HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Leipold, H. Articles by Worda, C. Search for Related Content PubMed PubMed Citation Articles by Leipold, H. Articles by Worda, C. Related Collections Diabetes

Fetal Hyperinsulinism and Maternal One-Hour Postload Plasma Glucose Level

Heinz Leipold, MD^*, Alexandra Kautzky-Willer, MD, Aral Özbal, MD^*, Dagmar Bancher-Todesca, MD^* and Christof Worda, MD^*

From the Departments of *Obstetrics and Gynecology and Internal Medicine III, University of Vienna Medical School, Vienna, Austria.

Address reprint requests to: Christof Worda, MD, Department of Obstetrics and Gynecology, University of Vienna Medical School, Waehringer Guertel 18–20, A-1090 Vienna, Austria; e-mail: christof.worda{at}akh-wien.ac.at.

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
OBJECTIVE: Fetal insulin concentrations reflect the intrauterine glucose load given the fetus by the mother. In this study, we assessed the association between maternal glucose levels during oral glucose tolerance testing and fetal cord insulin.

METHODS: Pregnant women with an oral glucose tolerance test (OGTT) result were included in this prospective study. The patients were divided into 3 groups according to their 1-hour OGTT glucose concentration: up to 160 mg/dL (control, group I), 160–179 mg/dL (intermediate, group II), and gestational diabetes mellitus (GDM, group III). Patients with GDM were assigned to insulin therapy if blood glucose levels were not in the preferable range.

RESULTS: Of the 930 patients who entered the study, 570 (61.3%) were assigned to group I, 76 (8.2%) to group II, and 284 (30.5%) to group III. The cord blood insulin value was significantly (P < .001, Mann–Whitney test) higher in group II (median, 12.8 µU/mL; range, 3–130 µU/mL) than in group I (median, 7.25 µU/mL; range, < 3–98 µU/mL). Cord blood insulin values were higher, albeit not significantly (P = .100, Mann–Whitney test), in group II than in group III (median, 9.9 µU/mL; range, < 3–61 µU/mL).

CONCLUSION: Children whose mothers had a 1-hour value between 160 and 179 mg/dL had significantly higher cord blood insulin values than offspring of women with a 1-hour value below 160 mg/dL.

LEVEL OF EVIDENCE: II-2

As a result of the postprandial hyperglycemia of the mother and fetus, a reactive increase in fetal insulin concentration is observed (Pedersen's hypothesis) because maternal glucose can freely pass the placental barrier whereas the larger-sized maternal insulin cannot.⁴ Fetal hyperinsulinism is the key symptom of diabetic fetopathia, and in animal studies has led to fetal death.⁵ Newborns with increased cord insulin levels demonstrate a higher frequency of poorer postpartum adaptation, hyperbilirubinemia, hypoglycemia, immature birth, and respiratory distress syndrome.⁶

Gestational diabetes mellitus is diagnosed by performance of an oral glucose tolerance test (OGTT), but there is no consensus on the diagnostic glucose concentration and lack of a threshold value both for the development of fetal hyperinsulinism and for an association with fetal morbidity and mortality. Two frequently used diagnostic criteria for GDM in Europe are the modified Carpenter-Coustan⁷ criteria, and criteria proposed by Weiss et al, which differ in regard to the 1-hour value and the omission of the 2-hour measurement.⁸

A close correlation has been shown between increasing neonatal morbidity and increasing cord blood insulin values.^6,9 It is well known that macrosomia at birth relates to elevated fetal insulin levels.^10–12 Metzger et al¹³ observed that offspring of diabetic mothers had a higher rate of childhood obesity when third-trimester amniotic fluid levels were increased. However, until now little is known about the influence of different glucose threshold levels of the OGTT and subsequent treatment and cord blood insulin levels at time of birth. This study aimed to elucidate the association between maternal glucose levels during oral glucose tolerance testing and fetal cord insulin.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
This cohort study was conducted from June 2001 to January 2003 at the Department of Obstetrics and Gynecology of the University of Vienna Medical School, a referral site for nearly all high-risk pregnancies from eastern Austria, and at the Division of Endocrinology of the Department of Internal Medicine. At initial contact, which occurs between 11 and 13 weeks of gestation, a risk assessment is performed and each patient is scheduled for an OGTT. Women fulfilling the following criteria were invited to participate in the study: 1) no maternal chronic or infectious diseases, 2) no multiple pregnancy, and 3) no preexisting diabetes mellitus. This study was conducted with the approval of our institutional review board, and informed consent was obtained from all participants. In women who were included in the study, an OGTT is performed using a standardized 75-g glucose solution (Glucodrink, Unipack GmbH, Wiener Neustadt, Austria) after a minimum fasting period of 8 hours, with blood samples being taken in the fasting state and after 1 and 2 hours. During the test, the patient is required to sit and to abstain from smoking. For OGTT evaluation, the guidelines of the German Society for Diabetes were implemented.¹⁴ According to these guidelines, the upper normal limit for the fasting serum glucose is 95 mg/dL; 180 mg/dL and 155 mg/dL represent the 1-hour and 2-hour values after glucose ingestion, respectively. Although the 75-g oral glucose load after an 8-hour fast is somewhat different than that used for the test most frequently performed in the United States, the physiologic principles shown in our study should also apply to different methods of glucose tolerance testing. If at least 1 value was exceeded, the woman was admitted to the gestational diabetes program.

The OGTT is performed if there are certain risk factors (GDM, intrauterine fetal death, birth of child weighing 4,000 g, glucosuria, body mass index [BMI] > 27 kg/m² in the first trimester), and is repeated at 24–28 weeks of gestation in the case of a physiologic OGTT test result. We excluded every woman with a pathological OGTT result before the 24th week of gestation to avoid inclusion of women with pregestational diabetes. In women with no apparent history or risk, the OGTT is exclusively performed between 24 and 28 weeks of gestation. The women were divided into 3 groups according to their OGTT result: group I: normal fasting glucose concentration, 1-hour glucose concentration up to 160 mg/dL, and normal 2-hour glucose concentration (control group); group II: normal fasting glucose concentration, 1-hour glucose concentration between 160 and 179 mg/dL, and normal 2-hour glucose concentration (intermediate group); and group III: GDM group. Groups I and II were classified as physiologic (according to German Society for Diabetes criteria); therefore, they did not receive any therapy. The cutoff value of 160 mg/dL between groups I and II was chosen in accordance with the findings of Weiss et al.¹⁵

Every patient in the study with an abnormal OGTT result (group III) was referred to dietary counseling and was advised to measure her blood glucose levels 4 times daily. The cutoff levels for blood glucose self-assessment are 90 mg/dL at fasting and 130 mg/dL 1 hour postprandially, according to the recommendation of the Austrian Diabetes Association and the Austrian Diabetes and Pregnancy Study Group.¹⁶ If a patient has 5 excess values per week in the blood glucose self assessment, insulin therapy is started. Every insulin-requiring patient with an acceptable metabolic function and normal fetal biometric data was examined at biweekly, and, if necessary, weekly intervals. Acceptable metabolic function was defined as having less than 5 excess values per week in the blood glucose self-assessment (capillary blood glucose concentrations were measured daily at home, once before breakfast and then 1 hour after breakfast, lunch, and dinner) and the fetal abdomen below the 95% percentile.¹⁷

Pregnant women with normal glucose tolerance were scheduled to return only if there were complications, or if they reported back to the clinic independently for pregnancy-related problems. All pregnant women received examinations in short time intervals by a gynecologist outside the hospital throughout their pregnancy, which are paid for by the Austrian Government (Mother-Child-Pass). Every patient was assigned to a cardiotocographic examination at term and was induced no later than 10 days after the term. Term is defined as the day 40 weeks after the last menstrual period and corrected if necessary by first-trimester ultrasonography.

To reduce the effects of potential confounders on the cord blood insulin, women with normal glucose tolerance did not receive intravenous hydration, except for those with prolonged labor and epidural anesthesia (group I: 86 women [15.1%], group II: 13 women [17.1%]; P = .61, ² test). These women received 1,000 mL Ringer's solution once. Women with GDM and women receiving cesarean delivery received a regimen of intravenous hydration. Glucose infusion was not given to any patient during birth.

Immediately after delivery, cord blood was obtained from the umbilical vein in all patients enrolled in the study. Insulin was determined immediately after sampling using a radioimmunoassay (Adaltis Italia S.p.A.; Casalecchio di Reno, Italy) and levels below the detection limit of this test (<1 µU/mL) were set to zero. Glucose was determined using an enzymatic in vitro test (Gluco-quant; Glucose/HK, Roche Diagnostics GmbH, Mannheim, Germany) on a Roche/Hitachi analyzer. The assay for insulin and glucose had intraassay precisions of less than 7.4% and less than 1.1%, respectively. The interassay precisions was less than 8% and less than 1.9%, respectively. The assay for insulin showed a cross-reaction of less than 14% with proinsulin. No other significant cross reactions for the insulin or glucose assays were reported. Hyperinsulinemia was defined as a cord blood insulin level exceeding 19 µU/mL.⁸

Statistical data were evaluated using SPSS 10.0 (SPSS Inc., Chicago, IL). The main outcome measure was the cord blood insulin concentration in the 3 groups. Minor outcome measures were fetal weight, birthweight above the 95% percentile, and cord blood glucose. Because the Kolmogorov-Smirnov test gave significant results for insulin and partus and the graphic distribution of the tested variables showed no normal distribution, nonparametric tests were used. Values are given as the median plus minimum and maximum. The Mann–Whitney test and the repeated Mann–Whitney test with Shaffer correction were applied for comparison between the independent groups.¹⁸The ² value was used for the comparison of proportions. Correlation analysis (Spearman rank) was used to identify correlations in the parameters tested. Potential influences of the 3 "groups" (according to their 1-hour OGTT glucose concentration), mode of delivery, birthweight, maternal age, BMI, gestational age at delivery, and arterial cord blood pH were assessed by a univariate analysis of variance. For this purpose, umbilical cord insulin was linearized by a log-transformation. Subgroups of the influencing factor "group" were compared with differences in umbilical cord insulin by the polynomial contrast method. The analysis of variance showed significant quadratic contrasts for the factor "group." The contrast estimations of log-umbilical cord insulin supplied P values of .008 for "group." No interactions were observed between the significant factors. All tests were 2-tailed and a statistical value of P < .05 was considered to be significant.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Overall 976 women were included in the study; 46 women were excluded for the following reasons: delivery in another hospital (n = 25), hemolytic cord blood (n = 7), or nonattendance of the recommended controls (n = 14). Among the 930 pregnant women who were included in the study, 570 women (61.3%) were assigned to group I), 76 women (8.2%) to group II, and 284 women (30.5%) to group (III). All women in group III, except for the 28 women who were treated with diet only, needed human insulin until birth. Patient characteristics are shown in Table 1. The high rate of women with GDM may also be caused by the fact that our clinic is a tertiary care center.

Group II had significantly (P < .001, Mann–Whitney test) higher cord blood insulin levels (median, 12.8 µU/mL; range 3–130 µU/mL) than group I (median, 7.25 µU/mL; range, 0–98 µU/mL). Cord blood insulin levels were higher in group II than in group III (median, 9.9 µU/mL; range, 0–61 µU/mL), although this difference was not significant (P = .10, Mann–Whitney test) (Fig. 1). Differences in fetal birthweight above the 95% percentile, but not in cord blood glucose values and birthweight, were significant in group I (Table 2). Cord blood insulin levels were significantly (P = .008, Mann–Whitney test) higher in newborns from mothers with normal glucose tolerance with birthweight above the 95% percentile (median, 9.4 µU/mL; range 3–130 µU/mL) than in newborns below the 95% percentile (median, 7.45 µU/mL; range, 0 –35 µU/mL). Group II had significantly more newborns above the 95% weight percentile than did group I (59 [10.3%] versus 16 [21.6%]; P = .046 by ² test).

View larger version (21K): [in this window] [in a new window]   Fig. 1. Box-and-whisker plots show cord blood insulin with the 3 different groups according to the 1-hour value of the oral glucose tolerance test (OGTT). Solid bar indicates median; upper and lower limits of box, 75th and 25th percentiles; upper and lower bars, maximum and minimum values; outliers are not shown. Group I, normal OGTT result with 1-hour value below 160 mg/dL; group II, normal OGTT result with 1-hour value between 160 and 179 mg/dL; group III, pathologic OGTT result with 1-hour value above 180 mg/dL. Leipold. Fetal Hyperinsulinism and GDM. Obstet Gynecol 2004.

There was a significant correlation (Spearman) between maternal BMI and number of pregnancies (0.231; P < .001), glucose concentration during oral glucose tolerance testing (0.14; P < .002), and cord blood glucose concentration (–0.201; P < .001). Cord blood glucose was inversely correlated to cord blood pH (–0.258; P < .001) but was not significantly correlated with cord blood insulin (0.002; P = .96).

Umbilical cord insulin in the subgroups of "group" and mode of delivery, adjusted by the covariates mode of delivery, maternal age, birthweight, BMI, gestational age at delivery, and arterial cord blood pH, revealed a significant increase from group I to group II and a nonsignificant decrease from group II to group III in the univariate analysis of variance (P = .005 and P = .176, respectively). Newborns delivered by cesarean had significantly higher umbilical cord insulin values than did infants delivered vaginally (P < .001).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Because of the constant supply of excess glucose to the fetus by mothers with impaired glucose metabolism, a reactive increase in insulin release from the ß cells of the fetal pancreas is observed. Between 12 weeks of gestation and the first 5 months of the newborn's life, a 130-fold increase in ß-cell mass and a special sensitivity to unphysiologic stress situations can be observed.¹⁹ It has been shown that cord blood insulin concentrations correlate with diabetes-associated neonatal complications.⁶

There is evidence that the 1-hour OGTT level seems to be most important for the diagnosis of GDM because it correlates best with birthweight and is the most reproducible value with better specificity and sensitivity regarding fetal complications.^15,20–22 This could also be the reason for the rather low 1-hour cutoff value (160 mg/dL) proposed by Weiss et al⁸ in contrast to other established criteria.

We have shown that newborns of women with 1-hour OGTT values between 160 and 179 mg/dL had significantly elevated cord blood insulin levels, were more likely to develop intrapartum hyperinsulinism, and were more likely to be macrosomic than newborns of women with 1-hour OGTT results below 160 mg/dL. Furthermore, newborns above the 95% birthweight percentile had significantly higher cord blood insulin concentrations than newborns below the 95% birthweight percentile, which is in accordance with a previous report.¹⁰ Interestingly, the cord blood insulin levels of the newborns of women with well-treated GDM were even lower than those of the newborns f women with a 1-hour value between 160 and 179 mg/dL (group II). This supports the hypothesis that maternal 1-hour glucose values above 160 mg/dL are a marker of increased postprandial glucose load to the fetus stimulating fetal insulin production.

Limitations of this study include the use of different treatment regimens and the possible influence of delivery on fetal insulin values. For ethical reasons, it is not possible to deny treatment to women with GDM to get a more homogenous sample for comparison with the two other groups. Although comparison of fetal insulin levels before birth obtained with late gestational amniocentesis might reduce the possible but improbable effect of birth on significant alteration of fetal insulin levels, it is difficult to justify an invasive procedure in the third trimester only for study reasons.

On a worldwide basis, there appears to be a consistent increase in the number of patients suffering from diabetes mellitus.²³ The genetic disposition of type 1 and non–insulin-dependent diabetes mellitus seems to be one of the causes of this phenomenon. The development of diabetes at a later stage in life may be influenced by the intrauterine glucose burden of the fetus in adoption to genetic heritage.

It was shown that glucotoxicity leads to reduced or insufficient irritability of ß cells in the pancreas.²⁴ The intrauterine damage to the pancreas may be responsible for the 20% incidence of impaired glucose tolerance in early childhood among infants of diabetic mothers.²⁵ Detection of all women with even subtle deterioration of glucose tolerance during pregnancy is the main prerequisite for a successful treatment. Because elevated fetal cord blood insulin values result from excessive intrauterine glucose loads, cord blood insulin reflects the intrauterine metabolic situation of the fetus and is therefore a good marker to establish threshold values for the OGTT.

Our study shows that the glucose load of the fetus is increased if the maternal 1-hour plasma glucose concentration exceeds 160 mg/dL. The utility of potential management strategies such as fetal growth monitoring or maternal dietary or pharmacologic interventions on these women are unknown.

	Footnotes

 
Received March 25, 2004. Received in revised form June 22, 2004. Accepted July 15, 2004.

doi:10.1097/01.AOG.0000142716.00040.bb

	REFERENCES

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Shoham I, Wiznitzer A, Silberstein T, Fraser D, Holcberg G, Katz M, et al. Gestational diabetes complicated by hydramnios was not associated with increased risk of perinatal morbidity and mortality. Eur J Obstet Gynecol Reprod Biol 2001;100:46–9.[Medline]

2. Cordero L, Treuer SH, Landon MB, Gabbe SG. Management of infants of diabetic mothers. Arch Pediatr Adolesc Med 1998;152:249–54.[Abstract/Free Full Text]

3. Vambergue A, Nuttens MC, Goeusse P, Biausque S, Lepeut M, Fontaine P. Pregnancy induced hypertension in women with gestational carbohydrate intolerance: the diagest study. Eur J Obstet Gynecol Reprod Biol 2002;102:31–5.[Medline]

4. Pedersen J. The pregnant diabetic and her newborn. 2nd ed. Baltimore (MD): Williams & Wilkins, 1977.

5. Philipps AF, Dubin JW, Matty PJ, Raye JR. Arterial hypoxemia and hyperinsulinemia in the chronically hyperglycemic lamb. Pediatr Res 1982;16:653–8.[Medline]

6. Weiss PA, Kainer F, Haas J. Cord blood insulin to assess the quality of treatment in diabetic pregnancies. Early Hum Dev 1998;51:187–95.[Medline]

7. Carpenter MW, Coustan DR. Criteria for screening tests for gestational diabetes. Am J Obstet Gynecol 1982;144:768–73.[Medline]

8. Weiss PA, Haeusler M, Tamussino K, Haas J. Can glucose tolerance test predict fetal hyperinsulinism? Br J Obstet Gynaecol 2000;107:1480–5.

9. Mehta A, Hussain K. Transient hyperinsulinism associated with macrosomia, hypertrophic obstructive cardiomyopathy, hepatomegaly, and nephromegaly. Arch Dis Child 2003;88:822–4.[Free Full Text]

10. Hoegsberg B, Gruppuso PA, Coustan DR. Hyperinsulinemia in macrosomic infants of nondiabetic mothers. Diabetes Care 1993;16:32–6.[Abstract]

11. Fallucca F, Gargiulo P, Pachi A. Amniotic fluid insulin, C peptide concentrations, and fetal morbidity in infants of diabetic mothers. Am J Obstet Gynecol 1985;153:534–40.[Medline]

12. Schaefer UM, Dupak J, Vogel M, Heinze T, Kjos SL, Buchanan TA, et al. Hyperinsulinism, neonatal adiposity and placental immaturity in infants born to women with one abnormal glucose tolerance test value. J Perinatal Med 1998;26:27–36.[Medline]

13. Metzger BE, Freinkel N. Amniotic fluid insulin as a predictor of obesity. Arch Dis Child 1990;65:1050.[Abstract]

14. Deutsche Diabetes-Gesellschaft. Diagnostik und Therapie des Gestationsdiabetes. Richtlinien der Deutschen Diabetes-Gesellschaft [Diagnostic and therapy of GDM. Guidelines from the German Diabetes Association]. Der Frauenarzt 1993;13–4.

15. Weiss PAM. Diabetes in pregnancy: lessons from the fetus. In: Dornhorst A, Hadden DR, eds. Diabetes and pregnancy: an international approach to diagnosis and management. New York (NY): John Wiley & Sons; 1996. p. 21–40.

16. Kautzky-Willer A, Bancher-Todesca D. Gestationsdiabetes. Wien Med Wochenschr 2003;153:478–84.[Medline]

17. Hadlock FP, Deter RL, Harrist RB, Park SK. Estimated fetal age: computer-assisted analysis of multiple fetal growth parameters. Radiology 1984;152:497–501.[Abstract/Free Full Text]

18. Shaffer JP. Modified sequentially reactive multiple procedures. J Am Stat Assoc 1986;81:826–31.

19. Hellerstrom C, Swenne I, Anderson A. Islet cell replication and diabetes. In: Lefebvre PJ, Pipeleers DG, eds. The pathology of the endocrine pancreas in diabetes. Heidelberg: Springer, 1988:141–71.

20. Jovanovic-Peterson L, Peterson CM, Reed GF, Metzger BE, Mills JL, Knopp RH, et al. Maternal postprandial glucose levels and infant birth weight: the Diabetes in Early Pregnancy Study. The National Institute of Child Health and Human Development–Diabetes in Early Pregnancy Study. Am J Obstet Gynecol 1991;164:103–11.[Medline]

21. Harlass FE, Brady K, Read JA. Reproducibility of the oral glucose tolerance test in pregnancy. Am J Obstet Gynecol 1991;164:564–8.[Medline]

22. Benjamin F, Wilson SJ, Deutsch S, Seltzer VL, Droesch K, Droesch J. Effect of advancing pregnancy on the glucose tolerance test and on the 50-g oral glucose load screening test for gestational diabetes. Obstet Gynecol 1986;68:362–5.[Medline]

23. King H, Aubert RE, Herman WH. Global burden of diabetes, 1995-2025: prevalence, numerical estimates, and projections. Diabetes Care 1998;21:1414–31.[Abstract]

24. Rasmussen H, Zawalich KC, Ganesan S, Calle R, Zawalich WS. Physiology and pathophysiology of insulin secretion. Diabetes Care 1990;13:655–66.[Abstract]

25. Plagemann A, Harder T, Kohlhoff R, Rohde W, Dorner G. Glucose tolerance and insulin secretion in children of mothers with pregestational IDDM or gestational diabetes. Diabetologia 1997;40:1094–100.[Medline]

This article has been cited by other articles:

				R. Retnakaran, Y. Qi, M. Sermer, P. W. Connelly, B. Zinman, and A. J.G. Hanley Isolated Hyperglycemia at 1 Hour on Oral Glucose Tolerance Test in Pregnancy Resembles Gestational Diabetes Mellitus in Predicting Postpartum Metabolic Dysfunction Diabetes Care, July 1, 2008; 31(7): 1275 - 1281. [Abstract] [Full Text] [PDF]

				R. Retnakaran, B. Zinman, P. W. Connelly, M. Sermer, and A. J.G. Hanley Impaired Glucose Tolerance of Pregnancy Is a Heterogeneous Metabolic Disorder as Defined by the Glycemic Response to the Oral Glucose Tolerance Test Diabetes Care, January 1, 2006; 29(1): 57 - 62. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Leipold, H. Articles by Worda, C. Search for Related Content PubMed PubMed Citation Articles by Leipold, H. Articles by Worda, C. Related Collections Diabetes