Fetal Left Ventricular Mass Determination on 2-Dimensional Echocardiography Using Area-Length Calculation Methods

ORIGINAL RESEARCH Fetal Left Ventricular Mass Determination on 2-Dimensional Echocardiography Using Area-Length Calculation Methods Xiao-Zhi Zheng, PhD, MD, Bin Yang, PhD, MD, Jing Wu, MD Received May 9, 2013, from the Department of Ultrasound, Jinling Hospital, Nanjing University School of Medicine, Nanjing, China (X.-Z.Z., B.Y.); and Department of Ultrasound, Fourth Affiliated Hospital of Nantong University (First People s Hospital of Yancheng), Yancheng, China (X.-Z.Z., J.W.). Revision requested May 22, 2013. Revised manuscript accepted for publication June 24, 2013. We thank Jue wen Wang, BS, En hui Xia, BS, Xiao qin Huang, and Gen xiang Fan at the Department of Ultrasound, First People s Hospital of Yancheng, for technical assistance and helpful discussion. Address correspondence to Bin Yang, PhD, MD, Department of Ultrasound, Jinling Hospital, Nanjing University School of Medicine, 305 East Zhongshan Rd, 210002 Nanjing, Jiangsu, China. Abbreviations AC, abdominal circumference; BPD, biparietal diameter; EFW, estimated fetal weight; FL, femur length; GA, gestational age; HC, head circumference; LV, left ventricular; LVAd SAX EPI, LV epicardial short-axis area at the level of the papillary muscle tips at end diastole; LVAd SAX PM, LV endocardial short-axis area at the papillary muscle level at end diastole; LVAs SAX EPI, LV epicardial short-axis area at the level of the papillary muscle tips at end systole; LVAs SAX PM, LV endocardial short-axis area at the papillary muscle level at end systole; LVd mass, LV mass at end diastole; LVLd, LV long-axis length at end diastole; LV (d-s) mass, difference between the LVd mass and LVs mass; LVLs, LV long-axis length at end systole, apical; LVs mass, LV mass at end systole; 3D, 3-dimensional; 2D, 2-dimensional doi:10.7863/ultra.33.2.349 Objectives Fetal cardiac examination is an important part of fetal malformation screening. The purposes of this study were to describe the left ventricular (LV) mass in the second and third trimesters by 2-dimensional echocardiography using area-length calculation methods and to examine the clinical usefulness of this procedure in evaluation of gestational age (GA)- and fetal weight-related LV mass changes. Methods Five hundred healthy fetuses were divided into 2 groups (250 participants per group): second- and third-trimester groups. The estimated fetal weight (EFW) was computed according to the Hadlock formula (Radiology 1984; 150:535 540). The LV mass at end diastole (LVd mass) and LV mass at end systole (LVs mass) were measured, and the difference between the LVd mass and LVs mass [LV(d-s) mass], LVd mass/efw ratio, and LVs mass/efw ratio were calculated. Results The EFW, LVd mass, LVs mass, and LV(d-s) mass were all significantly greater in the third-trimester group than the second-trimester group (P <.05), whereas the LVd mass/efw and LVs mass/efw ratios did not differ between the groups (P >.05). The LVd mass, LVs mass, and LV(d-s) mass all significantly correlated with GA and weight (P <.001), but the LVd mass/efw and LVs mass/efw ratios did not (P >.05). Conclusions Two-dimensional echocardiography using area-length calculation methods can effectively provide measurements for LV mass and can sensitively indicate fetal weight- and GA -related changes in LV mass. Fetal cardiac mass measurement is a useful parameter for evaluation of fetal heart development. Key Words area-length; fetal weight; gestational age; left ventricular mass; left ventricular mass-to-fetal weight ratio; obstetric ultrasound; 2-dimensional echocardiography Fetal cardiac examination is an important part of fetal malformation screening. Structural and functional cardiac disorders and extracardiac factors, such as left ventricular (LV) noncompaction, hypertrophy, congenital heart disease, gestational diabetes, and maternal hypertension, often result in an abnormal ventricular mass. In addition to conventional cardiac dimensions and Doppler measurements, myocardial mass measurement should be performed during the cardiac examination because it may be used with other cardiac parameters to ascertain the severity and prognosis, for termination counseling, or to determine the nature and timing of interventions. 1 3 2014 by the American Institute of Ultrasound in Medicine J Ultrasound Med 2014; 33:349 354 0278-4297 www.aium.org

Sonography is currently the primary screening technique for imaging the fetus, 4 7 which has played an important role in evaluation of fetal organ development and detection of fetal organ abnormalities in utero. To date, M-mode, 2-dimensional (2D), 3-dimensional (3D), and 4-dimensional echocardiography and a series of new techniques, such as spatiotemporal image correlation, have been used to evaluate the ventricular mass. Among these methods, 2D echocardiography using area-length calculation methods is a simple, commonly used modality. The LV mass as measured by the 2D area-length method was close to the results of cardiac magnetic resonance imaging in adults, 8 and necropsy weights in mice, 9,10 but fetal LV mass determination with 2D area-length methods remains unknown. Fetal organs, including the heart, are rapidly developing and changing their structure and function week by week during pregnancy. Just as the estimated fetal weight (EFW) is an important parameter for evaluation of fetal development, the myocardial mass may also be an important parameter for evaluation of fetal heart development. Previous studies 1,2,11 found that both the fetal weight and ventricular mass increase with gestational age (GA), but the ratio of ventricular mass (diastole and systole) to EFW and whether it changes with GA remain unknown. In this study, we assessed the changes in LV mass in healthy fetuses in a 2D echocardiographic (area-length) study to provide a simple, useful parameter for evaluation of fetal heart development. Materials and Methods Study Population The study population consisted of 500 healthy fetuses (East Asian race; 242 female and 258 males; mean age ± SD, 192 ± 36 days; range, 91 280 days) who were divided into 2 groups (250 participants per group): secondtrimester group (91 189 days) and third-trimester group (196 280 days). The fetuses were evaluated by clinical and physical assessments, chromosome examinations, sonographic examinations, and magnetic resonance imaging. The pregnant women (mean age, 29.26 ± 4.54 years; range, 23 44 years) were evaluated by clinical and physical assessments, laboratory data, electrocardiography, sonographic examinations, and magnetic resonance imaging. The inclusion criterion for the fetuses and pregnant women was the absence of any focal or diffuse disease at any of the examined organs. Pregnant women with risk factors, such as diabetes mellitus, hypertension, congenital heart disease, cardiomyopathy, and endocrine diseases, were excluded from the study. The study was approved by the local Human Research Ethics Committee, and informed consent was obtained from all pregnant women. The GA was calculated from the first date of the last menstrual period and confirmed by sonography. Echocardiographic and Biometric Measurements The echocardiographic data and biometric indices were acquired transabdominally with the following ultrasound systems: Vivid E9 (GE Healthcare, Horten, Norway) equipped with an M5S single-crystal matrix array transducer and Voluson E8 (GE Healthcare) equipped with a C1-5-D wideband convex transducer. All acquisitions were performed independently by 2 experienced operators. Data were obtained longitudinally; ie, the same pregnant women came in at 20, 24, 28, 32, and 36 weeks, and each time, data were stored digitally for offline analysis. The echocardiographic data were stored at a frame rate of 80 frames per second for subsequent analysis. In this process, width angles were kept at 60 to 120 ; gains were adjusted at the minimum optimal level to minimize noise; frequencies were adjusted to 1.7 to 3.3 MHz; and the filter settings were kept low (50 Hz). All values for each parameter were obtained by averaging 3 measurements. First, the fetal apical 4-chamber view was obtained. The apical LV long-axis length at end diastole (LVLd) and end systole (LVLs) were measured from the endocardial boundaries to the middle of the mitral annulus (Figure 1, A and C). Second, the LV short-axis view at the papillary muscle level was obtained. The LV epicardial short-axis area at the level of the papillary muscle tips at end diastole (LVAd SAX EPI), LV endocardial short-axis area at the papillary muscle level at end diastole (LVAd SAX PM), LV epicardial short-axis area at the level of the papillary muscle tips at end systole (LVAs SAX EPI), and LV endocardial short-axis area at the papillary muscle level at end systole (LVAs SAX PM) were measured. The LV mass (area-length) at end diastole (LVd mass) and LV mass (arealength) at end systole (LVs mass) were automatically calculated by the ultrasound system (Figure 1, B and D). The difference between the LVd mass and LVs mass [LV(d-s) mass] was calculated. Finally, biometric parameters were measured as follows 12,13 : The biparietal diameter (BPD) was measured from the proximal echo of the fetal skull to the proximal edge of the deep border (outer-inner) at the level of the cavum septi pellucidi. The head circumference (HC) was measured as an ellipse around the perimeter of the fetal skull. 1 The abdominal circumference (AC) was measured in the transverse plane of the fetal abdomen at the level of 350 J Ultrasound Med 2014; 33:349 354

the umbilical vein in the anterior third and the stomach bubble in the same plane; measurements were taken around the perimeter. The femur length (FL) was measured in a view in which the full femoral diaphysis was seen and was taken from one end of the diaphysis to the other, not including the distal femoral epiphysis. The EFW was computed according to the Hadlock formula on the basis of BPD, HC, AC, and FL 12,14 : log10 EFW = 1.3596 + 0.0064 (HC) + 0.0424 (AC) + 0.174 (FL) + 0.00061 (BPD) (AC) 0.00386 (AC) (FL). Then the ratio of LVd mass/efw and LVs mass/efw ratios were calculated. Reproducibility Intraobserver variability was assessed in 100 randomly selected participants by repeating the measurements on 2 occasions (3 days apart) under the same basal conditions. To test the interobserver variability, the measurements were performed on the same patient by a second blinded observer. Observers 1 and 2 both had PhD degrees. Observer 1 was a well-trained radiologist with 25 years of experience in fetal heart examinations, whereas observer 2 was a new learner with only 1 year of experience in fetal heart examinations. Variability was calculated as the mean percent error, derived as the difference between the 2 sets of measurements divided by the mean observations. Statistical Analysis Data were expressed as mean ± standard deviation. The differences between the 2 groups were tested by an unpaired 2-tailed t test. The LV mass was compared with GA and EFW by curvilinear regression analysis. P<.05 was considered statistically significant. All statistical analysis was performed with SPSS version 13 software for Windows (IBM Corporation, Chicago, IL). Results The echocardiographic and biometric measurements were successfully completed in all patients. As shown in Table 1, GA, EFW, LVd mass, LVs mass, and LV(d-s) mass were all significantly greater in the third-trimester group than the second-trimester groups (P <.05), whereas the LVd mass/efw and LVs mass/efw ratios did not differ between the second- and third-trimester groups (P >.05). Figure 1. Measurements of LV mass (area-length [A-L]) at end diastole and end systole using biplane 2D echocardiographic methods. A, Apical LVLd. B, LVd mass, LVAd SAX EPI, and LVAd SAX PM. C, Apical LVLs. D, LVs mass, LVAs SAX EPI, and LVAs SAX PM. A B C D J Ultrasound Med 2014; 33:349 354 351

Curvilinear regression analysis (model: growth) showed that there was a significant correlation between LVd mass and GA (r 2 = 0.55; P <.001) and weight (r 2 = 0.54; P <.001) and between LVs mass and GA (r 2 = 0.51; P <.001) and weight (r 2 = 0.56; P <.001), as well as between LV(d-s) mass and GA (r 2 = 0.23; P <.001) and weight (r 2 = 0.22; P <.001; Figure 2). However, the LVd mass/efw and LVs mass/efw ratios did not correlate with GA (r 2 = 0.011; P =.365; r 2 = 0.019; P =.271, respectively; Figure 3). Intraobserver and interobserver variability for the parameters measured are shown in Table 2. Intraobserver and interobserver variability rates for apical LVLd and LVLs ranged from 2.4% to 3.7%. Intraobserver and interobserver variability rates for LVAd SAX EPI, LVAd SAX PM, LVAs SAX EPI, and LVAs SAX PM were all less than 10%. Intraobserver and interobserver variability rates for BPD, HC, AC, and FL were all less than 5%. were significantly greater in the third trimester than in the second trimester and correlated significantly with GA and weight support this conclusion. These findings also show the fetal cardiac mass measurement is a useful parameter for evaluation of fetal heart development. Figure 2. Correlation between LV mass and estimated GA. A, LVd mass. B, LVs mass. C, LV(d-s) mass. A Discussion The results presented here indicate that 2D echocardiography using area-length calculation methods can effectively provide LV mass measurements and can sensitively indicate fetal weight- and GA-related changes in LV mass. It is a reliable modality for evaluation of fetal heart development. In the process of fetal development, with the increase in fetal volume, the fetal hemodynamics and blood flow change dramatically. To adapt to the augmented workload, the developing myocardium increases its cell number, which further results in cardiac growth and increased mass. Abnormal alterations in blood flow may lead to impaired cardiac growth and malformations. Since the fetal heart is in a proliferative state of development, it appears to respond rapidly with adaptive physiologic mechanisms. 1,15 In our study cohort, the findings that LVd mass and LVs mass B C Table 1. Comparative Features and Measurements According to GA Intervals Parameter Second Trimester Third Trimester GA, d 181.54 ± 5.44 211.7 ± 13.41 a EFW, g 782.67 ± 171.79 1326.14 ± 446.89 b LVd mass, g 2.48 ± 0.79 4.07 ± 1.06 b LVs mass, g 1.65 ± 0.55 2.84 ± 0.69 b LV(d-s) mass, g 0.84 ± 0.37 1.23 ± 0.53 a LVd mass/efw ratio, 10 3 3.9 ± 0.18 4.3 ± 0.39 LVs mass/efw ratio, 10 3 1.9 ± 0. 52 2.1 ± 0.45 Data are presented as mean ± SD. a P<.05; b P<.01, unpaired t test, compared to the second-trimester values. 352 J Ultrasound Med 2014; 33:349 354

Figure 3. Correlation between LV mass-to-fetal weight ratio and GA. A, LVd mass/efw ratio. B, LVs mass/efw ratio. A In our study, LV mass was measured by 2D echocardiography using area-length calculation methods. As commonly known, the ventricular cavity has different lengths and cross sections at end diastole and end systole; ie, the heart has different masses as determined by arealength calculation methods at end diastole and end systole, with the ventricular mass at end diastole being greater than that at end systole. Liang et al 16 verified this phenomenon in pigs. In our study cohort, the finding that the LVd mass differed from LVs mass and was significantly greater than it throughout gestation is consistent with the findings of Liang et al. 16 As the fetus continues to mature, the fetal heart needs stronger contractile function and more myocardial perfusion. Our previous study found that the difference in LV mass at end diastole and end systole in adults strongly correlated with myocardial perfusion. 17 Another study also confirmed that there is a positive relationship between the fetal heat contraction fraction and the ventricular mass difference in systole and diastole. 1 Our findings were consistent with the views mentioned above in that the LV(d-s) mass was significantly greater in the third trimester than in the second trimester and correlated significantly with GA and weight. However, the issue of the LV mass-to-fetal weight ratio is another matter. In this study, the ratios were constant values ( 4 10 3 for LVd mass and 2 10 3 for LVs mass throughout gestation). This finding can be explained by the simultaneous increase in LV mass and fetal weight during the process of fetal development. Interestingly, the constant LV mass-to-fetal ratios provide new reference values for evaluation of fetal development. Another question is whether 2D area-length methods are time-consuming processes. In a previous study, the time spent for each analysis in 2D area-length and 3D methods was compared. The time spent in the 2D arealength method was significantly less than in the 3D method (35 ± 13 versus 49 ± 21 seconds; P <.05; X.-Z.Z., B.Y., and J.W. unpublished data, May 2013). We think that the 2D area-length method is a time-saving modality for determination of fetal LV mass. Our study had some possible limitations. First, LV mass as measured by 2D echocardiography using arealength calculation methods has inherent defects, which assume that the LV has a prolate ellipsoid shape. 18 Formulas were developed for calculation of LV mass on 2D echocardiography that were based on regression equations for calculated mass from autopsy findings. This factor was a major limitation in this study. Every fetus will have a flat interventricular septum because of the equal right and left ventricular pressures during fetal life; thus, there is a systematic error in the application of area-length calculations to the fetus. Several factors, such as a non- B Table 2. Intraobserver and Interobserver Variability for the Parameters Measured Intraobserver Interobserver Parameter Variability, % Variability, % Apical LVLd 1.9 ± 0.9 1.8 ± 1.1 Apical LVLs 1.8 ± 1.3 1.9 ± 1.0 LVAd SAX EPI 6.5 ± 2.4 6.9 ± 2.1 LVAd SAX PM 6.7 ± 2.5 6.3 ± 2.6 LVAs SAX EPI 7.2 ± 1.9 7.4 ± 1.9 LVAs SAX PM 6.9 ± 2.1 7.5 ± 2.4 BPD 1.5 ± 0.9 1.6 ± 1.3 HC 2.4 ± 1.6 2.6 ± 1.7 AC 2.3 ± 1.7 2.5 ± 1.4 FL 1.9 ± 0.8 2.1 ± 1.1 Data are presented as mean ± SD. J Ultrasound Med 2014; 33:349 354 353

standard LV long- or short-axis view, unclear epicardial or endocardial boundaries, and poor acoustic windows, all affect the accuracy of the measurements. In our study, the correlations between LVd mass and GA and weight were significantly smaller than those reported by Bhat et al, 11 who used 3D echocardiography for determination of fetal ventricular mass (r 2 = 0.55 versus 0.81 for GA; P <.001; r 2 = 0.54 versus 0.82 for weight; P <.001). These differences may be relevant to the inherent defects of 2D echocardiography using area-length calculation methods. In addition, the number of participants and race were limited. New data need to be collected in subsequent studies. Even so, arealength biplane 2D echocardiography, being rapid, convenient, inexpensive, reliable, and noninvasive for assessment of LV mass, has potential for clinical applications. In conclusion, in this study, we assessed changes in LV mass relative to fetal weight and GA. We have demonstrated that LVd mass, LVs mass, LV(d-s) mass, and fetal weight increase throughout gestation, whereas LV massto-fetal weight ratios are constant values. Although our study had some limitations, as mentioned above, this method still holds considerable clinical promise for evaluation of fetal development. References 1. Messing B, Cohen SM, Valsky DV, et al. Fetal heart ventricular mass obtained by STIC acquisition combined with inversion mode and VOCAL. Ultrasound Obstet Gynecol 2011; 38:191 197. 2. St John Sutton MG, Gewitz MH, Shah B, et al. Quantitative assessment of growth and function of the cardiac chambers in the normal human fetus: a prospective longitudinal echocardiographic study. Circulation 1984; 69:645 654. 3. Zureik M, Bonithon-Kopp C, Lecomte E, Siest G, Ducimetiere P. Weights at birth and in early infancy, systolic pressure, and left ventricular structure in subjects aged 8 to 24 years. Hypertension 1996; 27:339 345. 4. Pinter SZ, Rubin JM, Kripfgans OD, et al. Three-dimensional sonographic measurement of blood volume flow in the umbilical cord. J Ultrasound Med 2012; 31:1927 1934. 5. Sepulveda W, Cafici D, Bartholomew J, Wong AE, Martinez-Ten P. Firsttrimester assessment of the fetal palate: a novel application of the Volume NT algorithm. J Ultrasound Med 2012; 31:1443 1448. 6. Lindell G, Källén K, Maršál K. Ultrasound weight estimation of large fetuses. Acta Obstet Gynecol Scand 2012; 91:1218 1225. 7. Uerpairojkit B, Witoonpanich P. Prenatal ultrasound diagnosis in Thailand. Southeast Asian J Trop Med Public Health 1999; 30(suppl 2):193 195. 8. Alfakih K, Bloomer T, Bainbridge S, et al. A comparison of left ventricular mass between two-dimensional echocardiography, using fundamental and tissue harmonic imaging, and cardiac MRI in patients with hypertension. Eur J Radiol 2004; 52:103 109. 9. Ghanem A, Röll W, Hashemi T, et al. Echocardiographic assessment of left ventricular mass in neonatal and adult mice: accuracy of different echocardiographic methods. Echocardiography 2006; 23:900 907. 10. Collins KA, Korcarz CE, Shroff SG, et al. Accuracy of echocardiographic estimates of left ventricular mass in mice. Am J Physiol Heart Circ Physiol 2001; 280:H1954 H1962. 11. Bhat AH, Corbett V, Carpenter N, et al. Fetal ventricular mass determination on three-dimensional echocardiography: studies in normal fetuses and validation experiments. Circulation 2004; 110:1054 1060. 12. Melamed N, Ben-Haroush A, Meizner I, Mashiach R, Glezerman M, Yogev Y. Accuracy of sonographic weight estimation as a function of fetal sex. Ultrasound Obstet Gynecol 2011; 38:67 73. 13. Melamed N, Yogev Y, Linder N, et al. Role of fetal length in the prediction of fetal weight. J Ultrasound Med 2012; 31:687-94. 14. 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; 150:535 540. 15. de Almeida A, McQuinn T, Sedmera D. Increased ventricular preload is compensated by myocyte proliferation in normal and hypoplastic fetal chick left ventricle. Circ Res 2007; 100:1363 1370. 16. Liang XC, Huang GY, Chen GZ. Experimental study on assessment of left and right ventricular mass by real-time three-dimensional echocardiography. BME Clin Med 2007; 11:9 12. 17. Zheng XZ, Ji P, Mao HW. Reduced difference in left ventricular mass at end diastole and end systole is a predictor of major stenosis of the left coronary artery territory. J Ultrasound Med 2012; 31:1437 1442. 18. Myerson S, Montgomery HE, World MJ, Pennell DJ. Left ventricular mass: reliability of M-mode and 2-dimensional echocardiographic formulas. Hypertension 2002; 40:673 678. 354 J Ultrasound Med 2014; 33:349 354