Long-term effects of diabetes during pregnancy on the offspring

# 2009 The Authors Journal compilation # 2009 Blackwell Munksgaard Pediatric Diabetes 2009: 10: 432–440 doi: 10.1111/j.1399-5448.2009.00507.x All rights reserved Pediatric Diabetes Original Article Long-term effects of diabetes during pregnancy on the offspring Wroblewska-Seniuk K, Wender-Ozegowska E, Szczapa J. Long-term effects of diabetes during pregnancy on the offspring. Pediatric Diabetes 2009: 10: 432–440. Background: Many epidemiological and experimental studies have proven that some adult diseases might have their origin in fetal life. It has been also hypothesized that intra-uterine environment in pregnancy complicated with diabetes might influence the development of obesity, type 2 diabetes, and cardiovascular diseases in the offspring. Objectives: To assess glucose metabolism, insulin secretion, and prevalence of obesity in the offspring of mothers with pregestational diabetes mellitus (PGDM) and gestational diabetes mellitus (GDM) and to evaluate the relationship between maternal metabolic control during pregnancy and metabolic disturbances in children. Subjects: Children of mothers with PGDM (n ¼ 43) and GDM (n ¼ 34) were examined at 4–9 yr of age and compared with the control group (n ¼ 108; metabolic parameters available for n ¼ 29). Methods: The incidence of overweight and obesity, impaired glucose tolerance, and insulin resistance were analyzed based on anthropometric and biochemical measurements. Statistical analysis was performed with STATISTICA package. Results: In children of GDM mothers, body mass index z-score (0.81
1.01 vs. 20.04
1.42 PGDM vs. 0.07
1.28 control group) and insulin resistance indices (homeostasis model assessment index – insulin resistance 1.112 vs. 0.943 PGDM vs. 0.749 control group) were significantly higher than in other groups. Obesity and insulin resistance were also most frequent in GDM group [not significant (NS)]. In addition, we observed the relationship between maternal hemoglobin A1c and mean glycemia in perinatal period and insulin resistance in children. There was not such correlation for the class of maternal diabetes. Conclusion: Children born to mothers with gestational diabetes seem to be at risk for obesity and metabolic disturbances. Type 1 diabetes mellitus is being diagnosed in 0.3% of women at the reproductive stage of their life (1). Additionally, in 3–6% of pregnancies, we observe the condition of glucose intolerance known as gestational diabetes mellitus (GDM) (2). Metabolic disturbances in the course of diabetes during pregnancy are known to affect the health of both the mother and her offspring. Diabetes in pregnancy is associated with increased rates of congenital anomalies, spontaneous abortions, perinatal complications, and neonatal deaths (3–6). Good metabolic control of pregestational diabetes mellitus (PGDM) at conception 432 Katarzyna WroblewskaSeniuka, Ewa Wender-Ozegowskab and Jerzy Szczapaa a Department of Neonatal Infectious Diseases, Chair of Neonatology, Poznan University of Medical Sciences, Poznan, Poland; and bDepartment of Obstetrics and Women’s Diseases, Poznan University of Medical Sciences, Poznan, Poland Key words: GDM – glucose intolerance – insulin resistance – obesity – PGDM Corresponding author: Katarzyna Wróblewska-Seniuk _ Klinika Zakazeń Noworodków ul. Polna 33 60-565 Poznań Poland. Tel: 1 48 607 393 463; fax: 1 48 618 419 650; e-mail: [email protected] Submitted 21 July 2008. Accepted for publication 3 February 2009 and during pregnancy seems to reduce malformation rates and clearly improves child survival (4, 5, 7–10). Most important for the normal pregnancy outcome and fetal development is to maintain the fasting glucose level of 60–90 mg/dL and the glycated hemoglobin level between 3.8 and 6.3% and to avoid the episodes of hypoglycemia and postprandial hyperglycemia (11). Maternal hyperglycemia in the second and third trimester of pregnancy may cause hyperplasia and hyperactivity of the pancreatic beta cells, leading to the insulin overproduction and, as a consequence, to the excessive fetal growth, pulmonary immaturity, Effects of diabetes on the offspring and metabolic disturbances in the neonatal period, such as hypoglycemia, hypocalcemia, hypomagnesemia, and hyperbilirubinemia (12–14). While most authors focus on altered glucose homeostasis in neonatal period, it remains unclear how maternal diabetes may influence later life of the offspring. Based on the clinical observations, in 1954, Hoet and Lukens suggested that prenatal hyperinsulinemia in fetuses of diabetic pregnancies might lead to the development of diabetes in adulthood (15). Some clinical investigations during childhood indicated elevated frequency of impaired glucose tolerance (IGT) in the offspring of mothers with diabetes during pregnancy (16). Some authors also reported alterations of insulin secretion, such as hyperinsulinemia, which is known to play a key role in the development of metabolic and cardiovascular disturbances in adults (16–19). It has been shown that the prevalence of obesity and type 2 diabetes is higher in the offspring of diabetic mothers than in children born to healthy mothers (20). The risk of type 2 diabetes in these children is around 4% at 5–9 yr and 12% at 10–16 yr, and the risk of IGT is 5.4% at 5–9 yr and 19.3% at 10–16 yr (21). Experimental studies in rats demonstrated long-term alterations of glucose tolerance and insulin secretion because of the induction of maternal gestational hyperglycemia, leading to fetal and neonatal hyperinsulinism (21). Although both the pathogenesis and the outcome of PGDM and GDM are different, there are only very few studies comparing glucose metabolism and insulin secretion in infants of mothers suffering from these conditions (22, 23). The aims of this study were to assess the features of glucose metabolism and insulin secretion as well as the prevalence of obesity in the offspring of diabetic mothers and to compare children born to mothers with PGDM and GDM. We also tried to evaluate the relationship between maternal metabolic control during pregnancy and metabolic disturbances in children by detecting maternal threshold values for glycemia and hemoglobin A1c (HbA1c) levels that correlate with disturbances in the offspring. Methods Subjects A group of 43 children of mothers with PGDM and 34 children of mothers with GDM were examined at 4– 9 yr of age during one visit in the outpatient clinic. The control group consisted of 108 children of the same age, born to mothers in whom glucose tolerance was normal during pregnancy [negative oral glucose tolerance test (OGTT) performed between 24th and 28th wk of gestation]. All children were born at the Clinical Pediatric Diabetes 2009: 10: 432–440 Hospital of Obstetrics and Gynecology in Poznan, Poland. The data concerning the metabolic control of mothers during pregnancy and the neonatal outcome of the offspring were collected retrospectively from the hospital medical records. Procedures used in the study were accepted by the ethics committee of the hospital. To enroll children in the study, we sent invitation letters clarifying the aim and the design of our study to all mothers in whom PGDM or GDM had been diagnosed during pregnancy and whose children were 4–9 yr old. Only 28% of mothers responded to our letters and came to the outpatient clinic for the scheduled visit. To form the control group, we invited to our study each three children born before and each three children born after the child planned to be included in the study groups, who were of the same gestational age and whose mothers were not diabetic. The rate of positive answers to our letters was 6.5%. Of 108 children of the control group who showed up for the visit, laboratory tests were performed in only 29 because of the lack of consent from other parents to take blood samples. To avoid the selection bias, we checked if the anthropometric parameters and the prevalence of obesity and overweight were comparable in children of the control group in whom metabolic parameters were or not assessed and found that the difference between the two groups was not significant. In the group of women with PGDM, 15 (34.9%) were diagnosed with type B, 20 (46.5%) with type C, 3 (7%) with type D, 3 (7%) with type R, and 2 (4.7%) with type F diabetes according to White’s classification. We have not analyzed whether it was type 1 or type 2 diabetes; the discriminatory factor was the time of onset of the disease. The group of mothers with GDM was further divided into GDM1 (n ¼ 27; 79.4%) and GDM2 (n ¼ 7; 20.6%), depending on if the woman was treated with diet only or insulin, respectively. All women with PGDM and GDM were monitored during pregnancy in the outpatient clinic at the hospital. Their HbA1c was measured at least once in every trimester and additionally before delivery. The first recorded HbA1c for PGDM women was from the first trimester of pregnancy, as soon as pregnancy was confirmed. In GDM mothers, first HbA1c was measured when GDM was diagnosed and afterward before delivery. The results that we show in the paper are based on the last measurement taken within 48 h before delivery. The mean daily glycemia that we analyzed was calculated from the daily glucose profile (seven measurements during 1 d) in the last week before delivery, measured in the hospital. Throughout the whole pregnancy, the glucose control was based mainly on self-monitoring, but each diabetic woman had several pregnancy checks in the hospital, where mean daily glycemia was also assessed. 433 Wroblewska-Seniuk et al. Anthropometric measurements In all children, height and weight were measured, and body mass index (BMI) and BMI z-scores were calculated according to the following formulas (24): (i) BMI ¼ body weight/(height)2 (kg/m2); (ii) BMI z-score ¼ (BMI 2 mean BMI)/SD. Obesity and overweight were diagnosed if BMI z-score was 1.6 (BMI .95th centile for gender and age) and between 1.1 and 1.6 (BMI between the 85th and the 95th centile for gender and age), respectively. The mean BMI used to calculate BMI z-score and the 85th and 95th centile of BMI used to define obesity and overweight were taken from the appropriate growth charts based on the population of Polish children (25, 26). Metabolic parameters In all children of the study groups and in 29 children of the control group, fasting levels of glucose, insulin, and HbA1c were measured, and oral glucose tolerance test was performed. In OGTT, each child was given 150 mL solution containing glucose in the amount of 1.75 g/kg body weight, and glucose and insulin levels were measured at 30, 60, and 120 min. The glucose level in venous blood was determined by means of the enzymatic (hexokinase) method with the Roche Diagnostics laboratory reagents on Hitachi 912 analyzer. Insulin level was measured with the immunodiagnostic test with microparticles (Microparticle Enzyme Immunoassay) (AxSYM Insulina; Abbott Laboratories, Abbott Park, Illinois, USA). The percentage of glycated hemoglobin (HbA1c) in capillary blood was estimated using the Roche Diagnostics Tina-quantÒ Hemoglobin A1c II test. Glucose metabolism disturbances were diagnosed according to the American Diabetes Association based on the following criteria (1): (i) Diabetes – fasting plasma glucose 126 mg/dL or 2-h plasma glucose .200 mg/dL during an OGTT; (ii) IGT – 2-h plasma glucose 140 mg/dL during an OGTT; (iii) Impaired fasting glucose (IFG) – fasting plasma glucose 100 mg/dL but ,126 mg/dL. The following indices were used to determine insulin resistance and insulin sensitivity: (i) Insulin/glucose (I/G) index (27): I/G ¼ insulin (mIU/mL)/glucose (mg/dL); (ii) Homeostasis model assessment index – insulin resistance (HOMA-IR) (28, 29): HOMA-IR ¼ [insulin (mIU/mL) 3 glucose (mmol/L)]/22.5; 434 (iii) Quantitative insulin sensitivity check index (QUICKI) (29, 30): QUICKI ¼ 1/[logI0 (mIU/ mL) 1 logG0 (mg/dL)]; (iv) Insulin sensitivity index – fasting glucose to insulin ratio (FGIR) (31): FGIR ¼ fasting glucose (mg/dL)/fasting insulin (mIU/mL). Statistical analysis Normally distributed data (according to Shapiro–Wilk test) are expressed as means
SD, and data without normal distribution are expressed as median with maximal and minimal values. The differences between groups were evaluated with Student’s t test, Mann– Whitney U test, or Kruskal–Wallis test with multiple comparisons test by Dunn, where appropriate. To examine the relation between groups, the Spearman’s correlation was used. Chi-squared test or Fisher’s exact test was used to compare frequencies in different groups, depending on their size, and the comparison of more than two groups was carried out by means of Fisher–Freeman–Halton test. To discriminate the threshold values of maternal glycemia and HbA1c level for children with and without insulin resistance, we have analyzed receiver operating characteristic (ROC), which is a graph that plots the true positive rate in function of the falsepositive rate at different cutoff points. In all statistical comparisons, p less than 0.05 was considered significant. Evaluations were accomplished using the STATISTICA v. 7.1 and MICROSOFT EXCEL 6.0 with ANALYSE-IT package. Results Table 1 shows the basic demographic data and the data of the neonatal period. The difference in gestational age at birth between the groups does not seem to be clinically important. Thirteen (30.2%) infants of mothers with PGDM were born large for gestational age (LGA) compared with 7 (20.6%) infants of mothers with GDM and 14 (13%) infants of the control group. The difference between the PGDM group and the control group was statistically significant (p , 0.05; Mann–Whitney U test). Small for gestational age (SGA) were two (4.7%) children of mothers with PGDM, one (2.9%) infant of mother with GDM, and eight (7.4%) infants of the control group. The differences between groups were not significant. Obesity and overweight were diagnosed more frequently in children born to mothers with GDM [9 (26.5%) and 3 (8.8%) children, respectively] than in other groups [PGDM: 5 (11.3%) and 3 (7.0%); control Pediatric Diabetes 2009: 10: 432–440 Effects of diabetes on the offspring Table 1. Demographic data and neonatal parameters in infants of gestational and pregestational diabetic mothers compared with infants of the control group Birth weight Gestational age at birth Boys/girls, n (%) Age at the time of study Mothers’ age at the time of delivery PGDM (n ¼ 43) GDM (n ¼ 34) Control group (n ¼ 108) 3500
510 38 (35–40) 3400
590 38 (33–41) 3450
520 39 (34–42) 21/22 (48.8/51.2) 6 (3–9) 18/16 (52.9/47.1) 6 (4.5–9) 51/57 (47.2/52.8) 6 (4–8.5) NS* ,0.05* (PGDM/GDM vs. control group) NS† NS* 27 (21–39) 26 (21–34) 26.5 (18–36) NS* p *Kruskal–Wallis test with multiple comparisons test by Dunn. †Fisher–Freeman–Halton test; NS, not significant. group: 17 (15.7%) and 8 (7.4%)], but the differences were not statistically significant. However, mean values of BMI and BMI z-score were significantly higher in children born to mothers with GDM in comparison to other groups, while there was not any difference between the children of mothers with PGDM and the control group (Fig. 1). Fasting glucose and HbA1c levels were not statistically different between the three groups of children. On the contrary, fasting insulin level was statistically higher in the offspring of mothers with GDM than in the control group (Table 2). The values of insulin resistance indices (HOMA-IR and I/G) were significantly higher, and the values of insulin sensitivity indices (QUICKI and FGIR) were significantly lower in children of mothers with GDM than in children born to mothers with PGDM and children of the control group (Table 2). However, the difference in prevalence of insulin resistance between the offspring of mothers with GDM and the other 3 BMI z-score 2 1 0 –1 –2 –3 PGDM GDM Control group Studied groups of children Compared groups p* PGDM vs. GDM p < 0.05 PGDM vs. control group NS GDM vs. control group p < 0.05 Fig. 1. Body mass index (BMI) z-score in the studied groups (Kruskal–Wallis test with multiple comparison test by Dunn). GDM, gestational diabetes mellitus; PGDM, pregestational diabetes mellitus. Pediatric Diabetes 2009: 10: 432–440 groups was not statistically significant. Severe insulin resistance was diagnosed in three (8.8%) children of GDM group and in one (3.4%) child of the control group [not significant (NS)]; moderate insulin resistance was observed in two (5.9%) children of GDM group, in five (11.6%) children of PGDM group, and in one (3.4%) child of the control group (NS). We have not diagnosed diabetes in any child of the studied groups. IGT was diagnosed in two (5.9%) children of mothers with gestational diabetes, and IFG was diagnosed in four (11.8%) children of mothers with GDM, five (11.6%) children of mothers with PGDM, and one (3.4%) child of the control group. The differences between groups were not statistically significant (Fisher’s exact test). The measurements suggesting IGT or IFG were confirmed in the second OGTT, and children were referred to the diabetology outpatient clinic. Because there was a difference in the number of children born LGA and SGA between the studied groups, we checked if being born LGA or SGA could influence the rate of obesity and insulin resistance. In children born LGA, we observed higher insulin and leptin levels, the highest values of insulin resistance indices and the lowest values of insulin sensitivity indices. However, the differences between the subgroups were not significant. In the GDM group, we compared children born to mothers with GDM treated with diet only (GDM1; n ¼ 27) with those born to mothers treated with insulin (GDM2; n ¼ 7), and we did not observe any statistically significant differences in terms of anthropometric and metabolic parameters. The group of children born to mothers with PGDM was further divided into two subgroups – of children born to mothers with and without vascular complications of diabetes (class D, R, F and B, and C, respectively). The only difference between these two subgroups was in the mean birth weight, which was significantly lower in children born to mothers with class D, R, or F diabetes (3100
200 vs. 3600
500 g). On the contrary, the class of diabetes 435 Wroblewska-Seniuk et al. Table 2. Glucose, insulin, and HbA1c levels and indices of insulin resistance and sensitivity in the studied groups (Kruskal–Wallis test with multiple comparison test by Dunn) PGDM (n ¼ 43) Parameter Glucose (mg/dL) Insulin (mIU/mL) GDM (n ¼ 34) 89.7 (75.5–106.9) 4.3 (1.4–12.7) Control group (n ¼ 29) 90.6 (74.2–103.2) 4.9 (1.8–61.3) p 89.0 (57.0–102.8) 3.5 (0.9–22.9) NS ,0.05 (control group vs. GDM) NS ,0.05 ,0.05 HbA1c (%) 5.4 (4.9–5.9) 5.3 (4.7–8.2) 5.2 (4.6–5.8) Insulin/glucose 0.046 (0.017–0.153) 0.054 (0.021–0.607) 0.041 (0.009–0.301) Homeostasis model 0.943 (0.203–2.794) 1.112 (0.367–15.272) 0.749 (0.220–4.293) assessment index – insulin resistance Fasting glucose to 21.705 (6.532–57.571) 18.583 (1.648–46.750) 24.516 (3.319–110.000) ,0.05 insulin ratio Quantitative insulin 0.384 (0.327–0.487) 0.377 (0.264–0.460) 0.403(0.309–0.513) ,0.05 sensitivity check index HbA1c, hemoglobin A1c; NS, not significant. 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 Glycemia 0.1 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1 – specificity (false-positive results) Fig. 2. Receiver operating characteristic curve assessing the effect of maternal mean daily glucose level in the perinatal period on the incidence of insulin resistance in the offspring. 436 with the highest sensitivity and the highest specificity on the glycemia curve (83.3 and 97 mg/dL, respectively). Discussion Diabetes in pregnancy is considered to be an important risk factor for both mother and her offspring. Perinatal morbidity of children born to diabetic mothers significantly exceeds the rate in the offspring of healthy mothers, the typical complications being congenital malformations, changes in the growth pattern such as macrosomy or intra-uterine growth restriction, and metabolic disturbances in the neonatal period. Most complications are directly because of the impaired metabolic control in mother, in the form of hyperglycemia and hyperinsulinemia. Many authors suggest that fetal hyperinsulinemia is also the factor predisposing to obesity, IGT, and type 2 diabetes in childhood (16, 32). In this paper, we analyzed anthropometric data and biochemical parameters of children born to mothers with pregestational or GDM and compared them with the results in children born to healthy mothers. Sensitivity (true positive results) Sensitivity (true positive results) did not have any impact on metabolic disturbances nor anthropometric parameters in childhood. We also analyzed the effect of maternal metabolic control during pregnancy on metabolic disturbances in the offspring. To determine the maternal metabolic control level, which might predispose to metabolic disturbances in children, we calculated the ROC curves for the first HbA1c measurement in pregnancy, its measurement before delivery, and the average daily glycemia before delivery, comparing children with insulin resistance (HOMA-IR . 2.2) and children without insulin resistance. We noted that mothers of children with insulin resistance had higher HbA1c level and the average daily glycemia before delivery (Figs 2 and 3, Table 3). We did not observe such relationship for the first HbA1c measurement during pregnancy. We estimated that the threshold value for HbA1c in the perinatal period, determining insulin resistance in the offspring, is 6.1%. This cutoff point was chosen as it had the highest sensitivity and specificity. We could not, however, determine such threshold value for the glucose curve as it did not demonstrate any clear cutoff point (Fig. 3). In the Table 3, we have shown the points 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 HbA1c 0 0 0.2 0.4 0.6 0.8 1 1 – specificity (false-positive results) Fig. 3. Receiver operating characteristic curve assessing the effect of maternal hemoglobin A1c (HbA1c) in the perinatal period on the incidence of insulin resistance in the offspring. Pediatric Diabetes 2009: 10: 432–440 Effects of diabetes on the offspring Table 3. Threshold points of maternal HbA1c and mean daily glucose level above which the insulin resistance in children is more likely to occur Curve AUC p 95% CI Threshold point Sensitivity (%) Specificity (%) HbA1c Mean glucose value 0.698 0.671 0.0087 0.0463 0.535–0.862 0.472–0.871 6.11%* 83 mg/dL 97.7 mg/dL 88.9 80 60 64.3 52.4 76.2 CI, confidence interval; HbA1c, hemoglobin A1c; AUC, area under the curve. *Bold value 6.11% is the threshold point for HbA1C that we think might determine insulin resistance in the offspring (further discussed in the text). Additionally, we tried to assess the influence of maternal metabolic control in pregnancy on further development of the offspring. We observed that children born to mothers with gestational diabetes had the tendency toward higher BMI and BMI z-score values when compared with the control group and the group of children born to mothers with PGDM. The incidence of obesity and overweight was also high in this group, although not statistically different from the two other groups. Data presented by other authors are very differentiated. Some show higher risk for obesity in children of mothers with gestational diabetes (33, 34). Many highlight that this tendency, as well as the incidence of IGT, increase with age, which means that it is not only temporary phenomenon (33, 34). Silverman et al. have shown that among 14- to 17-yr-old teenagers, the average BMI in the offspring of mothers with GDM was 26.0 kg/m2, while in the control group, it was 20.9 kg/m2 (19). Obesity at this age correlated with insulin level in amniotic fluid at delivery. What is more, these authors suggested that gestational diabetes influences the birth weight and BMI of children after the age of 4 yr, but it did not have any effect on BMI in the age of 1–3 yr (5, 19). This may suggest that GDM acts with delay and leads to obesity in childhood and early school age or that there are additional factors acting in childhood predisposing to obesity (5). The results of the Whitaker’s study were different. He observed that mild, diet-treated gestational diabetes did not influence the incidence of obesity in the offspring at all. He suggested that this could be because of the less severe glucose intolerance at the time of GDM diagnosis and in the perinatal period in the studied group of mothers (24). In one of the most recent reviews of studies examining offspring of mothers with a history of gestational diabetes and control mothers, Vohr and Boney conclude that the development of metabolic syndrome in children with increasing age is related to maternal GDM, maternal glycemia in the third trimester, maternal obesity, neonatal macrosomia, and childhood obesity (35). Dabelea et al. showed that the risk of obesity and diabetes development was significantly higher in children born after the onset of maternal Pediatric Diabetes 2009: 10: 432–440 diabetes than in their siblings born earlier (36). Such observation is in favor of the hypothesis that obesity and type 2 diabetes in children of diabetic mothers are stimulated not by genetic factors but by the intrauterine environment (36). Gillman et al. observed that the odds ratio of obesity in offspring of mothers with gestational diabetes was 1.4 (37). However, after adjusting for confounding factors, it occurred that birth weight weakens this relationship, which might suggest that GDM has the effect on the incidence of obesity by influencing the birth weight (37). In our study, on the contrary, we did not observe the effect of either large or small birth weight on the incidence of obesity and overweight as well as on the incidence of the glucose metabolism disturbances. Similarly, Silverman et al. did not observe higher incidence of IGT in children born LGA (16). We have not diagnosed diabetes in any child, and the incidence of IGT and IFG did not differ significantly between groups. We have not observed differences in terms of the mean fasting glucose level between groups either, while the fasting insulin level was significantly higher in children of mothers with GDM when compared with the control group. Additionally, in children of mothers with GDM, we observed significantly higher values of insulin resistance and significantly lower values of insulin sensitivity, which suggests the disturbances in the glucose homeostasis in this group. Among children at the age of 5–9 yr, studied by Plagemann et al., the incidence of IGT was 17.4% in offspring of mothers with PGDM and 20% in offspring of mothers with GDM (22). These values are much higher than in our group, where the IGT was diagnosed in only 5.9% children of mothers with GDM and was not observed in children of mothers with PGDM. Similar incidence of the IGT (5.4%) was observed by Silverman et al. (16) in the group of children born to mothers with both gestational and PGDM. The differences might be because of the fact that the abovementioned studies were carried in different populations as well as because of the improvement of perinatal care in diabetic pregnant women in the past years. Plagemann et al. have also observed that glucose tolerance disturbances were more frequent in the 437 Wroblewska-Seniuk et al. offspring of mothers with PGDM than in those with GDM (22). The results of our study are reverse. It might be because of the fact that insulin resistance is related to the BMI. Plagemann et al. did not observe differences in terms of BMI between the offspring of mothers with GDM and PGDM, while in our study, the difference in BMI between these two groups was significant. Obesity and overweight lead to insulin resistance and increase the incidence of IGT, which might be the cause of the differences between the studied groups of children. In several studies, it has been shown that fetal hyperinsulinemia and worse maternal metabolic control with higher HbA1c significantly increased the prevalence of the IGT in the groups of both children of mothers with PGDM and those with GDM (16, 27). It has been shown that there is a correlation between high amniotic insulin level and obesity and/or IGT in childhood and adolescence (16, 19). Increased amniotic insulin level several times increased the risk of IGT development in 10–16 yr of life (16). In other papers, it has been also shown that increased amniotic insulin level correlated with fasting insulin level and insulin resistance index I/G later in life (16, 22). It might be because of the permanent deregulation of insulin secretion as a result of fetal programming and might lead to the early development of type 2 diabetes mellitus in these children (20, 22). In our study, because of the lack of the appropriate retrospective data, we were not able to analyze the impact of fetal hyperinsulinemia on the development of IGT. However, we analyzed the relationship between the maternal metabolic control and the development of metabolic disturbances in the offspring. Based on the ROC curves, we have shown that there is a risk for insulin resistance in children whose mother had higher values of HbA1c level and of the average daily glycemia before delivery. Cutoff point for HbA1c determined on the ROC curve was 6.11%. However, the clear threshold value for average glycemia could not be determined. These results clearly show that the level of metabolic control of diabetes in the perinatal period significantly influences the development of metabolic disturbances, such as insulin resistance in the offspring. On the contrary, the advance of pregestational and gestational maternal diabetes mellitus did not have impact on anthropometric and/or metabolic parameters in the offspring. Observations of Silverman were similar. He did not find the relationship between the class of maternal diabetes and the development of IGT (16). The limitation of our study is that we were not able to analyze the confounding factors that can influence obesity and glucose tolerance in children. One of these factors is maternal prepregnancy BMI. It has been shown that children exposed to maternal obesity are at increased risk of developing obesity and glucose intolerance irrespective of the fact that their mothers 438 suffered from diabetes or not (38). Another confounder is breastfeeding, inversely associated with childhood obesity regardless of maternal diabetes status or weight status (39–41). Somm et al. have also proven that there is a direct association between fetal nicotine exposure and offspring metabolic syndrome with early signs of dysregulations of adipose tissue and pancreatic development (42). The results of our study, as well as those of other authors (5, 16, 22, 32, 33), are in agreement with the hypothesis of teratogenesis driven by the intra-uterine environment. It suggests that hyperglycemia and hyperinsulinemia in gestation complicated by diabetes lead to the improper programming of metabolic and neuroendocrine systems, increasing the susceptibility to glucose metabolism disorders and diabetes. Therefore, it is necessary to maintain the optimal metabolic control in women with PGDM and to perform routine screening for glucose intolerance in all pregnant women in order to early diagnose and treat GDM. Indispensable is also promotion of the healthy lifestyle in women who suffered from gestational diabetes as it is known that GDM predisposes for the development of type 2 diabetes. It is also advisable to extend the care for children of pregnancies complicated with diabetes beyond the neonatal period as it might help prevent the development of glucose intolerance and type 2 diabetes in this group or at least reduce incidence of these conditions. To conclude, we can say that in spite of the fact that early diagnosis of glucose intolerance in pregnancy and modern treatment approaches allowed for better metabolic control in the mother and her fetus, the group of children of diabetic mothers is still a group of higher risk for early and late metabolic complications. The relationship between maternal diabetes in pregnancy and complications in both newborns and older children should stimulate to continuous improvements in perinatal care of diabetic women and thorough monitoring of their offspring. Acknowledgement The study was sponsored by the Polish Ministry of Science and Higher Education, grant number 4PO5E 120 25. References 1. American Diabetes Association, Standards of Medical Care in Diabetes-2006. Diabetes Care 2006: 29 (Suppl.1): S4–S42. 2. Gestational diabetes mellitus. Diabetes Care 2004: 27 (Suppl. 1): S88–S90. 3. S CHAEFER -G RAF UM, B UCHANAN TA, X IANG A, SONGSTER G, MONTORO M, KJOS SL. Patterns of congenital anomalies and relationship to initial maternal fasting glucose levels in pregnancies complicated by type 2 and gestational diabetes. Am J Obstet Gynecol 2000: 182: 313–320. Pediatric Diabetes 2009: 10: 432–440 Effects of diabetes on the offspring 4. MOSES RG, CALVERT D. Pregnancy outcomes in women without gestational diabetes mellitus related to the maternal glucose level. Is there a continuum of risk? Diabetes Care 1995: 18: 1527–1533. 5. SILVERMAN BL, RIZZO T, GREEN OC et al. Long-term prospective evaluation of offspring of diabetic mothers. Diabetes 1991: 40 (Suppl. 2): 121–125. 6. BECERRA JE, KHOURY MJ, CORDERO JF, ERICKSON JD. Diabetes mellitus during pregnancy and the risks for specific birth defects: a population-based case-control study. Pediatrics 1990: 85: 1–9. 7. HARDER T, KOHLHOFF R, DORNER G, ROHDE W, PLAGEMANN A. Perinatal Ôprogramming’ of insulin resistance in childhood: critical impact of neonatal insulin and low birth weight in a risk population. Diabet Med 2001: 18: 634–639. 8. KITZMILLER JL, GAVIN LA, GIN GD, JOVANOVICPETERSON L, MAIN EK, ZIGRANG WD. Preconception care of diabetes. Glycemic control prevents congenital anomalies. JAMA 1991: 265: 731–736. 9. DAMM P, MOLSTED-PEDERSEN L. Significant decrease in congenital malformations in newborn infants of an unselected population of diabetic women. Am J Obstet Gynecol 1989: 161: 1163–1167. 10. GOLDMAN JA, DICKER D, FELDBERG D, YESHAYA A, SAMUEL N, KARP M. Pregnancy outcome in patients with insulin-dependent diabetes mellitus with preconceptional diabetic control: a comparative study. Am J Obstet Gynecol 1986: 155: 293–297. _ 11. WENDER-OZEGOWSKA E. Czynniki wp1ywaja˛ce na rozwój p1odu w przebiegu cia˛z_ y powik1anej cukrzyca˛. Kliniczna Perinatologia i Ginekologia 2000: 19 2000, Poznań: Ośrodek Wydawnictw Naukowych. 12. STENNINGER E, SCHOLLIN J, AMAN J. Early postnatal hypoglycaemia in newborn infants of diabetic mothers. Acta Paediatr 1997: 86: 1374–1376. 13. BALLARD JL, ROSENN B, KHOURY JC, MIODOVNIK M. Diabetic fetal macrosomia: significance of disproportionate growth. J Pediatr 1993: 122: 115–119. 14. OJOMO EO, COUSTAN DR. Absence of evidence of pulmonary maturity at amniocentesis in term infants of diabetic mothers. Am J Obstet Gynecol 1990: 163: 954–957. 15. HOET JP, LUKENS FD. Carbohydrate metabolism during pregnancy. Diabetes 1954: 3: 1–12. 16. SILVERMAN BL, METZGER BE, CHO NH, LOEB CA. Impaired glucose tolerance in adolescent offspring of diabetic mothers: relationship to fetal hyperinsulinism. Diabetes Care 1995: 18: 611–617. 17. FERRANNINI E, HAFFNER SM, MITCHELL BD, STERN MP. Hyperinsulinaemia: the key feature of a cardiovascular and metabolic syndrome. Diabetologia 1991: 34: 416– 422. 18. KOHLHOFF R, DORNER G. Perinatal hyperinsulinism and perinatal obesity as risk factors for hyperinsulinaemia in later life. Exp Clin Endocrinol 1990: 96: 105–108. 19. SILVERMAN BL, RIZZO TA, CHO NH, METZGER BE. Long-term effects of the intrauterine environment. The Northwestern University Diabetes in Pregnancy Center. Diabetes Care 1998: 21 (Suppl. 2): B142–B149. 20. DABELEA D. The predisposition to obesity and diabetes in offspring of diabetic mothers. Diabetes Care 2007: 30: S169–S174. 21. BARTELHEIMER H, KLOOS K. [Effects of experimental diabetes on pregnancy and fetus.]. Z Gesamte Exp Med 1952: 119: 246–265. 22. 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: 1094–1100. Pediatric Diabetes 2009: 10: 432–440 23. PERSSON B, GENTZ J, MOLLER E. Follow-up of children of insulin dependent (type I) and gestational diabetic mothers. Growth pattern, glucose tolerance, insulin response, and HLA types. Acta Paediatr Scand 1984: 73: 778–784. 24. WHITAKER RC, PEPE MS, SEIDEL KD, WRIGHT JA, KNOPP RH. Gestational diabetes and the risk of offspring obesity. Pediatrics 1998: 101: E9. _ 25. KRAWCZYŃSKI M, KRZYZANIAK A, WALKOWIAK T. Normy rozwojowe wysokości i masy cia1a dzieci i m1odzie_zy miasta Poznania w wieku od 3 do 18 lat. Pediatria Praktyczna 2000: 8: 341–353. _ 26. KRZYZANIAK A, KRAWCZYŃSKI M, WALKOWIAK T. Wskaźniki proporcji wagowo- wzrostowych w populacji dzieci i m1odzie_zy miasta Poznania. Pediatria Praktyczna 2000: 8: 355–364. 27. WEISS PAM, SCHOLZ HS, HAAS J, TAMUSSINO KF, SEISSLER J, BORKENSTEIN MH. Long-term follow-up of infants of mothers with type 1 diabetes. Diabetes Care 2000: 23: 905–911. 28. MATTHEWS DR, HOSKER JP, RUDENSKI AS, NAYLOR BA, TREACHER DF, TURNER RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985: 28: 412–419. 29. RADIKOVA Z. Assessment of insulin sensitivity/resistance in epidemiological studies. Endocr Regul 2003: 37: 189– 194. 30. KATZ A, NAMBI SS, MATHER K et al. Quantitative insulin sensitivity check index: a simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab 2000: 85: 2402–2410. 31. LEGRO RS, FINEGOOD D, DUNAIF A. A fasting glucose to insulin ratio is a useful measure of insulin sensitivity in women with polycystic ovary syndrome. J Clin Endocrinol Metab 1998: 83: 2694–2698. 32. AERTS L, HOLEMANS K, VAN ASCHE FA. Maternal diabetes during pregnancy: consequences for the offspring. Diabetes Metab Rev 1990: 6: 147–167. 33. PETTITT DJ, BAIRD HR, ALECK KA, BENNETT PH, KNOWLER WC. Excessive obesity in offspring of Pima Indian women with diabetes during pregnancy. N Engl J Med 1983: 308: 242–245. 34. KOSTALOVA L, LESKOVA L, KAPELLEROVA A, STRBAK V. Body mass, plasma leptin, glucose, insulin and Cpeptide in offspring of diabetic and non-diabetic mothers. Eur J Endocrinol 2001: 145: 53–58. 35. VOHR BR, BONEY CM. Gestational diabetes: the forerunner for the development of maternal and childhood obesity and metabolic syndrome? J Matern Fetal Neonatal Med 2008: 21: 149–157. 36. DABELEA D, HANSON RI, LINDSAY RS et al. Intrauterine exposure to diabetes conveys risks for type 2 diabetes and obesity. Diabetes 2000: 49: 2208–2211. 37. GILLMAN MW, RIFAS-SHIMAN S, BERKEY CS, FIELD AE, COLDITZ GA. Maternal gestational diabetes, birth weight, and adolescent obesity. Pediatrics 2003: 111: e221–e226. 38. BONEY CM, VERMA A, TUCKER R, VOHR BR. Metabolic syndrome in childhood: association with birth weight, maternal obesity, and gestational diabetes mellitus. Pediatrics 2005: 115: e290–e296. 39. MAYER-DAVIS EJ, RIFAS-SHIMAN SL, ZHOU L, HU FB, COLDITZ GA, GILLMAN MW. Breast-feeding and risk for childhood obesity: does maternal diabetes or obesity status matter? Diabetes Care 2006: 29: 2231–2237. 40. LI C, KAUR H, CHOI WS, HUANG TT, LEE RE, AHLUWALIA JS. Additive interactions of maternal prepregnancy BMI and breast-feeding on childhood overweight. Obes Res 2005: 13: 362–371. 439 Wroblewska-Seniuk et al. 41. GRUMMER-STRAWN LM, MEI Z. Does breastfeeding protect against pediatric overweight? Analysis of longitudinal data from the Centers for Disease Control and Prevention Pediatric Nutrition Surveillance System. Pediatrics 2004: 113: e81–e86. 440 42. SOMM E, SCHWITZGEBEL VM, VAUTHAY DM et al. Prenatal nicotine exposure alters early pancreatic islet and adipose tissue development with consequences on the control of body weight and glucose metabolism later in life. Endocrinology 2008: 149: 6289–6299. Pediatric Diabetes 2009: 10: 432–440