Matern Child Health J (2015) 19:578–582 DOI 10.1007/s10995-014-1541-9
Bone Density Among Infants of Gestational Diabetic Mothers and Macrosomic Neonates Irit Schushan-Eisen • Mor Cohen • Leah Leibovitch Ayala Maayan-Metzger • Tzipora Strauss
•
Published online: 20 August 2014 Ó Springer Science+Business Media New York 2014
Abstract Decreased bone density has been found among infants of diabetic mothers and among large-for-gestational-age newborns. To evaluate which etiologies (physical or metabolic effect) have the greatest impact on neonatal bone density. A case–control study was conducted that included two study groups: one comprising 20 appropriate-for-gestational-age (AGA) infants of gestational diabetic mothers (IGDM) and matched controls, and the other comprising 20 macrosomic infants (birth weight [ 4 kg) and matched controls. Bone density was examined along the tibia bone using quantitative ultrasound that measured speed of sound. Bone density among the group of macrosomic infants was significantly lower than among the control group (2,976 vs. 3,120 m/s respectively, p \ 0.005). No differences in bone density were found between infants of diabetic mothers and their controls (3,005 vs. 3,043 m/s respectively, p = 0.286). Low bone density was predicted only by birth weight (for every increase of 100 g) (OR 1.148 [CI 1.014–1.299], p = 0.003). Bone density was found to be low among macrosomic newborn infants, whereas among AGA– IGDM infants bone density was similar to that of the control group. These findings strengthen the hypothesis
I. Schushan-Eisen L. Leibovitch A. Maayan-Metzger T. Strauss (&) Department of Neonatology, The Edmond and Lili Safra Children’s Hospital, Sheba Medical Center, Tel Hashomer, 52621 Ramat Gan, Israel e-mail: [email protected] I. Schushan-Eisen M. Cohen L. Leibovitch A. Maayan-Metzger T. Strauss Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
123
that reduced fetal movements secondary to fetal macrosomia constitute the mechanism for reduced bone density. Keywords Bone density Infants of diabetic mothers Large-for-gestational-age Macrosomia
Introduction Gestational diabetes mellitus during pregnancy is a common complication that affects 5–10 % of pregnancies worldwide [1]. The prevalence of gestational diabetic mothers in our medical center was found to be 5.4 % of deliveries [2]. Infants of gestational diabetic mothers (IGDMs) may suffer from various complications, including congenital malformations, metabolic abnormalities (such as hypocalcaemia and hypomagnesaemia), hematologic abnormalities, cardiomyopathy, respiratory distress syndrome and birth trauma [3, 4]. Infants of gestational diabetic mothers tend to have a higher birth weight (BW) than infants born to non-diabetic mothers. Due to a secondary response to the high glucose levels transferred from their mothers these infants produce high insulin levels. Insulin is known to be a growth factor and therefore influences intrauterine growth. High birth weight is known as a risk factor for various complications, among them decreased fetal movements, a fact that may impact fetal bone density. Indeed, decreased bone density has been described among IGDMs [5–7]. Other studies have described lower bone density among large-for-gestational-age (LGA) newborn infants [8] and among infants born following breech presentation [9], which may imply an association to decreased intrauterine fetal movements. It remains to be determined whether the reason for decreased bone density among IGDMs is due mainly to
Matern Child Health J (2015) 19:578–582
decreased intrauterine fetal movements as a result of high weight (physical etiology) or whether might be secondary to various metabolic anomalies associated with the mother’s diabetic condition. The aim of the current study is to evaluate which of the etiologies (physical or metabolic effect) has the greatest impact upon neonatal bone density.
Methods The study population included two study groups, each containing 20 term (GA C 37 weeks) infants. All infants were born at Sheba Medical Center between 2010 and 2011. Group A included infants of gestational diabetic mothers (IGDM), all of whom with BW appropriate for gestational age (AGA). Group B included macrosomic infants with BW [ 4 kg without evidence of gestational diabetic mothers. For each infant in the study groups, a control infant was matched for gestational age (GA). Exclusion criteria were newborns with major congenital malformations, newborns of mothers with chronic diseases or diseases of pregnancy other than diabetes, newborns of mothers who took medications other than diabetes medications during pregnancy and infants with an active disease after birth (infectious, respiratory, cardiac, etc.). Bone density was measured during the immediate postnatal hospitalization at 1–4 days of life. Bone mineral density was measured using quantitative ultrasound (Sunlight Omnisence Premier, Tel Aviv, Israel), which is safe, radiation-free, and has been used in this age group [10–12]. The measurement obtained from the quantitative ultrasound is the speed of sound (SOS) that passes through the bone. Higher bone mineral density and additional parameters of bone quality and strength result in higher sound velocities [11]. Measurements were taken of the mid left tibial bone in all infants by a single investigator. The mean of five SOS measurements was used for data analysis. Agespecific SOS z-scores are provided by the Omnisence Premier program, based on pre-marketing international studies. Data recorded from the infants’ medical records and maternal reports included GA, BW, length (crown-heel), head circumference, gender, Apgar score at 1 and 5 min, mode of delivery, perinatal infant disease, GDM (no, A1 or A2) and day of life when examination was performed. The study was approved by the institutional research ethics board of the Chaim Sheba Medical Center. Statistical Analysis The sample size was calculated on the bases of previous studies showing values of 3,100 ± 200 m/s [5] to identify
579
a reduction of at least 7 % in bone SOS with a = 0.05 and power of 90 %. We there for calculated that *20 infants in each group are needed to identify this reduction. The statistical analysis was performed using SPSS version 17. For continuous variables independent t test was performed. For categorical variables v2 tests or Pearson exact tests were performed. Statistical significance was defined as p \ 0.05. A logistic regression model was used to predict the probability of low bone density. Factors that were found to have a statistical value of p B 0.1 on univariate analysis were included in the model of the multivariate analysis.
Results The study groups included 20 AGA infants of diabetic mothers (group A) and 20 macrosomic infants to non-diabetic mothers (group B) matched to 40 controls AGA infants to nondiabetic mothers. Data of study and control groups is presented in Table 1. Mean maternal age and percentage of 1st parity were similar in both study and control groups. Maternal diseases and medications included hypothyroidism treated with L-thyroxine (group A and control 1 mother each and 2 mothers in the control of group B) and one mother with Crohn’s disease in group B treated with Rafassal. In addition 4 mothers were treated with Clexane, one mother with Aspirin and one mother with Citalopram. No significant statistical difference was found between the maternal groups. Group A: IGDM Versus Controls Comparing the characteristics of the IGDM study group to characteristics of the controls (Table 1) showed that infants in the study group had a higher birth weight. A higher percentage of males and a higher rate of cesarean section (CS) were found among the study group compared to the controls. Length was similar in both groups. Bone density was similar in the study and control groups. Comparisons of the subgroup of IGDM of mothers treated with insulin (n = 7), the subgroup of IGDM mothers treated with diet (n = 13) and the controls revealed no significant differences in bone density among the three subgroups (2,985, 3,016, 3,040 m/s respectively, p = 0.5). Group B: Macrosomic Infants Versus Controls Table 1 shows the data parameters of the macrosomic infants compared to the controls. In addition to BW, among the macrosomic group length and head circumference were significantly higher compared to controls. The bone density of the study group was significantly lower than that of the control group.
123
580
Matern Child Health J (2015) 19:578–582
Table 1 Characteristics of IGDM (group A) and macrosomic infants (BW [ 4 kg) (group B) compared to control group Characteristics
IGDM (n = 20)
Control group (n = 20)
p value
BW [ 4 kg (n = 20)
Control group (n = 20)
GA (weeks)a
38.9 ± 0.9
38.9 ± 0.9
1
40.35 ± 0.98
40.35 ± 0.98
Range
37–41
37–41
38–42
38–42
BW (gr)a
3,441 ± 358
3,170 ± 378
Range
2,745–3,900
2,560–3,835
LGA (%)
1 (5)
0
Sex: male N (%)
12 (60)
5 (25)
Female N (%) Maternal agea
8 (40) 33.8 ± 5.1
Delivery 1st N (%)
p value 1
4,215 ± 150
3,324 ± 258
4,010–4,600
2,840–3,835
0.311
9 (45)
0
0.025
9 (45)
9 (45)
15 (75) 32.7 ± 3.9
0.43
11 (55) 32.0 ± 4.1
11 (55) 32.9 ± 4.8
0.511
8 (40)
8 (40)
1
3 (15)
5 (25)
0.695
PS
9 (45)
18 (90)
0.002
12 (60)
17 (85)
0.065
CS
11 (55)
2 (10)
8 (40)
2 (10)
0
1 (5)
0.005
Bone Density Among Infants of Gestational Diabetic Mothers and Macrosomic Neonates Irit Schushan-Eisen • Mor Cohen • Leah Leibovitch Ayala Maayan-Metzger • Tzipora Strauss
•
Published online: 20 August 2014 Ó Springer Science+Business Media New York 2014
Abstract Decreased bone density has been found among infants of diabetic mothers and among large-for-gestational-age newborns. To evaluate which etiologies (physical or metabolic effect) have the greatest impact on neonatal bone density. A case–control study was conducted that included two study groups: one comprising 20 appropriate-for-gestational-age (AGA) infants of gestational diabetic mothers (IGDM) and matched controls, and the other comprising 20 macrosomic infants (birth weight [ 4 kg) and matched controls. Bone density was examined along the tibia bone using quantitative ultrasound that measured speed of sound. Bone density among the group of macrosomic infants was significantly lower than among the control group (2,976 vs. 3,120 m/s respectively, p \ 0.005). No differences in bone density were found between infants of diabetic mothers and their controls (3,005 vs. 3,043 m/s respectively, p = 0.286). Low bone density was predicted only by birth weight (for every increase of 100 g) (OR 1.148 [CI 1.014–1.299], p = 0.003). Bone density was found to be low among macrosomic newborn infants, whereas among AGA– IGDM infants bone density was similar to that of the control group. These findings strengthen the hypothesis
I. Schushan-Eisen L. Leibovitch A. Maayan-Metzger T. Strauss (&) Department of Neonatology, The Edmond and Lili Safra Children’s Hospital, Sheba Medical Center, Tel Hashomer, 52621 Ramat Gan, Israel e-mail: [email protected] I. Schushan-Eisen M. Cohen L. Leibovitch A. Maayan-Metzger T. Strauss Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
123
that reduced fetal movements secondary to fetal macrosomia constitute the mechanism for reduced bone density. Keywords Bone density Infants of diabetic mothers Large-for-gestational-age Macrosomia
Introduction Gestational diabetes mellitus during pregnancy is a common complication that affects 5–10 % of pregnancies worldwide [1]. The prevalence of gestational diabetic mothers in our medical center was found to be 5.4 % of deliveries [2]. Infants of gestational diabetic mothers (IGDMs) may suffer from various complications, including congenital malformations, metabolic abnormalities (such as hypocalcaemia and hypomagnesaemia), hematologic abnormalities, cardiomyopathy, respiratory distress syndrome and birth trauma [3, 4]. Infants of gestational diabetic mothers tend to have a higher birth weight (BW) than infants born to non-diabetic mothers. Due to a secondary response to the high glucose levels transferred from their mothers these infants produce high insulin levels. Insulin is known to be a growth factor and therefore influences intrauterine growth. High birth weight is known as a risk factor for various complications, among them decreased fetal movements, a fact that may impact fetal bone density. Indeed, decreased bone density has been described among IGDMs [5–7]. Other studies have described lower bone density among large-for-gestational-age (LGA) newborn infants [8] and among infants born following breech presentation [9], which may imply an association to decreased intrauterine fetal movements. It remains to be determined whether the reason for decreased bone density among IGDMs is due mainly to
Matern Child Health J (2015) 19:578–582
decreased intrauterine fetal movements as a result of high weight (physical etiology) or whether might be secondary to various metabolic anomalies associated with the mother’s diabetic condition. The aim of the current study is to evaluate which of the etiologies (physical or metabolic effect) has the greatest impact upon neonatal bone density.
Methods The study population included two study groups, each containing 20 term (GA C 37 weeks) infants. All infants were born at Sheba Medical Center between 2010 and 2011. Group A included infants of gestational diabetic mothers (IGDM), all of whom with BW appropriate for gestational age (AGA). Group B included macrosomic infants with BW [ 4 kg without evidence of gestational diabetic mothers. For each infant in the study groups, a control infant was matched for gestational age (GA). Exclusion criteria were newborns with major congenital malformations, newborns of mothers with chronic diseases or diseases of pregnancy other than diabetes, newborns of mothers who took medications other than diabetes medications during pregnancy and infants with an active disease after birth (infectious, respiratory, cardiac, etc.). Bone density was measured during the immediate postnatal hospitalization at 1–4 days of life. Bone mineral density was measured using quantitative ultrasound (Sunlight Omnisence Premier, Tel Aviv, Israel), which is safe, radiation-free, and has been used in this age group [10–12]. The measurement obtained from the quantitative ultrasound is the speed of sound (SOS) that passes through the bone. Higher bone mineral density and additional parameters of bone quality and strength result in higher sound velocities [11]. Measurements were taken of the mid left tibial bone in all infants by a single investigator. The mean of five SOS measurements was used for data analysis. Agespecific SOS z-scores are provided by the Omnisence Premier program, based on pre-marketing international studies. Data recorded from the infants’ medical records and maternal reports included GA, BW, length (crown-heel), head circumference, gender, Apgar score at 1 and 5 min, mode of delivery, perinatal infant disease, GDM (no, A1 or A2) and day of life when examination was performed. The study was approved by the institutional research ethics board of the Chaim Sheba Medical Center. Statistical Analysis The sample size was calculated on the bases of previous studies showing values of 3,100 ± 200 m/s [5] to identify
579
a reduction of at least 7 % in bone SOS with a = 0.05 and power of 90 %. We there for calculated that *20 infants in each group are needed to identify this reduction. The statistical analysis was performed using SPSS version 17. For continuous variables independent t test was performed. For categorical variables v2 tests or Pearson exact tests were performed. Statistical significance was defined as p \ 0.05. A logistic regression model was used to predict the probability of low bone density. Factors that were found to have a statistical value of p B 0.1 on univariate analysis were included in the model of the multivariate analysis.
Results The study groups included 20 AGA infants of diabetic mothers (group A) and 20 macrosomic infants to non-diabetic mothers (group B) matched to 40 controls AGA infants to nondiabetic mothers. Data of study and control groups is presented in Table 1. Mean maternal age and percentage of 1st parity were similar in both study and control groups. Maternal diseases and medications included hypothyroidism treated with L-thyroxine (group A and control 1 mother each and 2 mothers in the control of group B) and one mother with Crohn’s disease in group B treated with Rafassal. In addition 4 mothers were treated with Clexane, one mother with Aspirin and one mother with Citalopram. No significant statistical difference was found between the maternal groups. Group A: IGDM Versus Controls Comparing the characteristics of the IGDM study group to characteristics of the controls (Table 1) showed that infants in the study group had a higher birth weight. A higher percentage of males and a higher rate of cesarean section (CS) were found among the study group compared to the controls. Length was similar in both groups. Bone density was similar in the study and control groups. Comparisons of the subgroup of IGDM of mothers treated with insulin (n = 7), the subgroup of IGDM mothers treated with diet (n = 13) and the controls revealed no significant differences in bone density among the three subgroups (2,985, 3,016, 3,040 m/s respectively, p = 0.5). Group B: Macrosomic Infants Versus Controls Table 1 shows the data parameters of the macrosomic infants compared to the controls. In addition to BW, among the macrosomic group length and head circumference were significantly higher compared to controls. The bone density of the study group was significantly lower than that of the control group.
123
580
Matern Child Health J (2015) 19:578–582
Table 1 Characteristics of IGDM (group A) and macrosomic infants (BW [ 4 kg) (group B) compared to control group Characteristics
IGDM (n = 20)
Control group (n = 20)
p value
BW [ 4 kg (n = 20)
Control group (n = 20)
GA (weeks)a
38.9 ± 0.9
38.9 ± 0.9
1
40.35 ± 0.98
40.35 ± 0.98
Range
37–41
37–41
38–42
38–42
BW (gr)a
3,441 ± 358
3,170 ± 378
Range
2,745–3,900
2,560–3,835
LGA (%)
1 (5)
0
Sex: male N (%)
12 (60)
5 (25)
Female N (%) Maternal agea
8 (40) 33.8 ± 5.1
Delivery 1st N (%)
p value 1
4,215 ± 150
3,324 ± 258
4,010–4,600
2,840–3,835
0.311
9 (45)
0
0.025
9 (45)
9 (45)
15 (75) 32.7 ± 3.9
0.43
11 (55) 32.0 ± 4.1
11 (55) 32.9 ± 4.8
0.511
8 (40)
8 (40)
1
3 (15)
5 (25)
0.695
PS
9 (45)
18 (90)
0.002
12 (60)
17 (85)
0.065
CS
11 (55)
2 (10)
8 (40)
2 (10)
0
1 (5)
0.005
