Abdominal Imaging
ª Springer Science+Business Media, LLC 2009 Published online: 24 September 2009
Abdom Imaging (2010) 35:596–601 DOI: 10.1007/s00261-009-9568-x
CT colonography versus double-contrast barium enema for screening of colorectal cancer: comparison of radiation burden Emanuele Neri,1 Lorenzo Faggioni,1 Francesca Cerri,1 Francesca Turini,1 Simone Angeli,1 Lorenzo Cini,1 Franco Perrone,2 Fabio Paolicchi,1 Carlo Bartolozzi1 1 2
Diagnostic and Interventional Radiology, University of Pisa, Via Paradisa, 2, 56100 Pisa, Italy Department of Physics, University of Pisa, Via Roma, 67, 56100 Pisa, Italy
Abstract Our aim is to compare the radiation dose associated with a low-dose CT colonography (CTC) protocol for colorectal cancer screening with that delivered by double-contrast barium enema (DCBE). CTC of twenty asymptomatic individuals (M:F = 10:10) participating to a colorectal cancer screening program and DCBE of fifteen patients (M:F = 6:9) were evaluated. For CTC, absorbed dose was determined by calculating the dose-length product for each CTC examination from measurements on a CT dose phantom equipped with a CT ion chamber. For DCBE, the free-in-air Kerma at the patient’s X-ray entry surface and the Kerma-area product during fluoroscopy and fluorography were measured with a Barracuda system, with fluoroscopy times being recorded blinded to the performing operator. Effective dose at CTC was 2.17 ± 0.12 mSv, with good and excellent image quality in 14/20 (70%) and 6/20 cases (30%), respectively. With DCBE, effective patient dose was 4.12 ± 0.17 mSv, 1.9 times greater than CTC (P < 0.0001). Our results show that effective dose from screening CTC is substantially lower than that from DCBE, suggesting that CTC is the radiological imaging technique of the large bowel with the lowest risk of stochastic radiation effects. Key words: Computed tomographic colonography— Multidetector computed tomography—Double-contrast barium enema—Radiation dose—Colorectal cancer screening
Colorectal cancer (CRC) is the second most common cause of cancer among men aged 40–79 years [1]. Computed tomographic colonography (CTC) is a modern Correspondence to: Emanuele Neri; email:
[email protected]
powerful technique for the evaluation of the entire colon, with a potential advantage over conventional optical colonoscopy and double-contrast barium enema (DCBE) due to its minimal invasiveness, that has recently been included by the American Cancer Society, the American College of Radiology, and the U.S. Multi-Society Task Force as a recommended option for detection of colorectal polyps and cancer [2]. On these grounds, it has been postulated that regular performance of CTC-based screening programs would produce a significant increase of CTC exams, leading to higher workload for radiologists and extra costs for national health care services, as well as to an increased radiation dose to the population [3]. This latter aspect is of particular relevance if one considers that asymptomatic individuals would be evaluated, thus raising the issue of radiological risk in this subject category. However, few data have been published in the literature on the effective dose associated with CTC. In addition, the widespread availability of multidetector CT (MDCT) scanners has determined, along with a significant improvement of image quality owing to the possibility to acquire images with submillimeter spatial resolution in all spatial directions, a trend to a further increase in radiation dose compared with single-slice scanners due to overbeaming and overranging effects. The aim of this study is to evaluate the radiation dose associated with a low-dose CTC acquisition protocol intended for CRC screening in an asymptomatic population and to compare radiation exposure related to CTC with that delivered by DCBE examination.
Materials and methods CT colonography Patient selection and preparation. Twenty asymptomatic individuals [age 55 ± 14 years (mean ± standard deviation), male: female = 10:10] participating to a CRC
597
E. Neri et al.: CT colonography versus double-contrast barium
screening program were selected for CTC between June 2008 and July 2008. The study was conducted upon approval of the local Institutional Review Board. Inclusion criteria for CTC were age greater than 50 years or familial history of CRC, while exclusion criteria were real or presumed pregnancy and diagnosis of acute inflammatory large bowel disease. Patients were instructed to follow a residue-free diet for 3 days before CTC examination and to assume a mild laxative (polyethyleneglycol: PEG macrogol 3350, MovicolÒ, NORGINE ITALIA Srl) starting 3 days prior CTC for colon cleaning purposes. Faecal tagging was also performed in order to improve diagnostic accuracy through oral administration of 50 mL iodinated contrast material (diatrizoate meglumine: GastrografinÒ, Bracco Diagnostics, Milan, Italy) 3 h before CTC. Immediately before CTC, an antispasmodic agent (hyoscine-N-butylbromide 20 mg, BuscopanÒ; Boehringer Ingelheim, Ingelheim, Germany) was administered in order to maximize colonic distention, reduce image artifacts, and improve patient comfort. Colonic distention was obtained by means of an automatic insufflator (Colon Insufflator System; E-Z-EM, NY) by gentle insufflation of 3.5 L carbon dioxide, after which a CT scout view of the abdomen was obtained in order to assess the degree of colonic distention. If this latter was not adequate, insufflation was repeated according to patient’s tolerance. CT scan parameters. CTC data were acquired on a 64row MDCT scanner (Lightspeed VCT, GE Healthcare, Milwaukee, WI) in both supine and prone position with the following parameters: tube voltage 120 kV, automatic tube current modulation (range 20–80 mA, noise index 50, z-axis and angular modulation), rotation time 500 ms, beam pitch 1.375:1, detector configuration 64 9 0.625 mm, slice thickness 1.25 mm, reconstruction increment 1 mm, standard reconstruction kernel. Scanning was performed in cranio-caudal direction during a single breath-hold covering the entire colon from the diaphragm dome to the pubic symphysis and the values of scan length (expressed in mm), CTDIvol (expressed in mGy), and DLP (expressed in mGy cm) relative to a 32-cm-large body phantom were automatically recorded by the scanner in a final dose log DICOM image. CTC images were exported in DICOM format via PACS to two dedicated workstations (CADCOLON, Im3D, Turin, Italy, and XELIS-COLON, Infinitt, Seoul, Korea) for 2D and 3D analysis including evaluation of axial source and multiplanar reformation (MPR) images, virtual dissection (VD) reconstructions, and virtual endoscopy (VE) views. Image quality was visually assessed by two readers with previous experience in CTC (EN, LF), reading CTC datasets in consensus with each other.
Estimation of CTC radiation dose. For CTC the absorbed dose was determined by calculating the dose-length product (DLP) from the acquisition parameters of each CTC examination as estimated on measurements on a CT dose phantom (Model 76–415 of Nuclear Associates, Carle Place, NY) equipped with a CT ion chamber (Barracuda System, RTI Electronic AB, Sweden). Effective dose values were obtained by multiplying DLP values for the proper standardized coefficient suggested in EUR 16262 EN of European guidelines for CT (0.015 mSv mGy-1 cm-1 for abdominal CT examinations) [4].
DCBE examination Patient selection and imaging protocol. DCBE examinations performed on 15 patients (6 male, 9 female, age 57 ± 12 years) were recorded in blind fashion to the performing operator in order to determine fluoroscopy time. DCBE was carried out using a MTX20E orthoclinoscope (Italray Srl, Florence, Italy) with the parameters and the projections (modified Altaras protocol) listed in Table 1 for computerized digital radiography (Computer Radiography FCR, FUJIFILM Corporation, Tokyo, Japan). Fluoroscopy was run at a tube voltage of 80 kV and a mA-second product value of 2 mAs; mean fluoroscopy time was 384 ± 80 s (range 280–520 s). Estimation of DCBE radiation dose. The free-in-air Kerma at the X-ray entry surface of the patient (Entrance Surface Air Kerma, ESAK) and the Kerma-area product (KAP) during fluoroscopy and fluorography were measured with a Barracuda System (RTI Electronic AB, Sweden). The entrance skin dose (ESD) was calculated as Eq. 1: ESD ¼ ESAK BSF ½len=qwa
ð1Þ
where BSF is the backscattering factor (which is assumed equal to 1.35) and [len/q]wa is the energy absorption Table 1. DCBE projections with their corresponding absorbed dose values Projection
kV
mA
Estimated entrance dose (mGy)
Direct Full column Caudocranial L-L Rectum (L-L) PA Craniocaudal Cecuma AP (orthostasis) Flexures
75 75 89 89 100 75 97 75 75 82
24 24 40 40 50 24 40 24 24 24
1.616 1.616 3.794 3.794 5.987 1.616 4.506 1.616 1.616 1.932
a The cecum projection was acquired twice with external compression and different patient inclination (supine and 20–30° extrarotation) in order to get complete depiction of the cecum L-L latero-lateral, PA postero-anterior, AP antero-posterior
E. Neri et al.: CT colonography versus double-contrast barium
598
coefficient relative to water and air, equaling 1.024. ESAK is obtained from the following empirical formula (Eq. 2): ESAK ¼ a mAs kV2p ðd=PFDÞ2
Table 3. Values of field-of-view, total scan length, DLP, and effective dose of each CTC examination Patients
Field of view (mm)
Scan length (mm)
DLP (mGy cm)
Effective dose (mSv)
300 300 300 300 300 300 300 380 300 360 380 400 380 360 360 346 360 349 341 300
1081 874 760 910 780 840 901 860 901 832 812 802 820 840 840 835 851 724 753 820
154.0 150.2 133.2 155.7 136.2 159.0 144.2 148.4 154.4 144.4 141.4 139.9 142.6 145.7 145.2 144.4 146.9 127.7 133.2 142.3
2.31 2.25 2.00 2.34 2.04 2.39 2.16 2.23 2.32 2.17 2.12 2.10 2.14 2.19 2.18 2.17 2.20 1.92 2.00 2.13
ð2Þ
where kVp is peak tube kilovoltage, PFD is the patientto-focus distance, and d is the distance at which the a parameter is measured; this latter is an X-ray equipmentspecific function that is weakly dependent from the kVp value. This formalism was established based on empirical data [5] and was validated several times from measurements. Combining Eqs. 1 and 2, the following expression is obtained (Eq. 3): ESD ¼ a mAs kV2p ðd=PFDÞ2 1:38
ð3Þ
The relative dose distribution was determined with an Alderson Rando phantom for a standard reference 80 kV X-ray beam with a 20 9 20 cm2 field-of-view and a 110 cm focus-to-skin distance by means of a XR-QA Gafchromic Film (International Specialty Products, Wayne, NJ). The average organ doses for a standard hermaphrodite patient model were estimated and converted into organ equivalent doses and effective doses by means of the appropriate conversion factors [4] (Table 2).
Statistical analysis For each CTC and DCBE examination, effective dose values (expressed in mSv) were recorded as mean ± standard deviation. Effective dose values for each group were compared by means of the two-tailed Mann– Whitney test; a P value less than 0.05 was considered statistically significant. Statistical analysis was carried out using software (GraphPad Prism 5.0b, GraphPad Software Inc, La Jolla, CA).
Results Values of field-of-view, total scan length, DLP, and effective dose of each CTC examination are tabulated in Table 3. Effective dose at CTC was 2.17 ± 0.12 mSv. All Table 2. Calculated distribution of radiation dose to target organs for DCBE examination Organs
Average absorbed dose (mGy)
Average effective dose (mSv)
Whole abdomen Colon Spine Kidneys Gonads Stomach Red marrow Skin Liver Other tissues
34.289 26.096 43.582 26.137 2 0.925 0.620 1.215 0.925 0.000
N/A 3.131 0.218 0.131 0.400 0.111 0.074 0.012 0.046 0
#1 #2 #3 #4 #5 #6 #7 #8 #9 #10 #11 #12 #13 #14 #15 #16 #17 #18 #19 #20
CTC images were of diagnostic quality; in particular, CTC image quality was judged as good in 14/20 cases (70%) and excellent in the remaining 6/20 (30%) (Figs. 1 and 2). With DCBE, patient average total entrance dose to patient resulted to be 83.2 mGy. Effective patient dose was 4.12 ± 0.17 mSv, i.e., 1.9 times greater than that associated with CTC (P < 0.0001).
Discussion Our data show that CTC performed following a lowradiation protocol allows to yield diagnostic images of the large bowel with a radiation burden (both in terms of absorbed and effective dose) lower than that associated with conventional DCBE. Few studies have compared the radiation dose of CTC and DCBE [6, 7]: in particular, Hirofuji et al. [6] measured a DCBE effective dose value of 12.7 mSv (decreasing by 12% when digital radiography equipment was used), while effective dose of CTC performed with a low-dose protocol was 5.7 mSv. The average effective dose value that we found for CTC is lower (2.17 mSv), partly due to the fact that automatic modulation of tube current was set up (corresponding, with the acquisition parameters used, to a mA value ranging between 7.3 and 29.1 mA), while Hirofuji et al. [6] employed a 27 mA protocol. The possibility to perform CTC with a substantially lower radiation dose while retaining adequate image quality by means of automated z-axis and angular tube current modulation, as with our acquisition protocol, has been demonstrated in the literature [8]. Our average CTC effective dose is also substantially lower than that reported by other authors: for instance, Macari et al. [9] delivered a mean effective dose
E. Neri et al.: CT colonography versus double-contrast barium
599
Fig. 1. CTC showing an 11-mm polyp located in the descending colon: A MPR on sagittal plane, B magnified view of (A), C virtual endoscopy view.
of 5.0 mSv for men and 7.8 mSv for women by using a four-row scanner and a low-dose, fixed tube current CTC acquisition protocol, while Schopphoven et al. [10] estimated effective dose from CTC to be as high as 7.5 mSv for male and 10.2 mSv for female individuals. The higher radiation dose of DCBE compared with CTC is likely to be partly due to the relatively large number of radiographic projections needed to get a complete depiction of the various segments of the large bowel. In addition, a substantial fraction of the radiation dose delivered to the patient during a DCBE examination is accounted for by fluoroscopy time, which is rarely shorter than 5 min and is necessary as to follow contrast medium replenishment and emptying of the colon, as well as to select the most appropriate patient position for the various radiographic projections. Such drawbacks are absent in CTC, as fluoroscopy is not needed and data acquisition is performed in the sole supine and prone position. Incidentally, it is under debate whether single-acquisition protocols can be used (i.e., in the paediatric population), which would imply half radiation exposure [11]. In addition, several works have been published that indicate the possibility to reduce CTC-related radiation exposure even further by using dedicated image reconstruction filters [12, 13]. In a study performed with a 4-row MDCT scanner [14], an acquisition protocol consisting of 120 kV tube voltage, 50 mA tube current, collimation of 1.25 mm with a detector pitch of 6 (i.e., beam pitch of 1.5) was suggested to be optimal for depiction of polyps sized 5 mm or larger, while Laghi et al. [15] have underscored the importance of thin collimation (down to 1 mm) with high pitch in order to maximize lesion detectability while reducing scan time and radiation exposure. On the other hand, McCullough et al. [16] have stated that thin collimation and low-noise
filters are not necessary for detection of polyps of clinically relevant sizes, potentially lessening the lower radiation dose limit affording diagnostic quality imaging. The advent of MDCT scanners with up to 64 detector rows allows for even faster imaging with submillimetric spatial resolution in the z-axis with the benefit of voxel isotropy, leading to better quality of 2D and 3D image reconstructions, as well as to a potentially higher polyp detection rate. Moreover, scanners with 16 detector rows and beyond have a progressively increased geometric efficiency due to reduction of overbeaming [17], thus reducing radiation dose at narrow collimation thicknesses in comparison with older generation MDCT equipment. It is also true that 64-MDCT scanners have a larger X-ray beam in the z-axis, implying greater overranging and penumbra effect than scanners with narrower beam width (such as 4-MDCT and 16-MDCT equipment); however, the difference between 64-MDCT and 16-MDCT scanners working at narrow collimation thickness is slight [18]. According to the radiological risk figures of the International Commission on Radiological Protection [4], the average effective dose value of 2.17 mSv that we found for CTC is associated with a theoretical 0.005% risk of lethal radiation-induced malignancy, which further decreases with increasing patient age. However, some differences exist between our study and the methods used by other authors for quantifying radiation dose. First, in our work the average patient absorbed dose was calculated from a pool of individuals, including cases in which higher-than-normal tube current values were used due to patient’s size, while Jensch et al. [19] and Liedenbaum et al. [20] computed the absorbed average patient dose from standard CTC protocols of each institution involved in the respective studies.
600
E. Neri et al.: CT colonography versus double-contrast barium
Fig. 2. CTC showing a 7-mm polyp localized in the cecum (arrow): A axial source image, B sagittal MPR, C coronal MPR, D virtual endoscopy view (asterisk). As can be seen,
high image noise due to the usage of a low radiation dose protocol does not prevent correct visualization of the lesion.
The effective dose value that we found for DCBE (4.12 mSv) is in line with literature data. For instance, Geleijns et al. [21] estimated a dose-area product (DAP) of 12.8 Gy cm2 and an effective dose of 2.6 mSv for men and 4.2 mSv for women by using a Monte Carlo-based algorithm, while Kemerink et al. [7] reported for DCBE an effective dose of 6.4 mSv. In this latter work, a relatively simple method was used for dose measurements, consisting of calculating DAP over the hypothetical trajectories covering the portion of the abdomen under investigation. Dose distribution had been established by measuring entrance dose at the center of every projection with thermoluminescent dosimeters placed on the patient’s skin; DAPto-effective-dose conversion coefficients based on Monte Carlo tables were then used to obtain the effective dose value from DAP figures from each projection. Our study has a potential limitation in that patients in both the DCBE and CTC group were not size-matched. Indeed, evidence exists [22] that radiation dose tends to be substantially higher in individuals with higher body weight, exceeding 100 mSv in extremely obese patients. However, the subjects included in both our experimental groups had similar weights (ranging from 68 to 83 kg), so that it can be assumed that our radiation dose values are, in this respect, representative of a typical population. In
addition, CTC image quality was not appreciably different among patients, even though the maximum tube current value allowed (80 mA) was the same for all CTC examinations. In conclusion, we found that the effective dose to individuals undergoing screening CTC is considerably lower than that related to DCBE (around 53%), suggesting that CTC is the radiological imaging technique of the large bowel with the lowest risk of stochastic radiation effects. These findings, together the high diagnostic confidence of CTC (comparable to that of conventional colonoscopy), support the usage of CTC as an appropriate and effective tool for CRC screening, definitively alternative and substitutive of DCBE. References 1. Jemal A, Murray T, Ward E, et al. (2005) Cancer statistics, 2005. CA Cancer J Clin 55(1):10–30 2. Nelson NJ (2008) Virtual colonoscopy accepted as primary colon cancer screening test. J Natl Cancer Inst 100(21):1492–1499 3. Brenner DJ, Georgsson MA (2005) Mass screening with CT colonography: should the radiation exposure be of concern? Gastroenterology 129(1):328–337 4. International Commission on Radiological Protection (1991) 1990 recommendations of the international commission of radiological protection, publication no. 60. In: Smith H, (ed). Annals of the ICRP, vol 21. Oxford: Pergamon Press, pp 1–3
E. Neri et al.: CT colonography versus double-contrast barium
5. Hospital Physicists’ Association (Diagnostic Radiology Topic Group) (1977) The physics of radiodiagnostics: scientific report series, vol 6. London 6. Hirofuji Y, Aoyama T, Koyama S, Kawaura C, Fujii K (2009) Evaluation of patient dose for barium enemas and CT colonography in Japan. Br J Radiol 82:219–227 7. Kemerink GJ, Borstlap ACW, Frantzen MJ, et al. (2001) Patient and occupational dosimetry in double contrast barium enema examinations. Br J Radiol 74:420–428 8. Graser A, Wintersperger BJ, Suess C, Reiser MF, Becker CR (2006) Dose reduction and image quality in MDCT colonography using tube current modulation. AJR Am J Roentgenol 187:695–701 9. Macari M, Bini EJ, Xue X, et al. (2002) Colorectal neoplasms: prospective comparison of thin-section low-dose multi-detector row CT colonography and conventional colonoscopy for detection. Radiology 224:383–392 10. Schopphoven S, Faulkner K, Busch HP (2008) Assessment of patient organ dose in CT virtual colonoscopy for bowel cancer screening. Radiat Prot Dosimetry 129:179–183 11. Sugiyama A, Ohashi Y, Gomi A, et al. (2007) Colorectal screening with single scan CT colonography in children. Pediatr Surg Int 23:987–990 12. Wang J, Wang S, Li L, et al. (2008) Virtual colonoscopy screening with ultra low-dose CT and less-stressful bowel preparation: a computer simulation study. IEEE Trans Nucl Sci 55:2566–2675 13. Branschofsky M, Vogt C, Aurich V, et al. (2006) Feasibility of ultra-low-dose multi-detector-row CT-colonography: detection of artificial endoluminal lesions in an in vitro-model with optimization of image quality using a noise reduction filter algorithm. Eur J Med Res 11:13–19
601
14. Taylor SA, Halligan S, Bartram CI, et al. (2003) Multi-detector row CT colonography: effect of collimation, pitch, and orientation on polyp detection in a human colectomy specimen. Radiology 229:109–118 15. Laghi A, Iannaccone R, Mangiapane F, et al. (2003) Experimental colonic phantom for the evaluation of the optimal scanning technique for CT colonography using a multidetector spiral CT equipment. Eur Radiol 13:459–466 16. McCollough CH (2002) Optimization of multidetector array CT acquisition parameters for CT colonography. Abdom Imaging 27:253–259 17. Dalrymple NC, Prasad SR, El-Merhi FM, Chintapalli KN (2007) Price of isotropy in multidetector CT. Radiographics 27:49–62 18. Luz O, Buchgeister M, Klabunde M, et al. (2007) Evaluation of dose exposure in 64-slice CT colonography. Eur Radiol 17:2616– 2621 19. Jensch S, van Gelder RE, Venema HW, et al. (2006) Effective radiation doses in CT colonography: results of an inventory among research institutions. Eur Radiol 16:981–987 20. Liedenbaum MH, Venema HW, Stoker J (2008) Radiation dose in CT colonography—trends in time and differences between daily practice and screening protocols. Eur Radiol 18:2222–2230 21. Geleijns J, Broerse JJ, Shaw MP, et al. (1997) Patient dose due to colon examination: dose assessment and results from a survey in the Netherlands. Radiology 204:553–559 22. Yanch JC, Behrman RH, Hendricks MJ, McCall JH. Increased radiation dose to overweight and obese patients from radiographic examinations. Radiology 2009. doi:10.1148/radiol.2521080141