Congenital heart disease (CHD) affects nearly 1% of newborns and represents a leading cause of infant mortality; 18% of affected infants die within the first year of life, which accounts for more than half of childhood deaths secondary to congenital malformations.1 Moreover, in a long-term infant mortality survey, the World Health Organization reports that 42% of infant deaths were attributable to cardiac defects.2
Major structural abnormalities can be detected by ultrasonography in the late first trimester.3,4 However, diagnostic accuracy is heavily influenced by several factors, such as ultrasonographer expertise,5 protocol used,6 equipment quality, severity of CHD,7 and gestational age at the time of ultrasound examination.8–10 In contrast with the second trimester, when conventional autopsy can be performed to evaluate fetal heart structures,11–13 first-trimester conventional autopsy is less accurate.14 Fetal autopsies after first-trimester loss or termination are hindered by the small cardiac size as well as the cardiac anatomy’s intricacy and great diversity of cardiac malformations. In addition, accuracy can be heavily influenced by the experience of the pathologist.
Imaging methods such as 9,4-T magnetic resonance imaging and microcomputed tomography15–17 have been proposed for confirmation of fetal heart anomalies detected by ultrasonography. The value of these imaging methods has been confirmed by comparing the results with those of stereomicroscopic examination or conventional autopsy at more than 13 weeks of gestation.18,19 However, although these methods offer high-quality images of the fetal heart, they require highly specialized equipment and training, which are not available in many settings.
Three-dimensional (3D) reconstruction of serial histologic slices represents a complex and informative imaging method20 that may improve the capacity to evaluate for cardiac malformations at earlier gestational ages. This study represents a continuation of our previous work,20 in which we successfully identified all anatomic elements of the normal first-trimester fetal heart using histologic 3D imaging reconstruction. We hypothesized that this technique would be useful for verification of fetal heart anomalies detected prenatally at the first-trimester anomaly scan.
This was a cohort study of pregnant women who elected first-trimester pregnancy termination in the setting of a fetus with suspected CHD on first-trimester ultrasound examination. Medical termination of pregnancy was performed using mifepristone and misoprostol, according to national guidelines. Before termination, at 12–13 weeks of gestation, a detailed ultrasound examination was performed by a team of experienced maternal–fetal medicine subspecialists using transabdominal ultrasonography following a previously published protocol6 using color or high-definition directional power Doppler.21 A transvaginal approach was used when transabdominal imaging was inadequate. Voluson E10 and E8 systems equipped with RM6C and RIC5-9-D convex transducers were used for the ultrasound examinations.
After medical termination of pregnancy, fetal autopsy was performed. The fetal hearts were removed and sent to the University of Medicine and Pharmacy Craiova's Research Centre for Microscopic Morphology and Immunology for further evaluation. The tissue was preserved in 10% neutral buffered formalin for 15 days, and paraffin embedding followed the standard protocol. Using a motorized HMB450 rotary microtome, each heart block was sectioned in serial 10-micrometer–thick sections. Sectioned slices were collected with a specific section transfer system on poly-l-lysine–prepared slides for improved adherence and left to dry at 37°C for 1 day. All slides were numbered to preserve slices in the correct sequence.
Subsequently, we used the standard histologic protocol for hematoxylin-eosin staining. Special coloration (periodic acid-Schiff stain, Masson's trichrome stain, and orcein) was used to confirm specific features, such as fibroelastosis in certain malformations. The colored slides were scanned using a Motic EasyScan scanner at 20× objective and saved in a proprietary format in the Motic Digital Slide Assistant package database. Image resolution was set to 72×72 dpi and quality to 10 using the “Batch image manipulation” plugin in GNU Image Manipulation Program (Fig. 1). This enabled faster manipulation in the 3D reconstruction software. We eliminated the slides with tissue roll or rupture during the histology process.
The images were imported into Amira Avizo software. The auto align function was used to align slices, and only minor manual adjustments were necessary. We used the “threshold” function with a masking range of 75–160 voxel value for segmentation. Artifact removal significantly improves the overall aspect of the volume easily and relatively fast using the “brush,” “blow,” or “lasso” tools. Segmentation then was performed to individualize heart structures using the “brush” tool, and each element of interest was assigned a different color (Fig. 1).
We used different sections through the reconstructed heart to evaluate the heart structure and highlight the defects. We generally aimed to section the fetal heart rendered volume according to the classic key planes of the ultrasound cardiac sweep (four-chamber view, left and right ventricular outflow tracts). In addition, certain functions can be applied to the volume to strengthen the results. To better visualize the differences between the left and right ventricles of the heart, we used “compute ambient occlusion” and “interactive threshold” to render the cavities' volumes. The 3D reconstructions and histologic slides (when needed) were analyzed and reviewed by a multidisciplinary team that included maternal–fetal medicine subspecialists and pathologists.
The ethical norms and good practice in scientific research were followed throughout the study. Therefore, the research ethics committees of the University of Medicine and Pharmacy in Craiova (no. 27/24.02.2021) and of the Emergency County Clinical Hospital of Craiova (no. 38680/13.09.2021) approved this study. Informed consent was obtained from all of the included patients.
Six fetuses with suspected heart malformations were investigated using histologic 3D imaging reconstruction: two with hypoplastic left heart syndrome, two with atrioventricular septal defects, one with an isolated ventricular septal defect, and one with transposition of the great arteries. Turner syndrome and trisomy 21 were detected with genetic analysis in the ventricular septal defect and atrioventricular septal defect cases, respectively.
After the segmentation process, we identified the cardiac structures and all of the defects detected by first-trimester fetal echocardiography. Even a small defect, such as the isolated ventricular septal defect (Fig. 2), was visualized using the lateral views of the septum from the right and left ventricles in 3D reconstruction (Fig. 3).
The cases with more severe defects detected on ultrasound examination, such as the atrioventricular septal defect (Fig. 4), were straightforward to confirm in 3D histologic reconstruction (Fig. 5). An axial plane through the heart reveals the incomplete ventricular and atrial septum. The abnormal anatomy of the atrioventricular valves is evident—only one cusp is visible on each side because of the common atrioventricular valve (Fig. 5). Furthermore, in this case, the systematic evaluation of the structures enabled the diagnosis of additional anomalies not detected at the time of the first-trimester ultrasound examination. We noted a bicuspid aortic valve (Appendix 1, available online at https://links.lww.com/AOG/D117) and a rare anatomical variant regarding the thymus position in relation to the left brachiocephalic vein. The right lobe of the thymus was located posteriorly to the left brachiocephalic vein, a rare variation that is important to be aware of in the advent of thymic interventions (Appendix 2, available online at https://links.lww.com/AOG/D117).
For the cases in which hypoplastic left heart syndrome was detected (Fig. 6), we emphasized the differences between the two ventricles and great vessels through multiple modalities. The fastest way to evaluate the ventricular size discrepancy was to investigate the axial sections of the reconstructed heart (Fig. 7), similar and parallel to the four-chamber view plane. Longitudinal long-axis views of the fetal heart highlight the size differences of both ventricular and outflow tracts (Appendix 3, part C, available online at https://links.lww.com/AOG/D117). The cavities' relative sizes can be further highlighted and compared using the volume modification functions (Appendix 3, part A, https://links.lww.com/AOG/D117). Also, a transverse section at the base of the heart illustrates the caliber difference between the normal pulmonary artery and hypoplastic aorta (Appendix 3, part B, https://links.lww.com/AOG/D117).
The ultrasound scan revealed a nonfunctional but thickened and echogenic left ventricle, suggesting myocardial hypertrophy and endocardial fibroelastosis. This led us to investigate the presence of cardiomyopathy and endocardial fibroelastosis using special histologic colorations (Appendix 4, available online at https://links.lww.com/AOG/D117). Ventricular wall hypertrophy was accompanied by increased subendothelial density of the collagen, fibrin, and elastic fibers, consistent with endocardial fibroelastosis.
In the transposition of the great arteries case (Fig. 8), the main characteristics of the anomaly were evident using histologic 3D imaging reconstruction: parallel vessel alignment (Fig. 9A) and ventriculoarterial discordance, where the aorta arises from the morphologic right ventricle and the pulmonary artery arises from the morphologic left ventricle (Fig. 9B). Furthermore, evaluation of the coronary arteries was possible, and we noted the circumflex artery arising from the right coronary artery (Fig. 9C).
Given the high incidence of CHD, a reliable and widely available method to confirm malformations detected prenatally by ultrasonography should be available. In this article, we demonstrate the ability to confirm CHD detected on first-trimester ultrasound examination in fetal specimens after termination of pregnancy or pregnancy loss using histologic 3D imaging reconstruction. The histologic 3D imaging reconstruction protocol used in our study is relatively low cost and employs generic equipment and easily acquirable software. Furthermore, the learning curve of each step is not steep, because the technique is widely used in general pathology practice. The process can be automated, further reducing the time needed to prepare and scan the histologic slides.22
The studies throughout the literature note that perinatal autopsy performed in the second or third trimester serves not only as an audit for prenatal ultrasound examination findings but also identifies additional anomalies.23–25 The same statement may now be applied to the first trimester using this technique. Histologic 3D imaging reconstruction allowed us to identify the ultrasound-detected anomalies in a small but diverse series of cases (cardiac chambers and conotruncal anomalies) and facilitated the diagnosis of additional findings (hypertrophic cardiomyopathy and fibroelastosis in the hypoplastic left heart syndrome case, bicuspid aortic valve and the anatomical variant of the thymus in the atrioventricular septal defect case, and the branching variation of the coronary arteries in the transposition of the great arteries case)26,27 that could not have been established by ultrasonography at the time of the first-trimester scan. These additional findings may affect counseling of the patient regarding recurrence risk.28,29
Perinatal autopsy complemented with histologic examination remains the gold standard for fetal anatomy assessment, even in the era of high-resolution computed tomography and magnetic resonance imaging.30 Histologic 3D imaging reconstruction is an important step in the field. It provides additional advantages, including medical teaching31–37 and telemedicine, because the reconstructions can be re-examined by various practitioners. This technique also facilitates remote analysis and reduces the time and costs of transfer of physical pathology samples between institutions.
Histologic 3D imaging reconstruction also provides an opportunity to retain specific slices for supplementary special stains, which can aid in evaluating the fetal heart38 or add valuable information to standard autopsy.24,39,40 In the hypoplastic left heart syndrome case, we improved the diagnosis by detecting endocardial fibroelastosis41,42 and characterized the associated hypertrophic cardiomyopathy. This finding is important in clinical care, because the literature suggests a genetic component of this disease43 as well as an association with maternal anti-Ro and anti-La antibodies.44,45
One limitation of this technique is that distortion can occur from one image to another due to tissue shrinkage throughout the process or inhomogeneous relaxation of sections before mounting.46–48 Methods are available to overcome this limitation,49–51 and automation of the slicing and scanning, user-friendly software, and experienced team members can overcome these technical issues.
In conclusion, we demonstrate the capacity to confirm fetal heart anomalies using 3D histologic imaging reconstruction of fetal hearts. The findings using histologic 3D imaging reconstruction consistently confirmed the first-trimester ultrasound imaging findings of congenital heart anomalies and, in some cases, identified additional anomalies.
1. Pinto NM, Keenan HT, Minich LL, Puchalski MD, Heywood M, Botto LD. Barriers to prenatal detection of congenital heart disease: a population-based study. Ultrasound Obstet Gynecol 2012;40:418–25. doi: 10.1002/uog.10116
2. Rosano A, Botto L, Botting B, Mastroiacovo P. Infant mortality and congenital anomalies from 1950 to 1994: an international perspective. J Epidemiol Commun Health 2000;54:660–6. doi: 10.1136/jech.54.9.660
3. Quarello E, Lafouge A, Fries N, Salomon LJ, Cfef T. Basic heart examination: feasibility study of first-trimester systematic simplified fetal echocardiography. Ultrasound Obstet Gynecol 2017;49:224–30. doi: 10.1002/uog.15866
4. Gembruch U, Knöpfle G, Bald R, Hansmann M. Early diagnosis of fetal congenital heart disease by transvaginal echocardiography. Ultrasound Obstet Gynecol 1993;3:310–7. doi: 10.1046/j.1469-0705.1993.03050310.x
5. Ye B, Wu Y, Chen J, Yang Y, Niu J, Wang H, et al. The diagnostic value of the early extended fetal heart examination at 13 to 14 weeks gestational age in a high-risk population. Transl Pediatr 2021;10:2907–20. doi: 10.21037/tp-21-255
6. Iliescu D, Tudorache S, Comanescu A, Antsaklis P, Cotarcea S, Novac L, et al. Improved detection rate of structural abnormalities in the first trimester using an extended examination protocol. Ultrasound Obstet Gynecol 2013;42:300–9. doi: 10.1002/uog.12489
7. Jicinska H, Vlasin P, Jicinsky M, Grochova I, Tomek V, Volaufova J, et al. Does first-trimester screening modify the natural history of congenital heart disease? Circulation 2017;135:1045–55. doi: 10.1161/CIRCULATIONAHA.115.020864
8. Gadsbøll K, Wright A, Kristensen SE, Verfaille V, Nicolaides KH, Wright D, et al. Crown-rump length measurement error: impact on assessment of growth. Ultrasound Obstet Gynecol 2021;58:354–9. doi: 10.1002/uog.23690
9. Schmidt P, Staboulidou I, Elsässer M, Vaske B, Hillemanns P, Scharf A. How imprecise may the measurement of fetal nuchal translucency be without worsening first-trimester screening? Fetal Diagn Ther 2008;24:291–5. doi: 10.1159/000158520
10. Hildebrand E, Gottvall T, Blomberg M. Maternal obesity and detection rate of fetal structural anomalies. Fetal Diagn Ther 2013;33:246–51. doi: 10.1159/000343219
11. Picazo-Angelin B, Zabala-Argüelles JI, Anderson RH, Sánchez-Quintana D. Anatomy of the normal fetal heart: the basis for understanding fetal echocardiography. Ann Pediatr Cardiol 2018;11:164–73. doi: 10.4103/apc.APC_152_17
12. Erickson LK. An approach to the examination of the fetal congenitally malformed heart at autopsy. J Fetal Med 2015;2:135–41. doi: 10.1007/s40556-015-0061-z
13. Shanmugasundaram S, Venkataswamy C, Gurusamy U. Pathologist's role in identifying cardiac defects—a fetal autopsy series. Cardiovasc Pathol 2021;51:107312. doi: 10.1016/j.carpath.2020.107312
14. Albu C, Staicu A, Popa-Stanilă R, Bondor C, Pop B, Chiriac L, et al. The evaluation of the four-chamber cardiac dissection method of the fetal heart as an alternative to conventional inflow–outflow dissection in small gestational-age fetuses. Diagnostics 2022;12:223. doi: 10.3390/diagnostics12010223
15. Lombardi CM, Zambelli V, Botta G, Moltrasio F, Cattoretti G, Lucchini V, et al. Postmortem microcomputed tomography (micro-CT) of small fetuses and hearts. Ultrasound Obstet Gynecol 2014;44:600–9. doi: 10.1002/uog.13330
16. Sandrini C, Boito S, Lombardi CM, Lombardi S. Postmortem micro-CT of human fetal heart—a systematic literature review. J Clin Med 2021;10:4726. doi: 10.3390/jcm10204726
17. Tang H, Zhang Y, Dai C, Ru T, Li J, Chen J, et al. Postmortem 9.4-T MRI for fetuses with congenital heart defects diagnosed in the first trimester. Front Cardiovasc Med 2021;8:764587. doi: 10.3389/fcvm.2021.764587
18. Votino C, Jani J, Verhoye M, Bessieres B, Fierens Y, Segers V, et al. Postmortem examination of human fetal hearts at or below 20 weeks' gestation: a comparison of high-field MRI at 9.4 T with lower-field MRI magnets and stereomicroscopic autopsy. Ultrasound Obstet Gynecol 2012;40:437–44. doi: 10.1002/uog.11191
19. Staicu A, Albu C, Popa-Stanila R, Chiriac L, Boitor-Borza D, Bondor C, et al. Potential clinical benefits and limitations of fetal virtopsy using high-field MRI at 7 Tesla versus stereomicroscopic autopsy to assess first trimester fetuses. Prenatal Diagn 2019;39:505–18. doi: 10.1002/pd.5457
20. Pichat J, Iglesias JE, Yousry T, Ourselin S, Modat M. A survey of methods for 3D histology reconstruction. Med Image Anal 2018;46:73–105. doi: 10.1016/j.media.2018.02.004
21. Ruican D, Petrescu AM, Ungureanu AL, Marinaş MC, Pirici D, Istrate-Ofiţeru AM, et al. Virtual autopsy and confirmation of normal fetal heart anatomy in the first trimester using three-dimensional (3D) reconstruction of histological sections. Rom J Morphol Embryol 2021;62:101–8. doi: 10.47162/RJME.62.1.09
22. Tudorache S, Cara M, Iliescu DG, Novac L, Cernea N. First trimester two- and four-dimensional cardiac scan: intra- and interobserver agreement, comparison between methods and benefits of color Doppler technique. Ultrasound Obstet Gynecol 2013;42:659–68. doi: 10.1002/uog.12459
23. Histopathology is ripe for automation. Nat Biomed Eng 2017;1:925. doi: 10.1038/s41551-017-0179-5
24. Kammoun A, Magdoud K, Frikha H, Karray I, Karoui A, Menjli S, et al. VP15.03: correlation between prenatal ultrasound and fetal autopsy findings: a retrospective study of second trimester termination of pregnancy [abstract]. Ultrasound Obstet Gynecol 2021;58:160. doi: 10.1002/uog.24260
25. Şorop-Florea M, Ciurea RN, Ioana M, Stepan
F, et al. The importance of perinatal autopsy. Review of the literature and series of cases. Rom J Morphol Embryol 2017;58:323–37.
26. Vogt C, Blaas HGK, Salvesen KÅ, Eik-Nes SH. Comparison between prenatal ultrasound and postmortem findings in fetuses and infants with developmental anomalies. Ultrasound Obstet Gynecol 2012;39:666–72. doi: 10.1002/uog.10106
27. Graham TR, Holden MP. A thymus gland behind the brachiocephalic vein in a patient with myasthenia gravis. Int J Cardiol 1986;12:99–102. doi: 10.1016/0167-5273(86)90104-x
28. Plaza OA, Moreno F. Anatomical variations of the thymus in relation to the left brachiocephalic vein, findings of necropsia. Int J Pediatr Otorhinolaryngol 2018;107:53–5. doi: 10.1016/j.ijporl.2018.01.019
29. Pedersen MW, Groth KA, Mortensen KH, Brodersen J, Gravholt CH, Andersen NH. Clinical and pathophysiological aspects of bicuspid aortic valve disease. Cardiol Young 2019;29:1–10. doi: 10.1017/S1047951118001658
30. van Engelen K, Bartelings MM, Gittenberger-de Groot AC, Baars MJH, Postma AV, Bijlsma EK, et al. Bicuspid aortic valve morphology and associated cardiovascular abnormalities in fetal Turner syndrome: a pathomorphological study. Fetal Diagn Ther 2014;36:59–68. doi: 10.1159/000357706
31. den Bakker MA. Is histopathology still the gold standard? [in Dutch]. Ned Tijdschr Geneeskd 2017;160:D981.
32. Anyanwu GE, Agu AU, Anyaehie UB. Enhancing learning objectives by use of simple virtual microscopic slides in cellular physiology and histology: impact and attitudes. Adv Physiol Educ 2012;36:158–63. doi: 10.1152/advan.00008.2012
33. Loke YH, Harahsheh AS, Krieger A, Olivieri LJ. Usage of 3D models of tetralogy of Fallot for medical education: impact on learning congenital heart disease. BMC Med Educ 2017;17:54. doi: 10.1186/s12909-017-0889-0
34. Harris T, Leaven T, Heidger P, Kreiter C, Duncan J, Dick F. Comparison of a virtual microscope laboratory to a regular microscope laboratory for teaching histology. Anatomical Rec 2001;265:10–4. doi: 10.1002/ar.1036
35. Blake CA, Lavoie HA, Millette CF. Teaching medical histology at the University of South Carolina School of Medicine: transition to virtual slides and virtual microscopes. Anatomical Rec 2003;275B:196–206. doi: 10.1002/ar.b.10037
36. Sieben A, Oparka R, Erolin C. Histology in 3D: development of an online interactive student resource on epithelium. J Vis Commun Med 2017;40:58–65. doi: 10.1080/17453054.2017.1332480
37. Chenoweth EM, Houston J, Burek Huntington K, Straley JM. A virtual necropsy: applications of 3D scanning for marine mammal pathology and education. Animals 2022;12:527. doi: 10.3390/ani12040527
38. Meyer-Szary J, Luis MS, Mikulski S, Patel A, Schulz F, Tretiakow D, et al. The role of 3D printing in planning complex medical procedures and training of medical professionals—cross-sectional multispecialty review. Int J Environ Res Public Health 2022;19:3331. doi: 10.3390/ijerph19063331
39. Lazda EJ, Batchelor WH, Cox PM. Immunohistochemical detection of myocardial necrosis in stillbirth and neonatal death. Pediatr Develop Pathol 2000;3:40–7. doi: 10.1007/s100240050005
40. Iliescu D, Comănescu A, Antsaklis P, Tudorache S, Ghiluşi M, Comănescu V, et al. Neuroimaging parameters in early open spina bifida detection. Further benefit in first trimester screening. Rom J Morphol Embryol 2011;52:809–17.
41. Nagy RD, Ruican D, Zorilă GL, Istrate-Ofiţeru AM, Badiu AM, Iliescu DG. Feasibility of fetal portal venous system ultrasound assessment at the FT anomaly scan. Diagnostics 2022;12:361. doi: 10.3390/diagnostics12020361
42. Fishbein MC, Ferrans VJ, Roberts WC. Histologic and ultrastructural features of primary and secondary endocardial fibroelastosis. Arch Pathol Lab Med 1977;101:49–54.
43. Cole CR, Eghtesady P. The myocardial and coronary histopathology and pathogenesis of hypoplastic left heart syndrome. Cardiol Young 2016;26:19–29. doi: 10.1017/S1047951115001171
44. Hodgson S, Child A, Dyson M. Endocardial fibroelastosis: possible X linked inheritance. J Med Genet 1987;24:210–4. doi: 10.1136/jmg.24.4.210
45. Nield LE, Silverman ED, Taylor GP, Smallhorn JF, Brendan J, Mullen M, et al. Maternal anti-Ro and anti-La antibody-associated endocardial fibroelastosis. Circulation 2002;105:843–8. doi: 10.1161/hc0702.104182
46. Raboisson MJ, Fouron JC, Sonesson SE, Nyman M, Proulx F, Gamache S. Fetal Doppler echocardiographic diagnosis and successful steroid therapy of Luciani-Wenckebach phenomenon and endocardial fibroelastosis related to maternal anti-Ro and anti-La antibodies. J Am Soc Echocardio 2005;18:375–80. doi: 10.1016/j.echo.2004.10.023
47. Burton RAb, Plank G, Schneider JE, Grau V, Ahammer H, Keeling SL, et al. Three-dimensional models of individual cardiac histoanatomy: tools and challenges. Ann N Y Acad Sci 2006;1080:301–19. doi: 10.1196/annals.1380.023
48. Streicher J, Weninger WJ, Müller GB. External marker-based automatic congruencing: a new method of 3D reconstruction from serial sections. Anatomical Rec 1997;248:583–602. doi: 10.1002/(SICI)1097-0185(199708)248:4<583::AID-AR10>3.0.CO;2-L
49. Malandain G, Bardinet É, Nelissen K, Vanduffel W. Fusion of autoradiographs with an MR volume using 2-D and 3-D linear transformations. Neuroimage 2004;23:111–27. doi: 10.1016/j.neuroimage.2004.04.038
50. Casero R, Siedlecka U, Jones ES, Gruscheski L, Gibb M, Schneider JE, et al. Transformation diffusion reconstruction of three-dimensional histology volumes from two-dimensional image stacks. Med Image Anal 2017;38:184–204. doi: 10.1016/j.media.2017.03.004
51. Saalfeld S, Fetter R, Cardona A, Tomancak P. Elastic volume reconstruction from series of ultra-thin microscopy sections. Nat Methods 2012;9:717–20. doi: 10.1038/nmeth.2072