Prediction of Fetal Macrosomia in Diabetic Pregnancies by Sonographic Measurement of Fetal Subcutaneous Fat Thickness

Prediction of Fetal Macrosomia in Diabetic Pregnancies by Sonographic Measurement of Fetal Subcutaneous Fat Thickness

1Mounica Schmidt-Fittschen¹, Ammar Al Naimi¹, Natalie Filmann², Franz Bahlmann¹

¹Department of Obstetrics & Gynecology, Buergerhospital Frankfurt am Main, Germany ²Institute of Biostatistics and Mathematical Modeling, Department of Medicine, Goethe University, Frankfurt

*Corresponding author: Dr. Mounica Schmidt-Fittschen, Department of Obstetrics & Gynecology, Buergerhospital Frankfurt, Nibelungenallee 37-41, 60318 Frankfurt, Germany, Tel: +49 69/1500412; Fax: +49 69/1500400; Email: f.bahlmann@buergerhospital-ffm.de

Abstract

Objective: To compare the usefulness of abdominal circumference (AC), anterior abdominal wall thickness (AAW), soft tissue thickness and subcutaneous fat tissue thickness of the upper arm (HumST, HumSC) and thigh (FemST, FemSC) as well as the ratios of subcutaneous fat tissue thickness to humerus (HumSC/HL) and femur length (FemSC/FL) in predicting fetal macrosomia in pregnancies with gestational diet-controlled diabetes, gestational insulin-dependent diabetes, pre-gestational Type 1 and Type 2 diabetes mellitus.

Methods: Basic fetal biometric measurements, fetal soft tissue and subcutaneous fat tissue thickness measurements, as listed above, were performed in a cross-sectional study of 138 women between 25 and 39 weeks of gestation with singleton pregnancies complicated by diabetes.

Results: Nineteen out of 138 neonates showed macrosomic growth. The distribution in the categories of diabetes indicated the highest risk of macrosomia in diabetes Type 1 and Type 2. A significant difference between macrosomic and appropriate-for-gestational age fetuses was shown for fetal humerus length, HumST, HumSC, and HumSC/HL (p= 0.034, p= 0.023, p= 0.002, p= 0.0003, respectively). Comparison of detection efficacy using areas under the receiver operating characteristic curves and corresponding p-values confirmed that HumST, HumSC, and HumSC/HL (AUC 0.729, p= 0.009; AUC 0.777, p<0.0001; AUC 0.787, p< 0.0001, respectively) presented the best detection efficacy for macrosomia.

Conclusion: Our data showed that measurement of fetal soft tissue thickness and subcutaneous fat tissue thickness of the upper arm as well as the ratio of subcutaneous fat tissue thickness to humerus length were good predictors of fetal macrosomia in diabetic pregnancies.

Keywords

prenatal ultrasoud; fetal subcutaneous fat tissue; gestational diabetes; fetal macrosomia

Abbreviations 

AAW: Abdominal Wall Thickness;

AC: Abdominal Circumference;

AGA: Appropriate-For-Gestational Age;

AUC: Area Under The Curve;

BMI: Body Mass Index;

Femsc: Subcutaneous Fat Tissue Thickness Thigh;

Femsc/HL: Ratio Of Subcutaneous Fat Tissue Thickness To Femur Length;

Femst: Soft Tissue Thickness Thigh;

GDM: Gestational Diabetes;

Humsc: Subcutaneous Fat Tissue Thickness Upper Arm;

Humsc/HL: Ratio Of Subcutaneous Fat Tissue Thickness To Humerus Length;

Humst: Soft Tissue Thickness Upper Arm;

LGA: Large-For-Gestational Age;

Mhz: Megahertz;

PGDM: Pre-Gestational Diabetes Mellitus;

ROC: Receiver Operating Characteristic;

SD: Standard Deviation;

Introduction

Gestational diabetes mellitus is defined as glucose intolerance diagnosed first during pregnancy using a 75g glucose tolerance test according to the International Association of the Diabetes and Pregnancy Study Group`s criteria [1]. Pre-gestational diabetes mellitus (PGDM) and gestational diabetes mellitus (GDM) are among the most common metabolic diseases during pregnancy. They are associated with an increased risk of fetal macrosomia and perinatal complications for the fetus and mother such as shoulder dystocia, prolonged labour, instrumental delivery, high-grade perineal tears and caesarean section [2,3]. Furthermore, macrosomic and large-for-gestational age (LGA) fetuses have an increased risk of developing a metabolic syndrome (obesity, dyslipidemia, hypertension, glucose intolerance) during childhood and adolescence [4,5]. Poor maternal glycemic control is correlated with fetal hyperinsulinemia leading to an increase in subcutaneous fat, which in turn is responsible for fetal macrosomia [6,7].

In order to be able to predict perinatal complications during childbirth several studies have addressed the diagnosis of fetal macrosomia using different ultrasound based methods [8]. Fetal weight and abdominal circumference, the latter supposedly the most sensitive individual parameter, are commonly used in the prediction of excessive fetal growth. Both estimates, however, show a wide range of variation and may differ from the actual fetal weight by 10-15% [3,9].

In neonates, skinfold measurements are commonly used to assess perinatal body composition and body fat mass [10,11]. Therefore, ultrasound measurement of fetal subcutaneous fat tissue thickness would seem to be a useful method to predict growth abnormalities such as fetal macrosomia in diabetic pregnancies, in which excessive growth is mainly associated with increased fetal subcutaneous fat tissue depots [9]. Measurements of subcutaneous fat tissue, femur and humerus length have the advantage that they are simple to perform and their accuracy generally is not affected by labour or rupture of membranes in contrast to fetal weight estimation which is based on the accuracy of measurements of femur, head and abdominal circumference.

Rotmensch et al [12] examined the screening efficacy of the ratio of subcutaneous fat tissue thickness of the thigh to femur length in non-diabetic pregnancies. They showed that the ratio was a poor predictor of fetal macrosomia in the non-diabetic pregnancy but they did hypothesize that fetal subcutaneous tissue thickness would show a better correlation to birth weight in diabetic mothers.

In this study we measured fetal abdominal circumference, anterior abdominal wall thickness, soft tissue thickness and subcutaneous fat tissue thickness of the upper arm (major tubercle level) and thigh (greater trochanter level), and the ratios between subcutaneous fat tissue thickness to the length of humerus and femur in diabetic pregnancies between 25 and 39 weeks of gestation to evaluate their usefulness in predicting fetal macrosomia.

Materials and Methods

One hundred and fifty-eight patients received one ultrasound examination, performed by the same investigator between February 2014 and May 2016. Inclusion criteria were gestational age between 25 and 39 weeks and a singleton pregnancy complicated by diet-controlled GDM, insulin-dependent GDM, PGDM Type 1 or PGDM Type 2. Exclusion criteria were fetal malformations, drug dependence, labour or if patients were incapable of giving consent. Twenty patients were excluded retrospectively from the final analysis due to fetal growth retardation, determined using Hadlock`s formula (Log₁₀ (weight) = 1.326 – 0.00326 AC × FL + 0.0107 HC + 0.0438 AC + 0.158 FL) [13,14] or incomplete perinatal data (moving out of the area). All patients provided written informed consent and the study was approved by our hospital`s ethics committee. Gestational age was calculated from the last menstrual period, confirmed by first trimester crown-rump length measurement and corrected when required [15]. All women had an uneventful pregnancy.

A glucose tolerance test using a 75g glucose load was performed between gestational weeks 24-28 according to the International Association of the Diabetes and Pregnancy Study Group`s criteria [1]. Diagnosis of GDM was based on at least one abnormal glucose measurement in the glucose tolerance test (fasting glucose ≥ 92 mg/dl, plasma glucose ≥ 180 mg/dl 1hour, ≥ 153 mg/dl 2 hours after glucose administration). Sixty-nine patients required insulin to treat their GDM, whereas 39 were able to achieve normal glucose levels with dietary counselling. Twenty patients had PGDM Type 1 and 10 patients had PGDM Type 2. BMI values ranged from 15.24 to 49.5. Each patient was examined by transabdominal ultrasound examination using an Aplio 500 ultrasound device (Toshiba Medical Systems Europe). Broad band curved array transducers (3.5 MHz and 6MHz) were used to perform routine fetal biometry measurements and measurements of subcutaneous tissue thickness. Hadlock`s formula was used to estimate fetal weight [13,14]. Abdominal circumference (AC) was measured in a cross sectional view of the upper abdomen visualising the stomach, one pair of ribs and the vertebrae as well as the umbilical vein at the level of the portal sinus. Measurements of anterior abdominal wall (AAW) thickness were made 2 cm from the umbilical vein from the skin surface to the anterior aspect of the liver (Fig 1). Adequate magnification was defined as the abdominal circumference filling the entire screen area. Soft tissue and subcutaneous fat tissue of the upper arm (HumST, HumSC) at the major tubercle level and thigh (FemST, FemSC) at greater trochanter level were measured in longitudinal views between skin surface and lean muscle below (Fig 2). Actual birth weights above the 90th percentile for gestational age were considered macrosomic [16].

Statistical analysis

AGA and macrosomic fetuses were compared using the non-parametric Wilcoxon-Mann-Whitney-U test and the Fisher-Freeman-Halton test as shown in Table 1 (demographic data) and Table 2 (sonographic measurements).

Receiver operating characteristic (ROC) curves were constructed by plotting the sensitivity of the listed parameters against their specificity. The curves illustrate sensitivity and specificity of the parameters for the detection of macrosomia at cut-off points defined in Table 3. The areas under the curve (AUC) were calculated and compared as described by DeLong et al. [17].

Total AGA Macrosomic p-value (AGA vs
Macrosomic)
Patients (n) 138 119 19
Diet-controlled GDM 39 (100%) 36 (92%) 3 (8%)
Insulin-dependent GDM 69 (100%) 62 (90%) 7 (10%)
Diabetes mellitus Type 1 20 (100%) 13 (65%) 7 (35%)
Diabetes mellitus Type 2 10 (100%) 8 (80%) 2 (20%)
Maternal age (years) 34.99± 4.36 35.08 ± 4.30 34.42 ± 4.77 0.35
Gravidity 2.40 ± 1.49 2.34 ± 1.45 2.78 ± 1.71 0.284
Parity 1.11 ± 1.29 1.04 ± 1.22 1.57 ± 1.64 0.161
Body Mass Index (BMI) 30.28 ± 6.86 30.32 ± 6.71 30.02 ± 7.93 0.699
Gest. age (wks) examination 32.84 ± 2.83 32.57 ± 3.4 32.85 ± 3.42 0.926
Gest. age (wks) birth 38.55 ± 1.57 38.73 ± 1.41 37.42 ± 2.03 0.003
Male newborns 92 (62%) 77 (65%) 10 (53%) 0.318
 Female newborns 57 (38%) 42 (35%) 9 (47%)

Table 1: Demographic data (mean values, ± standard deviation) in pregnancies resulting in appropriate-for- gestational age (AGA) and macrosomic newborns according to the Voigt growth chart [16].

 

 

AGA Macrosomic P value (AGA vs Macrosomic)
Patients (n) 119 19
Neonatal birth weight (g) 3405  ± 356.48 (n= 119) 4049 ± 505.16  (n= 19) 0.0001
HC (mm) 317.38 ± 21.95 (n= 119) 322.41 ± 30.27  (n= 19) 0.219
AC (mm) 287.81 ± 31.21 (n= 119) 302.33 ± 40.8 (n= 19) 0.057
AAW (mm) 5.28 ± 1.25 (n= 100) 5.96 ± 2.07 (n= 17) 0.219
FL (mm) 64.54 ± 5.90 (n= 119) 65.32 ± 8.2 (n=19) 0.386
Fem ST (mm) 9.31 ± 2.41 (n= 117) 10.68 ± 3.47 (n= 19) 0.111
Fem SC (mm) 5.11 ± 1.66 (n= 116) 5.91 ± 2.0 (n= 19) 0.119
Fem SC / FL ratio 0.07 ± 0.02 (n= 116) 0.08 ± 0.02 (n= 19) 0.07
HL (mm) 57.40 ± 4.66 (n= 99) 60.17 ± 5.83 (n=12) 0.034
Hum ST (mm) 7.8 ± 2.06 (n= 111) 9.74 ± 3.34 (n= 19) 0.023
Hum SC (mm) 3.43 ± 1.11 (n= 111) 4.40 ± 1.66 (n= 19) 0.002
Hum SC / HL ratio 0.05 ± 0.01 (n= 94) 0.12 ± 0.16 (n=12) 0.0003

Table 2: Mean values (± standard deviation) of neonatal birth weight and fetal biometric measurements. Head circumference (HC), abdominal circumference (AC), anterior abdominal wall thickness (AAW), femur length (FL), soft tissue (FemST) and subcutaneous fat tissue thickness thigh (FemSC), ratio of subcutaneous fat tissue thickness to femur length (FemSC/FL), soft tissue (HumST) and subcutaneous fat tissue thickness upper arm (HumSC), ratio of subcutaneous fat tissue thickness to humerus length (HumSC/HL) in appropriate-for-gestational age (AGA) and macrosomic offspring according to the Voigt growth chart [16].

Spearman`s rank correlation coefficient was used to assess the correlation of gestational age and birth weight percentiles with the fetal biometric measurements: abdominal circumference (AC), anterior abdominal wall (AAW) thickness, soft tissue thickness of the upper arm and thigh (HumST, FemST), subcutaneous fat tissue thickness of the upper arm and thigh (HumSC, FemSC), and the ratios of subcutaneous fat tissue thickness to humerus and femur length (HumSC/HL and FemSC/FL).

To adjust for correlations with gestational age, we utilized residuals obtained from linear regression models, where gestational age was used as an independent variable and measurement of AC, AAW thickness, HumST, FemST, HumSC, and FemSC thickness, and the ratios HumSC/HL, and FemSC/FL were the corresponding dependent variables. Residuals were checked for normality using the Kolmogorov-Smirnoff-Lilliefors-test.

All tests were two-sided and a p-value of ≤ 0.05 was considered statistically significant.

Statistical analyses were carried out using BiAS. Software (Version 11.04, Epsilon-Verlag, Darmstadt, Germany) and R Software (Version 3.2.4, The R Foundation for Statistical Computing, Vienna, Austria).

Results

Of the 138 patients examined mean maternal age at the time of investigation was 34.9 years (range 25-44 years), mean gestational age at examination was 32.8 weeks (range 25 – 38 weeks), mean gestational age of the new-borns was 38.5 gestational weeks (range 33 – 41 weeks) and mean birth weight was 3416 g (range 1840 g – 4810 g). At the time of the examination 39 GDM patients were able to achieve normal glucose levels with dietary counselling, 69 GDM patients were being treated with insulin, 20 patients had PGDM Type 1 and 10 had PGDM Type 2.

 

Table 1 shows demographic data for pregnancies resulting in macrosomic and appropriate-for-gestational age (AGA) new-borns. A higher percentage of macrosomic fetuses (35%) was found in patients with diabetes mellitus Type 1 (7 out of 20 fetuses), than in patients with GDM. A similar tendency albeit based on small patient numbers was found for diabetes mellitus Type 2 (20%, 2 out of 10 fetuses). Macrosomic newborns were younger at birth than AGA new-borns (37.4 vs 38.7 weeks, p= 0.003). Pregnancies resulting in macrosomic and normal-weight new-borns did not differ with regard to maternal age, gravidity, parity, BMI, gestational age at examination or sex of the new-borns.

 

Table 2 compares fetal biometric measurements and birth weights of macrosomic fetuses with those of AGA fetuses and new-borns. In the present study macrosomic fetuses had a mean birth weight of 4049 g and a mean abdominal circumference (AC) of 302.3 mm compared to 3405 g and 287.8 mm in AGA fetuses (p= 0.0001 and p= 0.057, respectively). A significant difference between macrosomic and AGA fetuses was also shown with respect to fetal humerus length (HL), HumST and HumSC thickness as well as the ratio HumSC/HL (60.1 mm vs 57.4 mm, p= 0.034; 9.74 vs 7.8 mm, p=0.023; 4.4 mm vs 3.43 mm, p= 0.002; 0.12 mm vs 0.05 mm, p= 0.0003, respectively).

 

The ROC curves for AC, AAW thickness, FemST, FemSC, the ratio FemSC/FL, HumST, HumSC and the ratio HumSC/HL are presented in Figures 3, 4 and 5, respectively. ROC curves were constructed by plotting sensitivity against specificity. The areas under the ROC curve, their respective p-values and the calculated cut-off points for fetal biometric measurements to predict fetal macrosomia, are listed in Table 3. The best possible prediction efficacy would be displayed by an area under the curve of 1.0 reflecting 100% sensitivity and 100% specificity.

 

Of all our measurements shown in Table 3, the HumST (AUC=0.729, p=0.009), the HumSC thickness (AUC=0.777, p < 0.0001) and the ratio HumSC/HL (AUC=0.787, p<0.0001) presented the best detection efficacy followed by AC (AUC=0.752, p=0.001), Fem SC (AUC 0.736, p=0.002), and the ratio Fem SC/FL (AUC=0.729, p=0.001).

 

 

Fetal biometric measurements

 

AUC

 

± SD (AUC)

 

95%-CI (AUC)

 

p-value (AUC)

 

Cut-off point

 

Sensitivity (95% CI)

 

Specificity (95% CI)

 

AC (mm)

 

0.752

 

0.080

 

0.594 – 0.909

 

0.001

 

312.7

 

0.47 (0.24-0.71)

 

0.79 (0.71-0.86)

 

AAW (mm)

 

0.621

 

0.114

 

0.396 – 0.846

 

0.290

 

7.4

 

0.35 (0.14-0.616)

 

0.94 (0.87-0.97)

 

FemST (mm)

 

0.691

 

0.100

 

0.494 – 0.887

 

0.056

 

12.5

 

0.42 (0.20-0.66)

 

0.90 (0.83-0.95)

 

FemSC (mm)

 

0.736

 

0.077

 

0.585 – 0.888

 

0.002

 

6.45

 

0.36 (0.16-0.61)

 

0.84 (0.76-0.90)

 

FemSC/FL

 

0.729

 

0.073

 

0.585 – 0.874

 

0.001

 

0.085

 

0.57 (0.33-0.79)

 

0.67 (0.57-0.75)

 

HumST (mm)

 

0.729

 

0.089

 

0.555 – 0.904

 

0.009

 

9.85

 

0.47 (0.24-0.71)

 

0.83 (0.74-0.89)

 

HumSC (mm)

 

0.777

 

0.061

 

0.657 –0.896

 

< 0.0001

 

3.55

 

0.68 (0.43-0.87)

 

0.70 (0.60-0.78)

 

HumSC/HL

 

0.787

 

0.069

 

0.650 – 0.924

 

< 0.0001

 

0.065

 

0.83 (0.51-0.97)

 

0.82 (0.72-0.89)

 

Table 3: Areas under the ROC curves (AUC) as described by DeLong et al [17], ± standard deviations, 95%

confidence intervals (CI), p-values and optimal cut-off points (sensitivity and specificity values with 95% confidence intervals) for the prediction of macrosomia with fetal biometric measurements: abdominal circumference (AC), anterior abdominal wall thickness (AAW), soft tissue (FemST) and subcutaneous fat tissue thickness thigh (FemSC), ratio of subcutaneous fat tissue thickness to femur length (FemSC/FL), soft tissue (HumST), and subcutaneous fat tissue thickness upper arm (HumSC) and ratio of subcutaneous fat tissue thickness to humerus length (HumSC/HL).

The fetal biometric measurements AC, AAW thickness, FemST, FemSC, HumST and HumSC thickness showed significant correlation to gestational age (Spearman`s correlation coefficient rho 0.82, 0.52, 0.45, 0.43, 0.43, 0.37 respectively; p < 0.0001 in all cases) whereas the ratios of subcutaneous fat tissue thickness to length of humerus and femur showed weak correlation (Spearman`s correlation coefficient rho 0.17, p= 0.057; rho 0.20, p= 0.014). We examined the correlation between fetal biometric measurements and birth weight percentiles [16] instead of neonatal birth weight at term to assure that preterm births that were macrosomic in their respective week of delivery were also included in the calculation, and found a significant correlation as well (Spearman`s correlation coefficient rho: AC 0.35, p < 0.0001; FemST 0.3, p=0.0002; FemSC 0.35, p < 0.0001; FemSC/FL 0.32, p=0.0001, HumST 0.29, p=0.0004; HumSC 0.34, p<0.0001; HumSC/HL 0.38, p<0.0001).

To adjust for correlations with gestational age, we utilized residuals obtained from linear regression models, with gestational age as an independent variable and the fetal biometric measurements as corresponding dependent variables. Normality could be assumed in AC, AAW thickness, HumST, FemST, FemSC, and the ratio of FemSC/FL. In HumST and FemST normality of the residuals was obtained after log-transformation, and in HumSC after log-log-transformation. However, no normality was demonstrated by this method for the ratio of HumSC/HL. Hence, the residuals could be interpreted as age-adjusted percentiles. The ROC-curves that were constructed using age-adjusted residuals of AC, HumST and FemSC showed larger areas under the curve (AC: AUC=0.838, SD=0.057, 95%-confidence interval (0.726-0.949), p=0.137; HumST: AUC=0.736, SD=0.088, 95%-confidence interval (0.563-0.909), p=0.87; FemSC: AUC=0.765, SD=0.071, 95%-confidence interval (0.626-0.904), p=0.539) than when using the non-adjusted residuals, but the difference was not significant.

Discussion

Our study compared the usefulness of several published parameters in predicting fetal macrosomia in diabetic pregnancies and found that fetal soft tissue thickness and subcutaneous fat tissue thickness of the upper arm (HumST and HumSC) as well as the ratio of subcutaneous fat tissue thickness to humerus length (HumSC/HL) showed significant differences between macrosomic and AGA fetuses. Calculation of the area under the ROC curve confirmed these three parameters as the strongest predictors of fetal macrosomia when compared to abdominal circumference (AC), anterior abdominal wall (AAW) thickness, soft tissue thickness of the thigh (FemST), subcutaneous fat tissue thickness of the thigh (FemSC), and the ratio of subcutaneous fat tissue thickness to femur length (FemSC/FL) in diabetic pregnant women.

Several previous studies have sought to find reliable indicators of fetal macrosomia (Table 4).

Sod et al. [18] and Landon et al. [19] reported that macrosomic neonates born to diabetic mothers show disproportionately larger shoulders, trunk and upper arms. This interesting observation agrees with our finding that macrosomic fetuses have longer humeri than AGA fetuses whereas the length of femora in both groups were similar. In the consideration which fetus might encounter perinatal complications, subcutaneous fat tissue thickness relative to the length of long bones is indeed an important parameter [12]: a short fetus with excessive fat deposition as frequently found in diabetes may be more prone to the complications than a long fetus of the same weight. Therefore, we tested the possibility of enhancing the sensitivity of sonographic soft tissue and subcutaneous fat tissue measurements by calculating the ratio of subcutaneous fat tissue to length of humerus and femur and found this parameter to be more sensitive than subcutaneous fat or bone length alone. Similar calculations were made by Santaloya-Forgas et al. [20] in normal pregnancies. They reported that the ratio of Fem SC/FL was an improvement in the detection efficacy for LGA fetuses over AGA fetuses with respect to AC and estimated fetal weight (EFW). In agreement with this group we find that the ratio of subcutaneous fat to the respective long bone improves the recognition of macrosomic fetuses although we find the ratio of HumSC/HL superior to the ratio of FemSC/FL reported by Santaloya-Forgas et al [20].

Chauhan et al [8] examined HumST and the ratio of FemSC/FL and found them to be inferior to sonographically estimated fatal weight or clinical  predictions in identifying macrosomic fetuses. The difference between their observations and ours may be explained by different measuring sites (Table 4) or different population characteristics (diabetic vs non-diabetic pregnancies), or their definition of macrosomic fetuses at an expected birth weight of >4000g, whereas we defined macrosomia as being above the 90th percentile. As shown in Voigt`s growth chart [16] birth weights below 4000g can still be above the 90th percentile for age at birth in preterm delivery. Also, the effective treatment of diabetes during pregnancy may have led to reduced prevalence of macrosomia at birth in our study.

Table 4: Comparison of the listed studies with respect to diabetes, gestational age at examination, measurements made and sites of subcutaneous fat tissue and soft tissue measurements. Cheek-to-cheek diameter (CCD), abdominal circumference (AC), anterior abdominal wall thickness (AAW), subcutaneous fat tissue thickness of the upper arm (HumSC) and thigh (FemSC), soft tissue thickness of the upper arm (HumST) and thigh (FemST)) and ratios calculated to predict macrosomia.

Conclusion

Our data show that fetal subcutaneous fat tissue thickness of the upper arm and the ratio of subcutaneous fat tissue thickness to humerus length are superior predictors of fetal macrosomia in diabetic pregnancies when compared to abdominal circumference, anterior abdominal wall thickness, soft tissue thickness and subcutaneous tissue thickness of the femur and the ratio of subcutaneous tissue thickness to femur length. Currently, established formulas for estimation of fetal weight do not include the disproportionately greater amount of subcutaneous fat tissue, which is so important in fetuses of diabetic mothers [18,19]. The consideration of subcutaneous fat may improve the prediction of fetal macrosomia in diabetic pregnancies.

Fig. 1: Sonographic measurement of anterior abdominal wall thickness (AAW) 2 cm from the umbilical vein.

Fig. 2: Sonographic measurement of subcutaneous fat tissue thickness of the thigh (greater trochanter level).

Fig. 3: Receiver operating characteristic curves for abdominal circumference (AC) and anterior abdominal wall thickness (AAW) for the prediction of macrosomia.

Fig. 4: Receiver operating characteristic curves for soft tissue thickness thigh (FemST), subcutaneous fat tissue thickness thigh (FemSC) and the ratio of subcutaneous fat tissue to femur length for the prediction of macrosomia.

Fig. 5: Receiver operating characteristic curves for soft tissue thickness upper arm (HumST), subcutaneous fat tissue thickness upper arm (HumSC) and the ratio of subcutaneous fat tissue to humerus length for the prediction of macrosomia.

References

 

  1. International Association of Diabetes and Pregnancy Study Groups Consensus Panel. International Association of Diabetes and Pregnancy Study Groups Recommendations on the Diagnosis and Classification of Hyperglycemia in Pregnancy. Diabetes Care. 2010. 33(3).
  2. Farah N, Hogan J, O`Dwyer V, Stuart B, Kennelly M, Turner MJ. Influence of maternal glycemia on intrauterine fetal adiposity distribution after a normal oral glucose tolerance test at 28 weeks gestation. Exp Diabetes Res. 2011, 2011: 951203.
  3. Higgins MF, Russel NM, Mulcahy CH, Coffey M, Foley ME et al. Fetal anterior abdominal wall thickness in diabetic pregnancy. Eur J Obstet Gynecol Reprod Biol. 2008, 140 (1): 43-47.
  4. Wroblewska-Seniuk K, Wender-Ozegowska E, Szczapa J. Long-term effects of diabetes during pregnancy on the offspring. Pediatr Diabetes 2009, 10: 432-440.
  5. 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(3): e290-296.
  6. Desoye G, Nolan CJ. The fetal glucose steal: an underappreciated phenomenon in diabetic pregnancy. Diabetologia. 2016, 59(6): 1089-1094.
  7. Whitelaw A. Subcutaneous fat in newborn infants of diabetic mothers: An indication of quality of diabetic control. Lancet 1977, 1(8001): 15-8
  8. Chauhan SP, West DJ, Scardo JA, Boyd JM, Joiner J et al. Antepartum detection of macrosomic fetus: clinical versus sonographic, including soft-tissue measurements. Obstet Gynecol. 2000, 95(5): 639-642.
  9. Larciprete G, Valensise H, Vasapollo B, Novelli GP, Parretti E et al. Fetal subcutaneous tissue thickness (SCTT) in healthy and gestational diabetic pregnancies. Ultrasound Obstet Gynecol. 2003; 22(6): 591-597.
  10. West DL, Brans YW. Maternal diabetes and neonatal macrosomia. Dynamic skinfold thickness measurements. Am J Perinatol. 1986, 3(1): 9-12.
  11. Brans YW, Sumners JE, Dweck HS, Cassady G. A non-invasive approach to body composition in the neonate: dynamic skinfold measurements. Pediatr Res. 1974, 8(4): 215-222.
  12. Rotmensch S, Celentano C, Liberati M, Malinger G, Sadan O et al. Screening efficacy of the subcutaneous tissue width/femur length ratio for fetal macrosomia in the non-diabetic pregnancy. Ultrasound Obstet Gynecol. 1999, 13(5): 340-344.
  13. Hadlock FP, Harrist RB, Carpenter RJ, Deter RL, Park SK. Sonographic estimation of fetal weight. The value of femur length in addition to head and abdomen measurements. Radiology. 1984, 150(2): 535-540.
  14. Hadlock FP, Harrist RB, Sharman RS, Deter RL, Park SK. Estimation of fetal weight with the use of head, body, and femur measurements – A prospective study. Am J Obstet Gynecol. 1985, 151(3): 333-337.
  15. Bahlmann F, Merz E, Weber G, Wellek S, Engelhardt O. Transvaginal ultrasound biometry in early pregnancy—a growth model. Ultraschall Med. 1997, 18(5): 196-204.
  16. Voigt M, Rochow N, Schneider KT et al. New percentile values for the anthropometric dimensions of singleton neonates: analysis of perinatal survey data of 2007–2011 from all 16 states of Germany; Z Geburtshilfe Neonatol. 2014, 218(5): 210-217.
  17. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988, 44(3): 837-845.
  18. Sood AK, Yancey M, Richards D. Prediction of fetal macrosomia using humeral soft tissue thickness. Obstet Gynecol. 1995, 85(6): 937-940.
  19. Landon MB, Sonek J, Foy P, Hamilton L, Gabbe SG. Sonographic measurement of fetal humeral soft tissue thickness in pregnancy complicated by GDM. Diabetes. 1991, 40 Suppl 2: 66-70.
  20. Santaloya-Forgas J, Meyer WJ, Gauthier DW, Kahn D. Intrapartum fetal subcutaneous tissue/femur length ratio: an ultrasonographic clue to fetal macrosomia. Am J Obstet Gynecol. 1994, 171(4): 1072-1075.

Be the first to comment on "Prediction of Fetal Macrosomia in Diabetic Pregnancies by Sonographic Measurement of Fetal Subcutaneous Fat Thickness"

Leave a comment

Your email address will not be published.


*


Select Language