Influence of breast characteristics in myocardial scintigraphy through the Monte Carlo method

Oliveira, Anderson; Meguerian, Berdj Aram; Mesquita, Cláudio Tinoco

doi:10.1590/S0066-782X2010005000162

Abstracts

BACKGROUND: By reducing the specificity associated with loss of information, the influence of attenuation of the breasts is very important in myocardial perfusion studies. However, although several studies have been conducted over the past years, little has been developed to determine accurately the influence of the characteristics of the breasts on the quality of myocardial scintigraphy, avoiding additional exposure to radiation. OBJECTIVE: The purpose of this study is to quantify the attenuation of photons by the breasts, in myocardial perfusion studies with 99mTc according to different sizes and compositions. METHODS: Each breast was assumed to be a cube composed of fibroglandular and adipose tissue. The data related to 99mTc photons were analyzed in a Monte Carlo model. We varied the thickness and composition of breasts and analyzed the interference in attenuation. The EGS 4 software was used in the simulations. RESULTS: Setting the thickness of a breast, the variation of its composition causes a maximum increase of 2.3% in the number of photons attenuated. By contrast, maintaining a fixed composition of breast tissue, the difference in photon attenuation was 45.0%, averaging 6.0% for each additional centimeter in the breast thickness. CONCLUSION: Monte Carlo simulation showed that the influence of the thickness of the breasts in the attenuation of photons in myocardial scintigraphy with 99mTc is much greater than the influence of their compositions.

Breast; myocardial reperfusion, injury; Monte Carlo method

FUNDAMENTO: Pela redução da especificidade associada à perda de informação, a influência da atenuação das mamas é de fundamental importância em estudos de perfusão do miocárdio. Entretanto, apesar de vários estudos terem sido realizados ao longo dos últimos anos, pouco se tem evoluído para determinar com acurácia a influência das características das mamas sobre a qualidade da cintilografia miocárdica, evitando exposições adicionais de radiação às pacientes. OBJETIVO: O objetivo deste estudo é quantificar a atenuação de fótons pelas mamas, em estudos de perfusão do miocárdio com 99mTc, de acordo com diferentes tamanhos e composições. MÉTODOS: Cada mama foi assumida como sendo um cubo composto de tecido adiposo e fibroglandular. Os dados referentes aos fótons de 99mTc foram analisados em um modelo de Monte Carlo. Variamos a espessura e a composição das mamas e analisamos as interferências na atenuação. Foi empregado o software EGS 4 para as simulações. RESULTADOS: Fixando a espessura de uma mama, a variação da sua composição acarreta um acréscimo máximo de 2,3% no número de fótons atenuados. Em contrapartida, mantendo-se uma composição do tecido mamário fixa, a diferença na atenuação de fótons foi de 45,0%, sendo em média de 6,0% para cada acréscimo de centímetro na espessura da mama. CONCLUSÃO: A simulação por Monte Carlo demonstrou que a influência das espessuras das mamas na atenuação de fótons em cintilografias do miocárdio com 99mTc é muito maior do que a influência das suas composições.

Mama; traumatismo por reperfusão miocárdica; método Monte Carlo

FUNDAMENTO: Por la reducción de la especificidad asociada a la pérdida de información, la influencia de la atenuación de las mamas es de fundamental importancia en estudios de perfusión del miocardio. Entre tanto, a pesar de que varios estudios han sido realizados a lo largo de los últimos años, poco se ha evolucionado para determinar con precisión la influencia de las características de las mamas sobre la calidad de la cámara gamma miocárdica, evitando exposiciones adicionales de radiación a las pacientes. OBJETIVO: El objetivo de este estudio es cuantificar la atenuación de fotones por las mamas, en estudios de perfusión del miocardio con 99mTc, de acuerdo con diferentes tamaños y composiciones. MÉTODOS: Cada mama fue asumida como siendo un cubo compuesto de tejido adiposo y fibroglandular. Los datos referentes a los fotones de99mTc fueron analizados en un modelo de Monte Carlo. Variamos el espesor y la composición de las mamas y analizamos las interferencias en la atenuación. Fue empleado el software EGS 4 para las simulaciones. RESULTADOS: Fijando el espesor de una mama, la variación de su composición acarrea un aumento máximo de 2,3% en el número de fotones atenuados. En contrapartida, manteniéndose una composición del tejido mamario fija, la diferencia en la atenuación de fotones fue de 45,0%, siendo en media de 6,0% para cada aumento de centímetro en el espesor de la mama. CONCLUSIÓN: La simulación por Monte Carlo demostró que la influencia de los espesores de las mamas en la atenuación de fotones en cámara gamma del miocardio con 99mTc es mucho mayor que la influencia de sus composiciones.

Mama; traumatismo por reperfusión miocárdica; método Monte Carlo

Comissão Nacional de Energia Nuclear, Hospital Pró-Cardíaco, Instituto Nacional do Câncer, Universidade Federal Fluminense, Universidade Gama Filho, Rio de Janeiro, RJ - Brazil

Mailing address

ABSTRACT

BACKGROUND: By reducing the specificity associated with loss of information, the influence of attenuation of the breasts is very important in myocardial perfusion studies. However, although several studies have been conducted over the past years, little has been developed to determine accurately the influence of the characteristics of the breasts on the quality of myocardial scintigraphy, avoiding additional exposure to radiation.

OBJECTIVE: The purpose of this study is to quantify the attenuation of photons by the breasts, in myocardial perfusion studies with ^99mTc according to different sizes and compositions.

METHODS: Each breast was assumed to be a cube composed of fibroglandular and adipose tissue. The data related to ^99mTc photons were analyzed in a Monte Carlo model. We varied the thickness and composition of breasts and analyzed the interference in attenuation. The EGS 4 software was used in the simulations.

RESULTS: Setting the thickness of a breast, the variation of its composition causes a maximum increase of 2.3% in the number of photons attenuated. By contrast, maintaining a fixed composition of breast tissue, the difference in photon attenuation was 45.0%, averaging 6.0% for each additional centimeter in the breast thickness.

CONCLUSION: Monte Carlo simulation showed that the influence of the thickness of the breasts in the attenuation of photons in myocardial scintigraphy with ^99mTc is much greater than the influence of their compositions.

Key words: Breast/radionuclide imaging; myocardial reperfusion; injury: Monte Carlo method.

Introduction

Recent studies have shown a growing concern about the radiation dose associated with medical imaging procedures^1,2. In the United States, the myocardial perfusion scintigraphy accounts for over 22.0% of total effective dose of radiation to the non-elderly adult population¹. Furthermore, due to the medical imaging procedures, almost 20.0% of men and 18.0% of women receive doses of up to 20 mSv per year, maximum annual allowable limit for workers exposed to ionizing radiation³.

A major limitation to the accuracy of myocardial scintigraphy is the attenuation of photons by soft tissues⁴. There is a consensus that the attenuation correction techniques reduce the number of false-positive myocardial perfusion examinations^5-7. The techniques of attenuation correction of photons consist in using coupled external radiation sources to generate non-uniform correction maps or using mathematical procedures that enable the attenuation correction, while the former are the most widely used procedures in clinical practice^8-10.

Because the methods used for attenuation of correction cause exposure to extra doses of radiation, studies of dose calculations with scintigraphic apparatuses coupled to computed tomography devices with low dose of X-rays have suggested an increase of up to 10.0% in the effective radiation dose for each patient¹⁰.

Monte Carlo simulation techniques have become very important in the study of medical physics in the last 50 years. This is a mathematical model used to evaluate various parameters of nuclear medicine images, since there is no analytical solution to solve the equations that describe the interaction of photons with non-uniform attenuating structures of the body and the complex geometry of detectors and collimators¹¹.

The purpose of this study is to determine the influence of photon attenuation by the breast, in myocardial perfusion studies with ^99mTc-sestamibi for breasts of different compositions and volumes, taking into consideration factors such as composition (affected by age) and thickness, using the Monte Carlo method to undertake the attenuation correction by mathematical procedures.

Methods

All simulations were performed using the EGS4 code (Electron Gamma Shower - version 4)¹². The study of photon attenuation by the breast in myocardial perfusion studies was performed using ^99mTc as a radiation source¹³.

In the supine position, in which the myocardial perfusion study is conducted, the breasts take a pyramidal shape, which by analogy can be simplified to a cubic form. This form was adopted in the reproduction. The cube had its side ranging from 5 cm to 10 cm at intervals of one centimeter, in order to simulate breasts of different sizes.

The breasts are heterogeneous and composed of several basic tissue components, with known elemental compositions¹⁴. Since some simplifications may occur, it is acceptable that the breasts be properly represented by a sum of its two main components: adipose and fibroglandular tissue. In this study, we considered only the presence of these two key tissues for the end of the simulations.

Knowing the chemical composition and density of each of these compounds, we may properly represent the breasts using equations (1) and (2):

Where C_m is the chemical composition of a breast, f_a is the fraction of adipose tissue weight, C_a is the chemical composition of adipose tissue, f_g is the weight fraction of fibroglandular tissue, C_g is the chemical composition of fibroglandular tissue, r_m m is the density of a breast, r_a is the density of adipose tissue and r_g is the density of fibroglandular tissue.

From equations (1) and (2), we simulated six types of breasts. Such simulations were aimed at checking not only the influence of thickness, but also the effects of the composition of the breasts in the attenuation of photons. The percentage of adipose was changed from 0 to 100.0%, at intervals of 20.0%. Hence, we adopted 0, 20, 40, 60, 80 and 100.0% of adipose tissue in the composition of a breast. Consequently, 100, 80, 60, 40, 20 and 0.0% of fibroglandular tissue, respectively. The densities and weight fractions for the elemental composition of fibroglandular and adipose tissue were taken from ICRU 44¹⁴ and are shown in Box 1.

Considering the breasts as being composed solely of adopose tissue, six simulations were performed. Each for a different size of cube thickness. Then, the procedure was repeated for the other situations. In all cases, we considered a beam of photons falling directly into a breast (cube) with energy of 140 keV.

Most SPECT equipment use scintillation crystals of sodium iodide with thallium impurities (NaI [Tl]). Once the photons need to lose energy in the crystal to be registered, only a small portion of these photons is utilized due to the poor detection characteristics of NaI(Tl)¹⁵. Furthermore, due to the attenuation of the medium between the source and the detector, most photons reach the scintillation crystal with energy below 140 keV. Therefore, it is customary to adopt a window around the photopeak energy. In this study, we considered an energy window¹⁶ of 14.0%.

Data are presented as mean ± standard deviation. Tests were used to calculate the linear correlation coefficient to evaluate the association between the number of photons attenuated and various breast thicknesses and compositions. The Student t test for unpaired data was used to analyze the average of undetected photons according to the varying thickness and composition of the breasts.

Probability values <0.05 were considered statistically significant.

Results

The first analysis was the correlation between the percentage of photons that crossed the breasts and the thicknesses and characteristics of these. Figure 1 shows the relationship between the number of photons crossing the breast, expressed in percentage, with their energies in a breast with 100.0% fibroglandular tissue and different tissue thicknesses. We observe that in the ^99mTc (140 keV) photopeak energy range, the thickness of 5 cm of fibroglandular tissue allows the passage of 55.7% of photons with energy of 140 keV. By varying such thickness to 10 cm, the passage of photons with 140 keV is reduced to 30.8%, showing a reduction of approximately 45.0% of previous values.

When we do the same calculations for photons with energy lower than the photopeak (< 140 keV), we observed that increased thickness of the gland from 5 cm to 10 cm leads to an increase in the percentage of photons recorded with energies < 140 keV, an effect opposite to that seen with photons of 140 keV. For the energy of 100 keV, the percentage of photons recorded increases from 12.2% to 17.2%.

Table 1 lists the percentage of undetected photons within the window of energy according to the attenuation of a breast. In that table, we consider variations in breast thickness and tissue composition. In Table 1, we see that for a given composition of a breast, there is an increase of almost 6.0% in the number of photons lost by increasing breast thickness from 5 to 6 cm. However, keeping the thickness unchanged, a variation of 20.0% in the composition of glandular tissue does not cause significant differences in the number of photons lost (the maximum loss was 0.6%).

Thumbnail

In an extreme case, by changing from one breast only composed of adipose tissue to one composed of fibroglandular tissue only increases the loss of photons by 2.3% at the most. This value is comparatively much smaller than the variation of only one-centimeter increase in the thickness of a breast (Table 1).

The loss of photons reaches more than 36.0% for a 5 cm breast thickness, reaching 60.5% when the breast is 10 cm thick. Such loss may induce an area of low uptake in areas that should be of normal uptake.

There was a strong correlation, inversely proportional, between the variation of the mammary gland content and the number of photons attenuated: r = - 0.999 for a thickness of 5 cm and r = -0.989 for a thickness of 10 cm.

Comparing the average number of undetected photons for thicknesses of 5 and 10 cm, we observed a significant increase in the numbers of undetected photons (35.45 ± 0.68 versus 59.48 ± 0.84), p < 0.00001. Comparing a variation of one centimeter in thickness, from 5 to 6 cm, we observed a significant increase in the average number of undetected photons (35.45 ± 0.68 versus 41.08 ± 0.78), p < 0.00001.

However, doubling the percentage of fibroglandular tissue from 20 to 40.0% (47.62 ± 9.02 versus 48.03 ± 9.04) and from 40 to 80.0% (48.03 ± 9.04 versus 48.77 ± 9.03), the average number of undetected photons had no significant changes, p = 0.4689 p = 0.4455, respectively. Considering an extreme case in which the percentage of fibroglandular tissue goes from 0 to 100.0% (47.03 ± 8.90 versus 49.15 ± 9.06), there are no statistically significant changes in the average number of undetected photons, p = 0.3459.

Discussion

Our results demonstrate that the Monte Carlo simulation can be extremely useful in analyzing the effects of photon attenuation in SPECT myocardial scintigraphy. We found that tests employing ^99mTc sestamibi have as their main determinant factor for the attenuation of photons from the interaction with a breast, the thickness and, to a lesser extent, composition.

A significant limitation for obtaining images of myocardial perfusion through scintigraphy is the frequent presence of attenuation artifacts in the images resulting from the interaction of photons during the passage through the soft tissues of the body. Common causes of attenuation artifacts are the breasts in women and subdiaphragmatic tissues in men¹⁷.

The Monte Carlo technique has been widely applied to studies on the influence of breast in imaging examinations^18-20 and the calculation of doses resulting from computed tomographies²¹ or radiotherapy²². Studies on the Monte Carlo technique have also been very valuable in the area of nuclear medicine²³. The application of these simulations to analyze the effects of photon attenuation in myocardial scintigraphy may contribute to the development of the quality of images and enable simulations that would be little practical in experimental studies with exposure to radiation for patients and researchers.

The real purpose of attenuation correction is to reduce the wide variability in the distribution of counts in myocardial walls²⁴. Hence, one of the potential applications of our study is the use of attenuation indexes to correct the effects of the breasts on the scintigraphic image by generating unit correction factors. With these factors, we can determine the attenuation for each type of breast, without any additional tests. The creation of these attenuation algorithms will enable a reduction or even the elimination of additional tests, which have been used to obtain the attenuation maps, which, besides exposing the patients to additional doses, increase the procedure time, maximizing the number of patients scanned on a day⁴.

In conclusion, the results showed that the influence of breast thicknesses in the attenuation of photons in myocardial scintigraphy with ^99mTc is much greater than the influence of their compositions. New studies, however, should be made to extrapolate these data into models of clinical applicability for correction of breast attenuation in myocardial perfusion scintigraphy.

Potential Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Sources of Funding

There were no external funding sources for this study.

Study Association

This study is not associated with any post-graduation program.

References

1. Fazel R, Krumholz HM, Wang Y, Ross JS, Chen J, Ting HH, et al. Exposure to low-dose ionizing radiation from medical imaging procedures. N Engl J Med. 2009; 361 (9): 849-57.
2. Alessio AM, Kinahan PE, Manchanda V, Ghioni V, Aldape L, Parisi MT. Weigth-based, low-dose pediatric whole-body PET/CT protocols. J Nucl Med. 2009; 50 (10): 1570-7.
³
Ministério da Ciência e Tecnologia. CNEN. Comissão Nacional de Energia Nuclear. (CNEN). CNEN NN - 3.01. Diretrizes básicas de proteção radiológica, Brasília; 2005.
4. Hendel RC, Corbett JR, Cullom SJ, DePuey EG, Garcia EV, Bateman TM. The value and practice of attenuation correction for myocardial perfusion SPECT imaging: a joint position statement from the American Society of Nuclear Cardiology and the Society of Nuclear Medicine. J Nucl Cardiol. 2002; 9 (1): 135-43.
5. Hendel RC, Berman DS, Cullom SJ, Follansbee W, Heller GV, Kiat H, et al. Multicenter clinical trial to evaluate the efficacy of correction for photon attenuation and scatter in SPECT myocardial perfusion imaging. Circulation. 1999; 99 (21): 2742-9.
6. Garcia EV, Esteves FP. Attenuation corrected myocardial perfusion SPECT provides powerful risk stratification in patients with coronary artery disease. J Nucl Cardiol. 2009; 16 (4): 490-2.
7. Duvernoy CS, Ficaro EP, Karabajakian MZ, Rose PA, Corbertt JR. Improved detection of left main coronary artery disease with attenuation-corrected SPECT. J Nucl Cardiol. 2000; 7 (6): 639-48.
8. Corbett JR, Ficaro EP. Clinical review of attenuation-corrected cardiac SPECT. J Nucl Cardiol. 1999; 6 (1 pt 1): 54-68.
9. Heller GV, Batemam TM, Johnson LL, Cullom SJ, Case JA, Galt JR, et al. Clinical value of attenuation correction in stress-only 99mTc sestamibi SPECT imaging. J Nucl Cardiol. 2004; 11 (3): 273-81.
10. Sawyer LJ, Starritt HC, Hiscock SC, Evans MJ. Effective doses to patients from CT acquisitions on the GE Infinia Hawkeye: a comparison of calculation methods. Nucl Med Commun. 2008; 29 (2): 144-9.
11. Silva MTS, Silva AMM. Simulações de Monte Carlo do modelo antropomórfico Zubal em aquisições de SPECT cerebral. In: 8ş Congresso Brasileiro de Física Médica. Anais. Porto Alegre: ABFM/PUCRS; 2003. p. 450-3.
12. Love PA, Lewis DG, Al-Affan IAM, Smith CW. Comparison of EGS4 and MCNP Monte Carlo codes when calculating radiotherapy depth doses. Phys Med Biol. 1998; 43 (5): 1351-7.
13. Ljungberg M, Strand SE. A Monte Carlo program simulating scintillation camera imaging. Compute Methods Programs Biomed. 1989; 29 (4): 257-72.
¹⁴
International Commission on Radiation Units and Measurements (ICRU) 44. Tissue substitutes in radiation dosimetry and measurement. ICRU Report 44. Bethesda, MD: ICRU; 1989.
15. Melcher CL. Scintillation crystals for PET. J Nucl Med. 2000; 41 (6): 1051-5.
16. Patton JA, Turkington TG. Coincidence imaging with a dual-head scintillation camera. J Nucl Med. 1999; 40 (3): 432-41.
17. Pádua RDS, Oliveira LF, Azevedo-Marques PM, DeGroote JJ, Castro AA, Wichert-Ana L, et al. Auxílio à detecção de anormalidade perfusional miocárdica utilizando atlas de SPECT e registro de imagens: resultados preliminares. Radiol Bras. 2008; 1: 397-402.
18. Boone JM. Dose spread functions in computed tomography: a Monte Carlo study. Med Phys. 2009; 36 (10): 4547-54.
19. Ng KP, Kwork CS, Tang FH. Monte Carlo simulation of x-ray spectra in mamography. Phys Med Biol. 2000; 45 (5): 1309-18.
20. Boone JM, Cooper NV 3rd. Scatter/primary in mammography: Monte Carlo validation. Med Phys. 2000; 27 (8): 1818-31.
21. Taibi A, Royle GJ, Speller RD. A Monte Carlo simulation study to investigate to potential of diffraction enhanced breast imaging. IEEE Transactions on Nuclear Science. 2000; 47: 1581-6.
22. Chen SW, Wang XT, Chen LX, Tang Q, Liu XW. Monte Carlo evaluations of the absorbed dose and quality dependence of Al2O3 in radiotherapy photon beams. Med Phys. 2009; 36 (10): 4421-4.
23. Silva MTS. Avaliação da quantificação em SPECT cardíaco utilizando mapas de atenuação com borramento tipo gaussiano. 2007. [Dissertação]. Porto Alegre: Pontifícia Universidade Católica do Rio Grande do Sul; 2007.
24. Chalela W, Meneghetti J. Correção de atenuação para sistemas SPECT e PET. Arq Bras Cardiol. 2006; 86 (supl.1): 21-4.