INTRODUCTION

Since the dawn of the age of computed  tomography/magnetic resonance imaging techniques, and the advent of plastination technology, efforts have been made to use both techniques with plastinated specimens as useful teaching tools. Sectional anatomy has been incorporated in the curricula of medical, veterinary and dental schools in order to facilitate learning. The understanding of sectional anatomy is crucial for professionals such as radiologists and neurosurgeons.

Most studies in the literature compare sheet plastinated specimens with images from computed tomography (CT) scanning and magnetic resonance (MRI) imaging. The source of the images may be from living subjects or cadaveric specimens (Cook, 1997; Entius et al., 1997; Reina de la Torre et al., 2015; Sora et al., 2019). Some studies compared plastinated slices to CT images obtained prior to plastination, from the same specimen (Latorre et al., 2006, 2003; Rodriguez et al., 2008, 2010; Parraga et al. 2013; Ottone et al., 2016). A few studies, including Shianti et al. (2015) showed the relationship between CT in pre- and post-plastinated specimens, emphasizing the behavior of X-Ray attenuation coefficient numbers.

The majority of studies showing 3D reconstruction of images of plastinated specimens are based on computerized models (Sora et al., 2007; Tunali et al., 2008; Latorre et al., 2008; Cerqueira et al., 2008 a, b; Rodriguez et. al, 2008; Arredondo et al., 2008 a, b; Sora et al., 2013). In 2012, Tiwari et al. performed a 3D reconstruction of a plastinated 24-week fetus directly from the 16 slice-CT scan machine.

Even though MRI is also important for diagnostic purpose, Baptista et al. (1990) showed that MRI scanning of plastinated specimens produced no signals, for reasons which are discussed later in this paper.

The objective of this study was to analyze the internal and external morphological aspects and integrity of silicone plastinated specimens (Biodur® S10) using tridimensional CT (3D-CT). The radiological aspects of images were studied by comparing the X-ray tissue attenuation coefficients of plastinated specimens by CT, to the coefficients of the same tissue in CT images made from living subjects.

MATERIALS AND METHODS

A human heart (Fig. 1a) and a diencephalon-brain stem (D-BS) (Fig. 1b), plastinated with silicone 34 years ago using the Biodur® S10 silicone standard technique, described by von Hagens (1979, 1985, 1986), were the subjects of this study. For 3D-CT scanning, a 64-multidetector Toshiba Aquilion CT Scanner (Fig. 2) was used and belonged to the Heart Institute of the College of Medicine of the University of São Paulo, Brazil. The radiological parameters of the images acquired are shown in Table 1. The scanner obtained 434 transverse cross-section images from the plastinated heart, and 317 images from the plastinated diencephalon-encephalic trunk (Figs. 3a, 3b).

Figure 1. Silicone plastinates: a: human heart, b: Diencephalon-brain stem	Figure 2. Toshiba Aquilion 64-multidetector CT scan from Radiology Department of Heart Institute – University of São Paulo, Brazil.	Table 1 - Radiological Parameters in Image Acquisition
Figure 3a. Set of CT scans of plastinated human heart cross-sections.	Figure 3b. A single slice CT cross-section of a plastinated human heart.	Figure 4. Workstation interface utilized to reconstruct the transverse sections into tridimensional images.

The cross-sectional images were reconstructed on a Terarecon® Aquarius Net Viewer Workstation by a technique known as Volume Rendering, producing tridimensional images, which can be rotated in all directions (Fig. 4).

The X-ray attenuation coefficients (mean and standard deviation) were obtained from transverse images of the plastinates, measured in Hounsfield units (HU), and compared to the attenuation coefficients obtained from archived images of the same region obtained in CT scans of living subjects (Archives of the Department of Radiology, Heart Institute, University of São Paulo) (Figs. 5a, 5b, 6a, 6b, 6c).

Figure 5a. Area of myocardial X-ray attenuation coefficients measurements: plastinated-Red Circle, b. Living human – Blue Dot	Figure 5b. Area of myocardial X-ray attenuation coefficients measurements: Living human – Blue Dot
Figure 6a. Area of X-Ray attenuation coefficients	Figure 6b. Area of X-Ray attenuation coefficients measurements of D-BS, gray and white matter: Living subject – green circle, Indicates area of gray matter	Figure 6c. Area of X-Ray attenuation coefficients measurements of D-BS, gray and white matter: Living subject –Red circle, Indicates area of white matter

Table 2. X-Ray tissue attenuation coefficients ( Preston R. Chapter1. Imaging the Head and Brain. In Diagnostic Imaging for the Emergency Physician. 2011, Pages 1-45.)

X-Ray attenuation in CT indicates the following: presence of fat or air inside the tissue, air reaching values about -1000 HU (shown as black on image); 0 HU is the attenuation of water (shown as medium gray on image); soft tissue other than fat is between 0 and +100 HU; hemorrhage, and especially bone and other calcified tissues are above +100 HU, bone reaching values about+1000 HU (shown as white on the image)  (Table 2).

The specimens were submitted to a 1.5 T MRI scan to evaluate their behavior in a strong magnetic field.

RESULTS

For the scanned plastinated heart, this study found the following:

1. Plastinated Heart-3D CT reconstructions showed the heart anatomy in detail: the valvular apparatus, including chordae tendineae and papillary muscles were well depicted (Fig. 7).
2. The whole 3D image was sectioned in coronal and sagittal planes (called slabs) showing the interatrial and interventricular septa as well as the great vessels with details (Fig. 8).
3. The myocardial tissue suffered homogenization and showed uniform density, thus individual myocardial fibers were not seen in detail.
4. The myocardial tissue showed an increased attenuation coefficient when compared to the CT-scan of the myocardium of a living subject. The myocardium attenuation values of the plastinate were higher (364.54 +/-17.52 HU) than the heart of a living subject (86.37+/- 6.32 HU).

Figure 7. 3D CT reconstruction of the plastinated heart (anterior view). (1- Right atrial appendage; 2- Left atrial appendage; A – Aorta; P - Pulmonary trunk; ca - Coronary artery; t - Trabeculae carneae)

Figure 8. Longitudinal section of heart with internal detail, 3D CT (slab image). Inner structure: AV - Aortic valve; M - Mitral valve; CT - Chordae tendineae; S - Interventricular septum; P - Papillary muscle; TV - Tricuspid valve)

For the D-BS plastinate, this study found the following:

1. 3D CT reconstructions showed that the inner and outer surfaces of the diencephalon/brain stem were well demarcated (Fig. 9).
2. The cavities (lateral and third ventricles, cerebral aqueduct), thalamus, hypothalamus, pons and medulla) were visible (Fig. 10).
3. Some cranial nerves were individually recognized.
4. It was dificult, in this study, to differentiate white from gray matter.
5. The brain tissue showed uniform density, except for some darker areas of low intensity in the region of the internal capsule. Attenuation values of the plastinate were higher (278.95+/- 7.37 HU) in the brightest areas and 158.88+/- 5.54 HU) in the darkest area. When compared to a living subject, the attenuation values were 25.16+/- 6.37 HU for the white matter in the centrum semiovale, and 34.19+/- 6.22 HU for the gray matter in the nucleus lentiformis.

The MRI scanning of the plastinates did not yield any images. There was no evidence of signal generation in the scan.

Figure 9. Anterior (left) and posterior (right) views of 3D CT reconstruction of diencephalon-brain stem plastinate. C- Cerebral peduncle; P- Pons; MO - Medulla oblongata; cb - Cerebellar peduncle’ cq - Corpora quadrigemina; t –talamus

Figure 10. Sagittal section of D-BS plastinate, 3D CT (slab image). Cc – Corpus callosum (splenium); SP- Septum pellucidum; LV- Lateral ventricle; T- Thalamus MB- Midbrain; P – Pons; oc – Optic chiasm; m – Mammillary body)

DISCUSSION

The internal and external morphology of the plastinated specimens was well characterized for both the heart and D-BS. Tiavri et al. (2012) concluded, after performing a CT scan of a 24-week plastinated fetus, that the three-dimensional morphology of the organs, soft tissue and muscle attachments was preserved.

There was an increase in the X-ray attenuation values of both the heart and D-BS CT images. Shianti et al. (2015) measured the attenuation coefficients of several organs (including liver, heart, lungs, and kidneys) before and after standard S10 plastination. Their study showed a significant increase in the coefficients of all organs studied, with minor differences occurring in bones and air-filled organs. The impregnated silicone seems to absorb more X-ray, making the specimen denser than the non-plastinated tissue, therefore raising the X-Ray attenuation values. These changes in attenuation values may be used to evaluate the degree of silicone impregnation. Once liquid is replaced by silicone, the changes in attenuation values may serve as an indicator of a complete or incomplete silicone impregnation. This phenomenon can explain the areas of low attenuation found inside the D-BS CT images. This finding is not related to the differences found in gray and white matter.

The MRI scans of the plastinated organs, as expected, produced no signal because of the absence of water and fat in the specimens. Baptista et al. (1990), in their study of MRI scans of plastinated specimens, concluded that the quality of the MRI image obtained from plastinated specimens had an indirect relationship to the time of hardening of the specimen. The specimens used for this study reported here were plastinated 34 years ago; consequently, there is no evidence of signals generated in the scan. Since MR images are generated from “vibrations” of hydrogen protons, the hardening process eliminates the possibility of such vibration.

We conclude that CT scans are an excellent method for examining plastinated specimens, especially inner or outer surfaces. Even though good images were obtained, anatomy of the wall architecture could not be determined. The impregnation of the specimens with Biodur S10 reduces the CT attenuation rates of the specimens.

CONCLUSION

ACKNOWLEDGEMENTS

We thank Professor Claudio Campi de Castro, MD, PhD, Chief Radiologist in the Magnetic Resonance sector at The Heart Institute, University of São Paulo, Brazil, for the use of the CT and MRI devices for this study.