discharge should occur, versus when it is likely that an ongoing hypoglycemic disorder exists, which requires further intervention and/or evaluation . Performing a safety fast prior to discharge may be helpful to make this determination, as discussed above. (See 'Discharge criteria' above.)
In cases with prolonged persistent hypoglycemia, consultation with a pediatric endocrinologist is recommended to help guide the evaluation and management.
The evaluation consists of a thorough history, physical examination, and, in some cases, laboratory evaluation. (See 'Laboratory evaluation' below.)
History — A thorough history can help determine the underlying cause of hypoglycemia (table 2). (See "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia", section on 'Pathogenesis of neonatal hypoglycemia'.)
●Prematurity – Preterm infants, as noted above, have poor nutrient reserves and immature counter-regulatory hormone systems, which increase their risk of hypoglycemia.
●Fetal growth restriction (FGR) – Infants with FGR are at risk for hypoglycemia due to poor nutrient reserves and hyperinsulinism.
●History of perinatal asphyxia or stress – Hyperinsulinism and increased metabolism in neonates with perinatal asphyxia or stress can contribute to hypoglycemia.
●History of maternal diabetes – Neonatal hyperinsulinism results in hypoglycemia. (See "Infants of women with diabetes", section on 'Hypoglycemia'.)
●Positive family history of an infant with neonatal hypoglycemia may be indicative of an underlying genetic disorder, including inborn errors of metabolism.
Physical findings — The physical examination may provide clues to an underlying cause of neonatal hypoglycemia as follows :
●Large or small for gestational age (LGA or SGA) (see "Large for gestational age newborn").
●Hemihypertrophy, macroglossia, and omphalocele are findings consistent with a diagnosis of Beckwith-Wiedemann syndrome (BWS) (see "Beckwith-Wiedemann syndrome", section on 'Clinical manifestations').
●Ambiguous genitalia, hypertension, hyponatremia, and hyperkalemia are features that may be seen in congenital adrenal insufficiency, which also is associated with hypoglycemia (see "Causes of primary adrenal insufficiency in children", section on 'Congenital adrenal hyperplasia').
●Hepatomegaly is seen in some glycogen storage and other hereditary metabolic diseases that can present with hypoglycemia, and with BWS.
●Midline facial defects and micropenis may be seen in cases of hypopituitarism.
Who should be tested? — Laboratory testing is warranted in patients who:
●Present with severe signs and symptoms of hypoglycemia (lethargy, coma, or seizures).
●Require prolonged and/or high rates of IV dextrose infusion for treatment, especially if there are no known risk factors for hypoglycemia.
●Have hypoglycemia that persists beyond is the expected course based upon history/underlying risk factors.
●Have historical or physical findings suggestive of an underlying genetic or developmental etiology (eg, family history of a genetic hypoglycemic disorder or if physical exam features suggest a syndromic hypoglycemic disorder such as BWS or congenital adrenal hyperplasia). In these patients, evaluation is warranted regardless of whether the infant has documented hypoglycemia).
Consultation with a clinician with expertise in managing neonatal hypoglycemia (ie, pediatric endocrinologist) should be considered in these cases, as the specific tests required to rule out a hypoglycemia disorder will vary based on the clinical setting (eg, family history and physical exam) .
Timing of "critical" blood test sampling — Because tests performed when the blood glucose (BG) levels are normal are generally not helpful in determining the underlying cause of hypoglycemia, critical blood test samples for the diagnostic evaluation should be obtained when BG levels are <50 mg/dL (2.8 mmol/L) when measured in a laboratory, or <40 mg/dL (2.2 mmol/L) when measured with a bedside glucometer. This difference in suggested blood glucose threshold is due to the inaccuracy of a bedside glucometer for diagnosing neonatal hypoglycemia. (See "Pathogenesis, screening, and diagnosis of neonatal hypoglycemia", section on 'How glucose testing is performed'.)
If a hypoglycemic episode does not develop spontaneously, it may be necessary to perform a six- to eight-hour fast (one skipped feed), which usually will result in an appropriately low BG. During the fast, frequent monitoring of vital signs, and BG levels (every hour) should be performed. If the BG concentration drops to <50 mg/dL (2.8 mmol/L) prior to six to eight hours, the critical blood samples should be obtained and then the fast terminated.
For patients who are receiving IV dextrose infusion, blood samples can still occur so long as the plasma glucose is <50 mg/dL (2.8 mmol/L). Interventions to treat hypoglycemia (ie, feeding or increasing the dextrose infusion rate) should be delayed until after the samples are obtained .
What tests to obtain? — In our practice, while the patient has a BG <50 mg/dL (2.8 mmol/L), we obtain the following tests:
●Confirmatory plasma glucose level measured in the laboratory
●Plasma insulin level
●Growth hormone level
●Other initial tests to consider include bicarbonate, lactate, and free fatty acids
These initial tests are used to distinguish diagnostic categories for neonatal hypoglycemia and help determine if other blood tests should be obtained, including plasma C-peptide, acylcarnitine profile, plasma free and total carnitine levels, serum amino acids, urine organic acids, or specific genetic tests (algorithm 1). These specific blood tests should be performed in consultation with a pediatric endocrinologist or other appropriate specialist. Changes in BG levels can rapidly alter many of these blood tests, particularly insulin, C-peptide, beta-hydroxybutyrate, cortisol, growth hormone, and free fatty acids .
Symptomatic hypoglycemia — Hypoglycemia with severe neuroglycopenic signs can result in brain injury that can be detected by magnetic resonance imaging (MRI). However, there are no available data that clearly define the glucose concentration or the duration of hypoglycemia associated with brain damage detected by MRI or other long-term neurologic sequelae.
A systematic review of the literature published in 2006 reported inconclusive evidence on the effect of neonatal hypoglycemia on neurodevelopment . Two subsequent studies using brain MRI suggest an association between hypoglycemia and brain injury in term infants [41,42]. Although it remains uncertain whether timely treatment of hypoglycemia will prevent brain injury and poor developmental outcome, experts in the field, including the author, recommend that hypoglycemia with severe neuroglycopenic signs should be aggressively treated given the potential significant adverse effects based on the available data [1,2,7].
Asymptomatic hypoglycemia — The outcome of children with asymptomatic neonatal hypoglycemia remains unclear.
The CHYLD (Children with HYpoglycemia and their Later Development) study was designed to examine the impact of neonatal hypoglycemia on infant and child development. It prospectively followed 528 infants who were born at ≥35 weeks gestation with risk factors for hypoglycemia (maternal diabetes, large for gestation, fetal growth restriction [FGR], or prematurity <37 weeks gestation) . All infants in the study were monitored for hypoglycemia and those with documented hypoglycemia (ie, blood glucose [BG] <47 mg/dL [2.6 mmol/L]) received treatment to maintain blood glucose above this threshold during the first 48 hours after birth. At two years of age, infants who received treatment for hypoglycemia had similar neurodevelopmental outcomes compared with those who did not require intervention. At follow-up in early childhood (mean age 4.5 years), children treated for neonatal hypoglycemia had higher rates of poor executive and visual motor function compared with those who did not require treatment . However, by school age (mean age 9.4 years), rates of low educational achievement were similar in both groups (47 versus 48 percent) . These rates are considerably higher than in healthy populations, suggesting that antenatal conditions that are associated with increased risk of neonatal hypoglycemia may have more of an impact on educational achievement than neonatal hypoglycemia itself.
A retrospective observational study of all newborn infants born at a single tertiary United States center reported that children who had experienced transient neonatal hypoglycemia (defined as BG <40 mg/dL [2.22 mmol/L]) had lower scores for literacy and math at fourth grade after adjusting for confounding factors . In addition, BG levels <45 mg/dL [2.5 mmol/L] were associated with lower literacy but not math scores.
Preterm infants — In preterm infants, controversy exists as to whether asymptomatic hypoglycemia causes neurologic injury and whether the threshold glucose concentrations for intervention should be lower in preterm than in term infants. Several studies report neurodevelopmental sequelae due to repeated or prolonged asymptomatic episodes of neonatal hypoglycemia in preterm patients. As a result, experts in the field, including the author, use a lower blood glucose level for intervention in preterm infants compared with term infants. (See 'Preterm infants' above.)
Supporting data for this approach include the following studies:
●Retrospective data analysis of a multicenter trial in preterm infants showed a correlation between prolonged hypoglycemia (BG <47 mg/dL [2.6 mmol/L] on five different days during the first two months of age) and lower Bayley mental and psychomotor scores at 18 months corrected age . Developmental delay or cerebral palsy was 3.5 times greater (95% CI 1.3-9.4) in the hypoglycemic infants. However, the frequency of glucose testing was variable and occurred more often in sicker infants. Furthermore, only arithmetic and motor scores were lower in hypoglycemic infants at 7.5 to 8 years of age .
●In a study of preschool-age children who were born moderately preterm (GA 32 to less than 36 weeks), multivariate analysis showed that hypoglycemia was associated with increased incidence of parent-reported developmental delay when the children were 43 to 49 months old (odds ratio [OR] 2.19, 95% CI 1.08-4.46) .
●The long-term impairment of neurodevelopment may be greater in preterm infants with repeated hypoglycemic episodes who are also small for gestational age (SGA). In a prospective study of 85 SGA preterm infants, repeated episodes of hypoglycemia were associated with a smaller head circumference at 18 months corrected age and lower psychometric testing scores at 3.5 and 5 years of age .
In contrast, results from two studies showed no association between hypoglycemia and increased risk of neurodevelopmental impairment:
●A population-based study from northern England of preterm infants (GA <32 weeks) born in 1990 and 1991 reported no differences in developmental status or physical disability based on psychometric assessment between 47 patients who had neonatal BG ≤45 mg/dL (2.5 mmol/L) on ≥3 days compared with matched control patients during follow-up at 2 and 15 years of age . In this cohort, daily measurements of blood glucose were obtained at a fixed time each morning for the first 10 days of life.
●Multivariant secondary analysis of longitudinal data from the multicenter Infant Health and Development Program study of preterm infants (GA <32 weeks) showed no difference between children with and without neonatal hypoglycemia in intellectual and cognitive skills, or academic achievement at 3, 8, and 18 years of age . For this study, hypoglycemia was defined as any blood glucose level that was ≤45 mg/dL (2.5 mmol/L).
Based on the available data, we use a BG threshold of 50 mg/dL for intervention in preterm infants until there is conclusive evidence that establishes a level that accurately predicts long-term outcome. (See 'Preterm infants' above.)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Hypoglycemia in the neonate".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topic (see "Patient education: Newborn hypoglycemia (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Management approach in term and late preterm newborns ‒ Treatment of neonatal hypoglycemia is a stepwise process depending on the presence or absence of symptoms and signs, and the response of the infant at each step. (See 'Management approach' above.)
•Thresholds and targets for treatment – Thresholds for intervention and target blood glucose (BG) levels are summarized in the table (table 1).
•Severe symptomatic hypoglycemia – Infants with severe symptomatic hypoglycemia (eg, lethargy, coma, and seizures) require prompt treatment with intravenous (IV) dextrose. Therapy should be started while awaiting laboratory confirmation.
IV dextrose is given with an initial bolus (0.2 g/kg) over 5 to 15 minutes (2 mL/kg of 10% dextrose in water [D10W]), followed by continuous infusion at an initial rate of 5 to 8 mg/kg per minute. If hypoglycemia persists, the infusion rate should be increased as needed. (See 'IV dextrose infusion' above.)
For the rare patient who fails to maintain BG in the target range despite maximal glucose infusion rate, we suggest glucagon (Grade 2C). The initial dose is 20 mcg/kg given via intramuscular or subcutaneous injection, or slow IV push. (See 'Glucagon' above.)
•Asymptomatic and mild symptomatic hypoglycemia – For newborns with documented hypoglycemia below the threshold values (table 1) who are asymptomatic (eg, at-risk infants identified through BG screening) or have only mild symptoms (eg, jitteriness), we suggest buccal dextrose gel followed by oral feeding rather than either intervention alone (Grade 2C). We encourage breastfeeding in this setting as we do for all newborns. This use of buccal dextrose gel followed by breastfeeding is preferable to providing formula as it promotes breastfeeding while also reducing the risk of recurrent hypoglycemia. (See 'Oral feeds' above and 'Dextrose gel' above.)
BG should be measured 30 to 45 minutes after completing the feed. If the subsequent BG increases to within the target range (table 1), oral feedings should continue every two to three hours while monitoring preprandial BG measurements.
If the subsequent BG remains below threshold, or if BG remains <45 mg/dL (2.5 mmol/L) after three oral feedings, we suggest IV dextrose (Grade 2C). (See 'Asymptomatic and mildly symptomatic infants' above.)
•At-risk newborns without documented hypoglycemia – For term and late preterm newborns who are at risk for hypoglycemia (maternal diabetes, small or large for gestational age, late preterm), we suggest early initial oral feeding (ie, within the first hour after birth) rather than prophylactic use of buccal dextrose gel (Grade 2C). Prophylactic dextrose gel may modestly reduce the incidence of hypoglycemia, but it does not appear to reduce the need for IV dextrose, nor does it reduce the risk of neurodevelopment impairment. (See 'Oral feeds' above and 'Dextrose gel' above.)
Blood glucose concentrations should be measured frequently, starting 30 minutes after the initial feed or within 90 to 120 minutes after birth if the first feeding is delayed and then before subsequent feedings. (See 'Asymptomatic and mildly symptomatic infants' above.)
•Discharge criteria ‒ Prior to discharge, the infant should be able to maintain BG in normal range through cycles of feeding and fasting, as demonstrated by a preprandial BG >50 mg/dL (2.8 mmol/L) through three feed-fast cycles for infants <48 hours of age, and >60 mg/dL (3.3 mmol/L) in those who are ≥48 hours of age. (See 'Discharge criteria' above.)
●Preterm infants – Preterm infants born at ≤34 weeks gestation are at increased risk for hypoglycemia. The management for asymptomatic preterm infants who are able to receive sufficient nutrition enterally entails early feeds and monitoring of BG levels. For those who are not expected to be able to receive enough enteral nutrition due to prematurity, parenteral nutrition, which includes dextrose, should be started promptly. These issues are discussed separately. (See "Approach to enteral nutrition in the premature infant" and "Parenteral nutrition in premature infants".)
Acute episodes of symptomatic hypoglycemia in preterm neonates are generally managed with IV dextrose (0.2 g/kg given over 5 to 15 minutes [2 mL/kg of D10W]).
●Further diagnostic testing – Further testing may be warranted in newborns who have persistent hypoglycemia or other concerning findings. Consultation with a pediatric endocrinologist is advised in such cases. (See 'Evaluation of infants with persistent hypoglycemia' above.)
•Whom to test – Additional testing is generally warranted in newborns with any of the following findings (see 'Who should be tested?' above):
-Presentation with severe signs of hypoglycemia (lethargy, coma, or seizures)
-Requirement of prolonged and/or high rates of IV dextrose infusion for treatment, especially if there are no known risk factors for hypoglycemia
-Persistent hypoglycemia (ie, beyond what is expected based upon history/underlying risk factors)
-Family history of a genetic hypoglycemia disorder
-Physical exam features suggestive of a syndromic hypoglycemia disorder (eg, Beckwith-Wiedemann syndrome, congenital adrenal insufficiency)
•Timing of evaluation – Testing is generally deferred until at least 96 hours after birth because it is difficult to distinguish pathologic hypoglycemia disorders (transient or permanent) from normal physiologic transitional changes since the biochemical features (mild hyperinsulinism) are similar. (See 'Definition and timing of evaluation' above.)
•Laboratory evaluation ‒ Laboratory testing is based on "critical samples" obtained when BG is <50 mg/dL (2.8 mmol/L). Initial tests include plasma insulin, beta-hydroxybutyrate, cortisol, and growth hormone. Other initial tests to consider include bicarbonate, lactate, and free fatty acids. The results of these tests help categorize the type of disorder and determine if additional testing is warranted (algorithm 1). (See 'Laboratory evaluation' above.)
●Neurodevelopmental outcome ‒ Symptomatic neonatal hypoglycemia has been associated with brain damage, demonstrated on magnetic resonance imaging, and poorer developmental outcome. However, the available data do not clearly establish the glucose concentration or the duration of hypoglycemia that correlate with long-term neurologic sequelae. (See 'Neurodevelopmental outcome' above.)