Fetal Interventricular Septum Volume Evaluated by Three-Dimensional Ultrasound Using Spatiotemporal Image Correlation and Virtual Organ Computer-Aided Analysis in Fetuses From Pre-Gestational Diabetes Mellitus Pregnant Women.

BACKGROUND: To assess the interventricular septum (IVS) volume of fetuses from pre-gestational diabetes mellitus (DM) pregnant women by 3-dimensional ultrasound using spatiotemporal image correlation (STIC) and virtual organ computer-aided analysis (VOCAL) methods.
METHODS: This was a prospective cross-sectional study of 45 fetuses from pre-gestational DM and 45 fetuses from healthy pregnant women (controls). Only singleton pregnancies between 20 and 34 + 6 weeks of gestation were included. The fetal IVS volumes were obtained off-line using STIC and VOCAL methods. To analyze differences among variables, the Student's t-test and Mann-Whitney U test were used. The correlation among continuous variables was determine using Spearman's correlation test (r).
RESULTS: The median of fetal IVS volume was significantly higher in pre-gestational DM than in healthy pregnant women (0.3 cm3 vs. 0.2 cm3, p = 0.032). A strong positive correlation was observed between fetal IVS volume and gestational age at the time of ultrasound examination (r = 0.75, R² = 0.48, p < 0.0001) and between fetal IVS volume and estimated fetal weight (r = 0.63, R² = 0.37, p < 0.0001). No significant correlation was noted between fetal IVS volume and glycated hemoglobin levels (r = -0.16, R² = 0.01, p = 0.540) in the pre-gestational DM pregnant women.
CONCLUSIONS: Significant differences were observed in fetal IVS volumes between pre-gestational and healthy mothers, with higher values in the fetuses of pre-gestational DM pregnant women.

INTRODUCTION

Diabetes mellitus (DM) is a health problem that affects 1 (16.8%) in 6 of all pregnancies worldwide; however, its prevalence may vary depending on the population characteristics and diagnostic criteria. Pregnant women with DM and their fetuses are exposed to long-term consequences with transgenerational effects such as children with a higher risk of obesity and its related complications.1)

Furthermore, DM is associated with fetal cardiac defects and a higher risk of stillbirth. A wide variety of factors can result to this adverse outcome. Fetal cardiomyopathy is reported to be an important etiology of intrauterine fetal demise due to the findings on fetal autopsy examinations in DM, compared with controls.2) A great variety of studies have demonstrated the prevalence of myocardial hypertrophy in fetuses of mothers with DM in comparison to non-DM pregnancies presenting a risk of heart failure in utero.3)4)5)

Classically, myocardial thickness can be assessed by the measurements of ventricular walls and interventricular septum (IVS) using M-mode and a 2-dimensional (2D) ultrasound. Reference ranges of fetal interventricular thickness in low-risk population using the 2D ultrasound have been published. The reference ranges have enabled important assessments for conditions at risk of myocardial hypertrophy such as DM.6)7) In addition to the reference ranges for the IVS expressed in percentiles, reference ranges in Z-scores were also published.8)

However, the 2D technique may have limited sensitivity, especially for mild cases. The 3-dimensional (3D) ultrasound with spatiotemporal image correlation (STIC) and virtual organ computer-aided analysis (VOCAL) methods allows a better identification of the fetal heart ventricular wall edges. Studies demonstrate that 3D ultrasound provides more accurate measurements of the fetal IVS thickness which may enable earlier diagnosis of myocardial hypertrophy.9)10) Reference ranges for fetal IVS volumes in low-risk pregnant women were obtained using 3D ultrasound with STIC and VOCAL methods, which could enable early diagnosis of myocardial hypertrophy.11) However, to the best of our knowledge, no studies have evaluated the fetal IVS volumes in pre-gestational DM mothers.

Therefore, this study aimed to evaluate the fetal IVS volumes in pre-gestational DM and healthy pregnant women using 3D ultrasound with STIC and VOCAL methods.

METHODS

We conducted a prospective cross-sectional study to assess fetal IVS volumes using 3D ultrasound with STIC and VOCAL methods in the fetuses of pre-gestational DM and healthy pregnant women with a gestational age ranging from 20 to 34 + 6 weeks of gestation. The study protocol was approved by the Research Ethics Committee of the Federal University of São Paulo (UNIFESP), (CAE: 87111116.4.0000.5505). On October 2016 and July 2019, pregnant women were recruited from the Diabetes Sector of the Department of Obstetrics at the UNIFESP in São Paulo-SP, Brazil. Consent forms were obtained from all the participants.

The data was divided into 2 groups, i.e., pre-gestational DM, and controls. The first group comprised fetuses of pregnant women with pre-gestational DM (types 1, 2 or MODY—Maturity-Onset Diabetes of the Young), and the second group comprised normal fetuses from healthy mothers. The inclusion criterion was singleton pregnancies with gestational age being between 20 and 34 + 6 weeks of gestation. The gestational age was based on the last menstrual period and confirmed by an ultrasound examination performed in the first trimester using the crown-rump length measurement. The exclusion criteria incorporated fetuses whose mothers had other comorbidities besides pre-gestational DM, fetuses with cardiac, extra cardiac anomalies or chromosomal abnormalities, an estimated fetal weight (EFW) of less than 10% for the gestational age and 3D fetal cardiac volumes of inadequate quality i.e. fetal movements, acoustic shadows and mother breathing.

All pregnant women included in this study were examined once using the Voluson E8 apparatus (General Electric Medical System, Zipf, Austria) with a convex (3.5–7.0 MHz) and volumetric (RAB 4-8 L) probes. Fetal echocardiograms were performed along with the ultrasound examinations which included the morphology, amniotic fluid assessment and fetal biometry. All fetal echocardiograms were performed by one specialist in perinatal cardiology with expertise in 3D ultrasound. Initially, a systematic evaluation of the fetal heart using 2D ultrasound with color Doppler had been obtained in all patients. Subsequently, the 3D cardiac volumes were acquired at the 4-chamber view level, using the STIC method. The 3D cardiac volumes were obtained during fetal rest and maternal apnea, according to the 3D scan acquisition standard, proposed by Gonçalves et al.12) The following parameters were evaluated: speed time ranging from 7.5 to 15 seconds and an open angle ranging from 20º to 25º and 35º to 40º in the second and third trimesters, respectively. From each examination, three 3D cardiac volumes were obtained and stored, according to the method proposed by Paladini.13) Previously, only one of three 3D cardiac volumes was selected by a single examiner to be analyzed off-line using maximum quality 3D images.

The axial plane of the multiplanar mode (plane A) was selected to perform the off-line measurements of the fetal IVS using the 4D view version 10.0 (General Electric Medical System). Sequentially, the cinellop mode reduced the velocity to 6% and retrieved it to obtain the IVS at the ventricular end of both ventricles (identified by the larger ventricular lumen). Sepia color selection and gamma curve adjustments were made to obtain the best contrast between the edges of the inner walls of the ventricles and their internal content. The 3D cardiac volume was rotated around the “z” axis to position the apex of the fetal heart to 12 o’clock. The caliper was positioned on the inner side of the atrioventricular valves (crux of the heart) and on the mid-point of the fetal heart apex, being respectively defined as the upper and the lower border of fetal IVS. The VOCAL button was activated, and the calipers were positioned to the upper and lower borders of the fetal IVS. Sequentially, 6 sequential planes at a 30° rotation were manually delimited. After the final plan, the software provided the fetal IVS volume at the end of the ventricular diastole (Figure 1).

Figure 1

Fetal IVS volume using STIC and VOCAL methods in the fetus of a pregnant woman with pre-gestational diabetes mellitus at 29 + 5 weeks of gestation. Fetal IVS volume was obtained from the end-diastole and was determined based on the maximum diameter of the ventricles. The calipers were positioned on the inner side of the atrioventricular valves (crux of the heart) and on the fetal heart’s apex mid-point of the, being respectively defined as the IVS upper and lower borders. After 6 sequential planes, the VOCAL methods provided the fetal IVS volume and the reconstructed 3D image.

D: dimensional, HR: heart rate, IVS: interventricular septum, STIC: spatiotemporal image correlation, VOCAL: virtual organ computer-aided analysis.

These clinical variables were collected; maternal age, number of previous pregnancies, parity, miscarriage, gestational age at the time of the ultrasound examination, EFW, and the fetal heart rate. Glycated hemoglobin (HbA1c) level was obtained from pre-gestational DM pregnant women during the first trimester.

A power analysis was performed to calculate the sample size on the basis of the Cohen effect of 0,5 to achieve a power of 80% and an alpha of 5% to detect the differences in the evaluated parameters. Using the software G 3.1, the results suggested a sample size of 44 fetuses from each group. The data were stored in an Excel spreadsheet 2010 (Microsoft Corp., Redmond, WA, USA) using the Prisma Graph Pad version 7.0 (Graph Pad Software, San Diego, CA, USA). The quantitative variables were initially subjected to the D'Agostino & Pearson normality test. The parametric distribution variables were presented from the mean and standard deviations. The nonparametric distribution variables were presented from the median, minimum and maximum values. To assess the difference between variables, the Student’s t-test and Mann-Whitney U test were used. To analyze the correlation between continuous variables, Spearman's correlation test (r) was used. The level of significance being (p) < 0.05.

RESULTS

Ninety-five singleton pregnancies were eligible, however five of them were excluded due to a poor quality of the 3D cardiac volume acquisition. Finally, 45 pre-gestational DM and 45 healthy mothers were included. There was significant effect of pre-gestational DM on maternal age (p = 0.0195), number of pregnancies (p < 0.0001), number of deliveries (p = 0.0032), number of miscarriages (p = 0.0009) and fetal IVS volume (p = 0.0032). Mean maternal age was significantly lower in patients with pre-gestational DM compared with controls (28.7 vs. 32 years, respectively). Pre-gestational DM pregnant women had significantly higher median number of pregnancies compared with controls. The number of pregnancies were 2.0 vs. 1.0, respectively, whereas the number of deliveries were 1.0 vs. 0.0, respectively. Significant differences were noted in fetal IVS volumes of pre-gestational DM compared with healthy mothers. The median value of the fetal IVS volume was significantly higher in pre-gestational DM than in controls (0.3 cm3 vs. 0.2 cm3). The clinical and ultrasonographic characteristics of the study population are described in Table 1. The equations for the fetal IVS volume according to the gestational age at the time of ultrasound examination, EFW, and HbA1 levels are shown in Table 2 and Figure 2.

Table 1

Clinical and ultrasonographic characteristics of the studied population

Maternal and ultrasonographic characteristics	Healthy (n = 45)				Pre-gestational diabetes mellitus (n = 45)				p-value
	Median	Min	Max	Mean ± SD	Median	Min	Max	Mean ± SD
Age (years)	33.0	19.0	43.0	32.0 ± 6.3	28.0	17.0	43.0	28.7 ± 6.6	0.0195*
Gestation	1.0	1.0	4.0	1.4 ± 0.8	2.0	1.0	6.0	2.5 ± 1.4	< 0.0001†
Parity	0.0	0.0	3.0	0.4 ± 0.8	1.0	0.0	3.0	1.0 ± 1.1	0.0032†
Miscarriage	0.0	0.0	1.0	0.0 ± 0.2	0.0	0.0	5.0	0.6 ± 0.1	0.0009†
Gestational age (weeks)	27.1	20.0	33.5	27.3 ± 4.1	27.5	20.2	34.6	27.6 ± 4.0	0.7283†
Glycated HB (%)					7.6	4.8	15.5	8.0 ± 2.6
EFW (g)	949.0	322.0	2,137.0	1,065.0 ± 530.0	1,143.0	333.0	2,689.0	1,230.0 ± 632.0	0.1926*
IVS (cm3)	0.2	0.0	0.5	0.2 ± 0.1	0.3	0.1	1.0	0.3 ± 0.2	0.0032†
FHR (bpm)	142.0	121.0	160.0	142.0 ± 10.8	144.0	113.0	166.0	142.0 ± 12.4	0.9501†

EWF: estimated fetal weight, FRH: fetal heart rate, HB: hemoglobin, IVS: interventricular septum, Max: maximum, Min: minimum.

*Student t-test; †Mann-Whitney U test.

Table 2

Equations for predicting fetal interventricular septum volume based on gestational age and estimated fetal weight, considering all cases included, and glycated HB value in pre-gestational diabetes mellitus pregnant women

Variables	Equation	R2	r	p-value
Gestational age (weeks)	Y = 0.02613 × X − 0.4781	0.48	0.75	< 0.0001
EFW (g)	Y = 0.0001512 × X + 0.05666	0.37	0.63	< 0.0001
Glycated HB (%)	Y = −0.007878 × X + 0.3638	0.01	−0.16	0.54

Spearman correlation test, p < 0.05.

EWF: estimated fetal weight, HB: hemoglobin.

Figure 2

Correlations among fetal IVS volume, gestational age, EFW and glycated HB in healthy and pre-gestational diabetes mellitus mothers. (A) The correlation between the gestational age at the time of the ultrasound examination and fetal IVS volume in both pre-gestational diabetes mellitus (red dots) and the healthy (blue dots) mothers. (B) The correlation between EFW and fetal IVS in both pre-gestational diabetes mellitus (red dots) and healthy (blue dots) mothers. (C) The correlation between glycated hemoglobin (glycated HB) and fetal IVS among mothers with pre-gestational diabetes mellitus (red dots). Spearman's correlation test, p < 0.05.

EFW: estimated fetal weight, HB: hemoglobin, IVS: interventricular septum.

A strong positive correlation was observed between the fetal IVS volume and the gestational age at the time of the ultrasound examination (r = 0.75, R2 = 0.48, p < 0.0001) (Figure 2A), and the EFW (r = 0.63, R2 = 0.37, p < 0.0001) (Figure 2B). There was no significant correlation between fetal IVS volume and HbA1 level (r = −0.16, R2 = 0.01, p = 0.540) in fetuses from pre-gestational DM pregnant women (Figure 2C).

In controls, fetal IVS volume increased by 0.0212 cm3 (Y = 0.0212 × X − 0.3838, p < 0.0001, R2 = 0.65, r = 0.79) per week. In pre-gestational DM fetuses, the one-week increase of the fetal IVS volume was by 0.0298 cm3 (Y = 0.0298 × X − 0.542, p < 0.0001, R2 = 0.48, r = 0.69) (Figure 3). Regarding the correlation between the EFW and the fetal IVS volume, an increase in one gram of the EFW increases the fetal IVS volume by 0.00012 cm3 (Y = 0.00012 × X + 0.04505, p < 0.0001, R2 = 0.54, r = 0.73) in controls and by 0.00015 cm3 (Y = 0.00015 × X + 0.09542, p = 0.0001, R2 = 0.31, r = 0.51) of the fetal IVS volume in the pre-gestational DM group (Figure 4).

Figure 3

The correlation between the gestational age at the time of the examination and fetal IVS volume was stronger in healthy (blue dots, A) than pre-gestational diabetes mellitus mothers (red dots, B). Spearman's correlation test, p < 0.05.

IVS: interventricular septum.

Figure 4

Correlations between fetal IVS volume and EFW in healthy and pre-gestational diabetes mellitus mothers. (A) The correlation between EFW and fetal IVS volume among healthy (blue dots). (B) The correlation between EFW and fetal IVS volume among mothers with pre-gestational diabetes mellitus (red dots). Spearman's correlation test, p < 0.05.

EFW: estimated fetal weight, IVS: interventricular septum.

DISCUSSION

The prevalence of fetal hypertrophic myocardiopathy, conotruncal anomalies, ventricular septal defects, and other cardiac malformations is increased in mothers with DM and is associated with higher rates of perinatal morbidity and mortality.2)14)15) The hypertrophy of IVS is the most common structural cardiac anomaly in fetuses from mothers with DM, and with a deleterious risk on the fetal heart function. In this scenario, several studies have been published including ultrasound/echocardiographic fetal cardiac functional parameters by M-mode, Doppler, 2D and 3D ultrasound in the fetuses of mothers with DM.4)16)17)18)19) Recently, a metanalysis reviewed 39 studies on fetal IVS thickness and Doppler cardiac functional parameters focusing on analyzing the effects of DM on the fetal heart. In regard to the 2D ultrasound and Doppler cardiac function parameters in this review, the E/A ratio was decreased in the DM group. However, no differences were observed in the myocardial performance index between DM and control groups. Moreover, fetal cardiac hypertrophy was more prevalent in pre-gestational and gestational mothers with DM than healthy mothers, mainly during the third trimester.3)

It has been well established that DM is associated with the fetal myocardial hypertrophy. The measurements of the fetal IVS are usually performed at the 4-chamber view using 2D ultrasound and M-mode.20)21) However, since the irregular shape of the fetal IVS, its measurements may be more accurately performed by a 3D ultrasound other than by a conventional 2D ultrasound. In this scenario, Rolo et al.11) showed that a 3D ultrasound with STIC and VOCAL methods is a reproducible strategy and offers a positive correlation between the fetal IVS volume and the gestational age. In the current study, fetal IVS volume was higher in DM group with a positive correlation between gestational age and fetal IVS volume in both groups.

Reference ranges for fetal heart atrial and ventricular walls volumes using 3D ultrasound with STIC and VOCAL methods have previously been published.22)23) Bravo-Valenzuela et al.19) evaluated the fetal atrial and ventricular volumes of 106 fetuses between 20 and 34 weeks of gestation by 3D ultrasound with STIC and VOCAL methods. In this study, the ventricular volumes were similar both in the DM and control groups. The left atrial volumes were significantly lower in the pregnant women with pre-gestational DM compared with controls (0.62 mL in DM vs. 0.68 mL in control, p < 0.001). Fetal IVS volume was not, however, assessed in this study. Recently, Melo Júnior et al.24) assessed the total myocardial volume of the fetal heart by 3D ultrasound with STIC and VOCAL methods and did not observe a significant difference in the mean fetal myocardial volume between mothers with pre-gestational DM and healthy mothers (p = 0.64).

Rolo et al.11) using 3D ultrasound with STIC and VOCAL methods, established reference values for the fetal IVS volume in low-risk pregnant women. The fetal IVS volume ranged from 0.13 ± 0.03 cm3 at 18 weeks to 1.33 ± 0.37 cm3 at 33 weeks and 6 days of gestation. The authors suggested that the inclusion of the assessment of the fetal IVS volume in the routine might be useful for early detection of myocardial hypertrophy in mothers with DM. Similar to Rolo et al.,11) we also observed a positive correlation between fetal IVS volume and gestational age in pre-gestational DM compared to healthy pregnant women with higher mean values of fetal IVS volume in pre-gestational DM mothers.

Studies on fetal myocardial hypertrophy in gestational DM and its correlation with metabolic control have shown inconsistent results. Although, some studies have shown positive correlation between fetal IVS thickness and HbA1, since the myocardial hypertrophy can also occur in fetuses of pregnant woman with well-controlled DM.20)25)26)27) As per our research, no significant correlation between the IVS volume and HbA1 levels were observed in fetuses from pregnant women with pre-gestational DM. Recently, a study about cardiac dysfunction in pregnant women with gestational and pre-gestational DM showed that increased levels of biochemical markers such as B-type natriuretic peptide and troponin I obtained through an umbilical cord sample, were associated with increased fetal myocardial thickness.28)

The limitations of this study were that the patients were evaluated once and there was an exclusion of pregnant women with over 34 weeks of gestation. Therefore, after the gestational period no follow-up was done for cardiac hypertrophy progression and there were no perinatal outcomes available. Furthermore, heterogeneous samples including well-controlled and uncontrolled pre-gestational DM in pregnant women may have been interfered with in the results about the impact of metabolic disturbances in the fetuses.

In conclusion, fetal IVS volume assessed by 3D ultrasound with STIC with VOCAL methods showed significant differences in fetal IVS volumes between pre-gestational DM and healthy mothers between 20 to 34 + 6 weeks of gestation with higher values for pregnant women with pre-gestational DM.