FUKUSHIMA JOURNAL OF MEDICAL SCIENCE
Online ISSN : 2185-4610
Print ISSN : 0016-2590
ISSN-L : 0016-2590
Case Report
Serial investigation of continuous glucose monitoring in a very low birth weight infant with transient late-onset hyperglycemia
Toshihiko NakamuraDaisuke HatanakaMari NakamuraMichiko KusakariHidehiro TakahashiTakashi Kamohara
Author information
JOURNAL OPEN ACCESS FULL-TEXT HTML

2016 Volume 62 Issue 2 Pages 108-111

Details
Abstract

Transient late-onset hyperglycemia was detected in a very low birth weight (VLBW) infant (gestational age 28 weeks, birth weight 1,082 g) by routine point-of-care glucose monitoring. The infant had no clinical symptom. Serial continuous glucose monitoring (CGM) was conducted for 3 days at 31, 35, and 39 weeks’ post conceptual age. The difference values between the maximum and minimum blood glucose levels during the interval from one enteral feeding to the next enteral feeding were 32.3±14.3 mg/dL, 47.5±22.9 mg/dL, and 27.5±12.9 mg/dL for the 1st, 2nd, and 3rd CGM, respectively. The serial change in the values was statistically significant (p<0.01).

CGM is widely used as a routine clinical practice, which is true even in VLBW infants. Hyperglycemic events detected by only once of CGM in otherwise healthy preterm infants have already been reported on larger numbers of patients. To our knowledge, this is the first report that the change of glucose intolerance in a VLBW infant with transient late-onset hyperglycemia was investigated by serial CGM.

Introduction

Continuous glucose monitoring (CGM) is used safely and reliably in neonates, including very low birth weight (VLBW) infants, in whom it detects many more episodes of both low and high glucose concentrations than intermittent blood glucose measurements1). Most previous studies have been conducted in newborns with hypoglycemia, and a few studies have reported blood glucose variability in VLBW infants with transient neonatal hyperglycemia yet2-4). We performed serial CGM in a VLBW infant with late-onset hyperglycemia detected by routine screening.

Case

A 1,082 g (+0.08 SD) female infant was born at 28 weeks of gestation by normal vaginal delivery. Her clinical course was unremarkable, and total parenteral nutrition was stared on day 1 of life. The maximum intravenous glucose infusion rate was 7.3 mg/kg/min on day 3. She received fortified mother’s milk (HMS-1®, Morinaga Milk Ind. Tokyo, Japan) via gastric tube every 2 hour and no intravenous glucose on day 17 (31 weeks’ post conceptual age: PCA). After her admission, her blood glucose level was checked at least twice a day by point-of-care testing (Radiometer ABL835). As the treatment for apnea of prematurity, aminophylline, which may be potentially affecting blood glucose levels, was administered from day 0 to day 74 of life. Other drugs such as steroids used for prevention of chronic lung disease were not given.

On day 20 of life, a high fasting blood glucose level of 206 mg/dL was measured. Since then, her blood glucose level remained high. However, there was no dehydration or body weight loss, and laboratory tests revealed negative findings for urine ketones and occasional positive finding of glucosuria. Except for hyperglycemia, there were no signs of stress such as infection or intracranial hemorrhage. The patient met the diagnostic definition of late-onset neonatal hyperglycemia rather than neonatal diabetes mellitus (Table 1.5)). Thus, we conducted the 1st CGM starting on day 24 (31 weeks’ PCA) for three continuous days. The 2nd and 3rd CGM episodes subsequently performed from 35 and 39 weeks’ PCA, respectively, showed relatively stable glycemic patterns. We used fortified mother’s milk or preterm infant formula that has a higher concentration of protein and carbohydrate. During the 1st CGM period, full enteral nutrition was provided at 120-130 ml/kg/day by slow injection for about 1 hour through a syringe every 2 hours. On the first day of the 2nd CGM period, feeding interval was changed 2 to 3 hourly. During the 3rd CGM period, she was fed on demand with 20-25 ml/kg/feeding approximately every 3 to 4 hourly. Her blood glucose level was in the normal range from 35 weeks’ PCA, and she had no hyperglycemic episodes. The hospital course of her fasting blood glucose levels and body weights is illustrated in Fig. 1. During follow-up at 3, 6, 9, 12, and 18 months of age, she presented with normal growth and development. From this clinical course, we have had the final diagnosis as transient late-onset neonatal hyperglycemia.

Table 1.

Criteria for neonatal transient late-onset hyperglycemia

Fig. 1.

The hospital course of her blood glucose levels and body weights.

The black line represents serial body weight and the bar represents the fasting blood glucose level.

Interestingly, regardless of the fasting glucose level, the body weight gain is almost constant during the hospital course.

Materials and Methods

The Japanese Red Cross Musashino Hospital ethics committee approved this study, and informed consent was obtained from the child’s parents.

The CGM system (iPro®2 Professional CGM; Medtronic MiniMed, Northridge, CA, USA) uses a disposable glucose oxidase-based platinum electrode sensor called the Sofsensor®, which was manually inserted into the subcutaneous tissue in the patient’s lateral thigh. The sensor catalyzes interstitial glucose, thereby generating an electrical current that is recorded on a monitor every 5 min as an averaged glucose value. To calibrate this monitor, capillary blood samples were intermittently analyzed with a point-of-care hand-held glucometer (the Nova StatStrip; Nova Biomedical, Waltham, MA, USA) at least four times daily6,7). For accuracy of CGM data for high glucose levels, calibration was performed at postprandial timing several times during each CGM. The patient underwent each CGM for 3 days in our hospital. There was no evidence of local edema, inflammation, or infection at the sensor site in the baby. The CGM system was well tolerated. The baby did not make any attention to put at ether the sensor or the lead.

Statistical analyses were performed using one-way analysis of variance (ANOVA) to compare the difference in the measurements between groups. We performed statistical analysis with SPSS 11.5 software (SPSS Inc., Cary, NC). Differences were considered significant at p<0.05.

Results (Fig. 2)

The 1st CGM results showed almost the same pattern of daily glycemic change over the 3 days. Even the lowest blood glucose revel exceeds 150 mg/dL on the 1st day during initial CGM.

We calculated the differences between the maximum and minimum blood glucose levels during the interval from one enteral feeding to the next enteral feeding and named it “glucose variability between feedings”7). The respective values were 32.33±14.3 mg/dL, 47.5±22.9 mg/dL, and 27.5±12.9 mg/dL for the first, second, and third CGM periods. The serial change in the values was statistically significant (p<0.01).

Fig. 2.

Continuous glucose monitoring (CGM) data over 3 days in this patient.

Each panel has time (representing a 24-hour period) on the horizontal axis and glucose level (mg/dL) recorded on the vertical axis. The dotted line indicates the serial change of the average value of the data for the 3 days. (1) The patient at 31 weeks’ postconceptional age (PCA) undergoing tube feeding every 2 h (>150 mg/dL: 89%, mean 172 mg/dL). (2) The patient at 35 weeks’ PCA undergoing tube feeding from every 2 to 3 h on the first day of CGM (>150 mg/dL: 10%, mean 117 mg/dL). (3) The patient at 39 weeks’ PCA undergoing oral self-feeding every 3 to 4h (>150 mg/dL: 0%, mean 95 mg/dL).

Discussion

Hyperglycemia in VLBW infants has been studied mainly in the first week of life using CGM2,3). To date, there are few reports of CGM in stable preterm infants4,6-8). These reports were considerations of the cases by which only several % will be hyperglycemic revels throughout the day. But in this case, even the lowest blood glucose revel exceeds 150 mg/dL on the 1st day during initial CGM. Therefore a differential diagnosis with neonatal diabetes was required. The present study uniquely using repeated CGM results to demonstrate the clinical course of late-onset hyperglycemia in a stable VLBW infant. The etiology of late-onset neonatal hyperglycemia is while being still unclear5).

The first CGM results showed almost the same pattern of daily glycemic change over the 3 days. Based on the fact that the blood glucose level fluctuates regularly in accordance with the time of feeding, the following factors may be implicated in hyperglycemia: 1) inadequate endogenous insulin secretion in response to blood glucose level and 2) lack of sufficient insulin-dependent tissues (fat and muscle)9). The second CGM period showed that glucose variability between feedings was higher than that of the first CGM period. On the first day of this CGM period, the interval of tube feeding was changed from every 2 hours to 3 hours, which might have revealed remarkable glycemic variability but showed only a 10% incidence of hyperglycemia. The last CGM period during self-feeding (maximum 200 ml/kg/day, orally every 3 to 4 hourly) revealed the glucose variability between feedings was lower than that of the second CGM period. The patient had no episodes of abnormal glucose level.

The maturation of glucose tolerance in VLBW infants with late-onset transient neonatal hyperglycemia as well as transient neonatal hyperinsulinemic hypoglycemia10) may be able to be confirmed based on the range of fluctuation of serial CGM measurements. If hyperglycemia is detected in a stable, fully enteral-fed VLBW infant with point-of-care glucose monitoring, CGM is a safe and great clinical diagnostic tool which avoidance of extremely high glucose level.

References
  • 1.   Beardsall K. Measurement of glucose levels in the newborn. Early Hum Dev, 86: 263-67, 2010.
  • 2.   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, 90: F307-F310, 2005.
  • 3.   Beardsall K, Vanhaesebrouck S, Ogilvy-Stuart AL, et al. Prevalence and determinants of hyperglycemia in very low birth weight infants: Cohort Analyses off the NIRTURE Study. J Pediatr, 157: 715-719, 2010.
  • 4.   Pertierra-Coetada A, Ramom-Krauel M, Iriondo-Sanz M, Iglesias-Platas I. Instability of glucose values in very preterm babies at term postmenstrual age. J Pediatr, 165: 1146-1153, 2014.
  • 5.   Watanabe Y, Nishida H, Oishi M, et al. Transient hyperglycemia in extremely immature infants-clinical findings and measurement of serum C-peptide level- (in Japanese). J Jpn Neon Med, 27: 714-720, 1991.
  • 6.   Mola-Schenzle E, Staffler A, Klemme M, et al. Clinical stable very low birthweight infants are at risk for recurrent tissue glucose fluctuations even after fully established enteral nutrition. Arch Dis Child Fetal Neonatal Ed, 100: F126-F131, 2015.
  • 7.   Nakamura T, Hatanaka D, Yoshioka T. A study of the usefulness of continuous glucose monitoring (CGM) in preterm infants at NICU. (in Japanese) J Jpn Soc Premature Newborn Med, 27: 59-64, 2015.
  • 8.   Mizumoto H, Kawai M, Yamashita S, Hata D. Intraday glucose fluctuation is common in preterm infants receiving intermittent tube feeding. Pediatr Int 2, Oct 21. doi: 10. 1111/ped. 12838, 2015.
  • 9.   Sabzehei MK, Afjeh SA, Shakiba M, Alizadeh P, Shamshiri AR, Esmaili F. Hyperglycemia in VLBW infants; incidence, risk factors and outcome. Arch Iran Med, 17: 429-34, 2014.
  • 10.   Nakamura T, Nomura T, Hatanaka D, Kusakari M, Takahashi H, Kamohara T. A study of the utility of continuous glucose monitoring (CGM) in the NICU: the 2nd report; in neonatal transient hyperinsulinemic hypoglycemia. J Jpn Scoi Neonat Health and Dev, 28: 57-62, 2016.
 
© 2016 The Fukushima Society of Medical Science

This article is licensed under a Creative Commons [Attribution-NonCommercial-ShareAlike 4.0 International] license.
https://creativecommons.org/licenses/by-nc-sa/4.0/
feedback
Top