- Email: [email protected]
Cumulative Radiation Dose in Patients With Hereditary Hemorrhagic Telangiectasia and Pulmonary Arteriovenous Malformations
Canadian Association of Radiologists Journal xx (2013) 1e6 www.carjonline.org
Thoracic and Cardiac Imaging / Imagerie cardiaque et imagerie thoracique
Cumulative Radiation Dose in Patients With Hereditary Hemorrhagic Telangiectasia and Pulmonary Arteriovenous Malformations Kate Hanneman, MDa,*, Marie E. Faughnan, MD, MScb,c,d, Vikramaditya Prabhudesai, MBBS, MS, FRCRa a
Department of Medical Imaging, St Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada b Division of Respirology, Department of Medicine, University of Toronto, Toronto, Ontario, Canada c Toronto Hereditary Hemorrhagic Telangiectasia Program, Division of Respirology, Department of Medicine, St Michael’s Hospital, Toronto, Ontario, Canada d Keenan Research Centre and the Li Ka Shing Knowledge Institute, St Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada
Abstract Purpose: To determine the cumulative effective dose (CED) of radiation from medical imaging and intervention in patients with hereditary hemorrhagic telangiectasia (HHT) who have pulmonary arteriovenous malformations and to identify clinical factors associated with exposure to high levels of radiation. Methods: All patients with at least 1 pulmonary arteriovenous malformation were identified from the dedicated patient database of a tertiary HHT referral center. Computerized imaging and electronic patient records were systematically examined to identify all imaging studies performed from 1989-2010. The effective dose was determined for each study, and CED was calculated retrospectively. Results: Among 246 patients (mean age, 53 years; 62.2% women) with a total of 2065 patient-years, 3309 procedures that involved ionizing radiation were performed. CED ranged from 0.2-307.6 mSv, with a mean of 51.7 mSv. CED exceeded 100 mSv in 26 patients (11%). Interventional procedures and computed tomography (CT) were the greatest contributors, which accounted for 51% and 46% of the total CED, respectively. Factors associated with high cumulative exposure were epistaxis (odds ratio 2.7 [95% confidence interval, 1.1-6.3]; P ¼ .02), HHT-related gastrointestinal bleeding (odds ratio 2.0 [95% confidence interval, 1.0-3.8]; P ¼ .04) and number of patient-years (P < .0001). Conclusions: Patients with HHT are exposed to a significant cumulative radiation dose from diagnostic and therapeutic interventions. Identifiable subsets of patients are at increased risk. A proportion of patients receive doses at levels that are associated with harm. Imaging indications and doses should be optimized to reduce radiation exposure in this population. Resume Objet: Determiner la dose efficace cumulative de radiation emanant des examens d’imagerie medicale et des interventions chez les patients atteints de telangiectasie hemorragique hereditaire et presentant des malformations pulmonaires arterio-veineuses, et definir les facteurs cliniques associes a une forte radioexposition. Methodes: Tous les patients presentant au moins une malformation pulmonaire arterio-veineuse qui figuraient dans la base de donnees sur les patients d’un centre de soins tertiaires specialise en telangiectasie hemorragique hereditaire ont ete identifies. Une revue systematique des images informatisees et des dossiers electroniques de ces patients a ensuite permis de relever tous les examens d’imagerie ayant ete realises entre 1989 et 2010. Enfin, la dose efficace a ete determinee pour chaque examen, puis a mene au calcul retrospectif de la dose efficace cumulative. Resultats: 3 309 interventions comportant de la radiation ionisante ont ete realisees chez 246 patients (^age moyen de 53 ans; 62,2 % de sexe feminin) representant un total de 2 065 annees-patients. Les calculs ont revele une dose efficace cumulative variant de 0,2 a 307,6 mSv et une moyenne de 51,7 mSv. Une dose efficace cumulative superieure a 100 mSv a ete constatee chez 26 patients (11 %). Par ailleurs, les interventions et les examens de tomodensitometrie (TDM) ont eu une incidence decisive sur les resultats, representant respectivement 51 % et 46 % de la dose efficace cumulative. Enfin, les facteurs suivants ont ete associes a une radioexposition cumulative elevee : l’epistaxis (rapport
* Address for correspondence: Kate Hanneman, MD, Department of Medical Imaging, St Michael’s Hospital, 30 Bond Street, Toronto, Ontario M5B 1W8, Canada.
E-mail address: [email protected] (K. Hanneman).
0846-5371/$ - see front matter Ó 2013 Canadian Association of Radiologists. All rights reserved. http://dx.doi.org/10.1016/j.carj.2013.02.007
2
K. Hanneman et al. / Canadian Association of Radiologists Journal xx (2013) 1e6
des cotes de 2,7 [intervalle de confiance de 95 %, de 1,1 a 6,3]; P ¼ 0,02), les saignements gastro-intestinaux lies a la telangiectasie hemorragique hereditaire (rapport des cotes de 2,0 [intervalle de confiance de 95 %, de 1,0 a 3,8]; P ¼ 0,04) et le nombre d’annees-patients (P < 0,0001). Conclusions: Les patients atteints de telangiectasie hemorragique hereditaire sont exposes a une dose cumulative significative emanant des interventions a visee diagnostique et therapeutique. Certains sous-groupes de patients presentent un risque accru. Une certaine proportion de patients a rec¸u des doses dont le niveau est associe a des effets nocifs. Les indications et les doses en matiere d’imagerie doivent ^etre optimisees afin de reduire la radioexposition que subit ce type de patients. Ó 2013 Canadian Association of Radiologists. All rights reserved. Key Words: Telangiectasia; Hereditary hemorrhagic; Arteriovenous malformations; Radiation; Ionizing; Radiation dosage; Multidetector computed tomography
Hereditary hemorrhagic telangiectasia (HHT), historically known as Rendu-Osler-Weber syndrome, is a rare hereditary disorder that may be complicated by arteriovenous malformations (AVM) in the brain, lung, gastrointestinal (GI) tract, and liver. Pulmonary AVMs (PAVM) are frequently present in HHT, occurring in 15%-45% of patients (Figure 1) [1]. Adults with untreated PAVMs are at risk of life-threatening hemorrhage and neurologic complications [1]. Routine screening for PAVMs in all patients with HHT is recommended, given the serious nature of complications that may occur and the availability of preventative treatment with transcatheter embolotherapy [2]. Given the potential for multiorgan involvement, as well as the role of interventional radiology, standard care for patients with HHT results in repeated imaging procedures, and patients may be exposed to a cumulative dose of radiation much higher than with the average patient. Unenhanced thoracic computed tomography (CT) is recommended to confirm the presence of PAVMs after an initial positive screening with transthoracic contrast echocardiography. Small PAVMs may be followed with repeated serial CTs because growth is demonstrated in up to 18% of small PAVMs, and these may eventually warrant treatment [2e4]. Significant PAVMs are treated with transcatheter embolotherapy, then follow-up CT is recommended at
12 months after treatment, and then every 1-3 years (depending on the size of residual PAVMs). Long-term follow-up is recommended to detect growth of previously small, insignificant PAVMs but also to detect reperfusion of embolized PAVMs [2]. However, high radiation exposure is an established cause of cancer, and there is evidence from observational studies of excess risks from fractionated exposures in the dose range that would be received from repeated CTs [5e8]. The purpose of this study was to quantify the cumulative effective dose (CED) associated with diagnostic, interventional, and follow-up imaging in a cohort of patients with HHT and with at least 1 PAVM, and to identify factors associated with exposure to high levels of diagnostic radiation. The lifetime radiation dose received by patients with HHT has not been studied previously. Materials and Methods Overview To characterize radiation exposure in this population, all patients with at least 1 PAVM were identified from the dedicated patient database of a tertiary HHT referral center. Demographic and clinical data were obtained from the
Figure 1. A 30-year-old woman with hereditary hemorrhagic telangiectasia. (A) Frontal chest radiograph and (B) coronal computed tomography image, demonstrating a right upper lobe pulmonary arteriovenous malformations (arrows).
Radiation dose in patients with HHT / Canadian Association of Radiologists Journal xx (2013) 1e6
database. The number of patient-years was calculated from the time of the first recorded imaging procedure to the study cutoff date. Patient genotype was recorded for patients for whom these data were available. High cumulative exposure to radiation was defined as CED that exceeded 50 mSv, which is an effective dose (ED) equivalent to approximately 500 standard chest radiographs [9]. This study was approved by the institution’s research ethics board. Subjects A total of 251 patients were identified as potentially eligible. Patients who had no recorded diagnostic or therapeutic imaging procedures were excluded (n ¼ 4). A patient who, on review of his clinical data, did not have a PAVM was also excluded (n ¼ 1). A total of 246 patients were included, 93 men (37.8%) and 153 women (62.2%). The mean age was 53 years (range, 17-92 years). The mean number of patientyears was 8.4 years (range, 1-22 patient-years). Methods Computerized imaging records (electronic radiology information system) were systematically examined in April 2011 to identify all imaging studies performed from January 1, 1989, to December 31, 2010, including fluoroscopyguided interventional procedures. Studies performed for non-HHTerelated indications and studies performed at other institutions that had been imported into the picture archiving and communication system were not excluded. The date of each study and the anatomic region examined were recorded. CTs were read to determine the following technical parameters when available: displayed kilovolt value, anodic current (in milliamperes), slice thickness, and dose length product (DLP). For CT studies in which DLP was available (338/982 CTs), ED was calculated by using published conversion coefficients [10]. For thoracic CTs in which DLP was the only imaging parameter not available, studies performed by using the same protocol (including kilovolt peak, anodic currant, and slice thickness) were assigned the calculated average ED of all other studies performed when using that same protocol for which DLP was available (66/982 CT studies). To estimate radiation exposure for CTs for which technical parameters were not available, and for all other imaging modalities that involved ionizing radiation, including interventional and fluoroscopic procedures and nuclear medicine studies, we obtained
3
estimates of EDs from the published literature [9]. Mean ED estimates for interventional procedures ranged from 1-15 mSv [9,11]. Statistical Analysis Data compilation and statistical analyses were performed by using Microsoft Excel 2007 (Microsoft Corporation, Redmond, WA), and Stata version 10.0 (StataCorp, College Station, TX). All continuous data are expressed as mean (standard deviation [SD]) unless noted otherwise. In addition to standard descriptive statistics, a generalized linear model was used to identify determinants of CED. To normalize the distribution of the residuals, CED was first log-transformed. Analyses of the difference between groups of nonparametric data, including genotype, were performed by using the KruskalleWallace test. Univariate analysis of the associations among patient clinical factors (including sex, the presence of cerebral and liver AVMs, and a history of epistaxis and GI bleeding), and high exposure (defined as CED higher than 50 mSv) were performed by using the Mantel-Haenszel test (expressed as odds ratios [OR]). Results Number of Studies Among 246 patients with a total of 2065 patient-years, 3309 procedures that involved ionizing radiation were performed, with a mean of 13.5 (range, 1-78) per patient. The distribution of the number of studies and CED according to imaging modality is presented in Table 1. Cumulative Dose The mean CED was 51.7 mSv, with an asymmetrical distribution (range, 0.2-307.6 mSv) (Figure 2). CED exceeded 50 mSv and 100 mSv in 84 (34%) and 26 (11%) patients, respectively. The mean CED per patient-year was 7.3 mSv (range, 0.03- 44.6 mSv). The ED per patient-year according to dose categories is presented in Table 2, with reference values for comparison. Imaging Interventional procedures and CT were the greatest contributors to CED, which accounted for 51% and 46%
Table 1 The number of procedures performed, total number of procedures and patients, and CED according to imaging modality Imaging modality
No. procedures, mean (range) Total no. procedures Total no. patients who received imaging Mean (range) CED, mSv CED ¼ cumulative effective dose.
Radiography
Computed tomography
Fluoroscopy and/or interventional
Nuclear medicine
All modalities
7.0 (0-56) 1734 243 0.9 (0-12.1)
4.0 (0-20) 982 243 23.8 (0-232.0)
2.2 (0-16) 552 213 26.2 (0-210)
0.2 (0-6) 43 27 0.8 (0-33.9)
13.5 (1-78) 3311 246 51.7 (0.2-307.6)
4
K. Hanneman et al. / Canadian Association of Radiologists Journal xx (2013) 1e6
Figure 2. Distribution of individual cumulative effective dose (CED) estimates for total study cohort (n ¼ 246). CED exceeded 50 mSv and 100 mSv in 84 (34%) and 26 (11%) patients, respectively.
of total CED, respectively (Figure 3). Plain radiographs represented 43% of the studies but accounted for only 2% of CEDs. In total, 982 CTs were performed with 243 patients, of which the majority were thoracic CTs (825 [84.0%]). The mean number of CTs per patient was 4.0 (range, 0-20). Only 3 patients had not undergone a CT. The mean (SD) CED from CTs was 23.8 23.3 mSv per patient. The mean (SD) ED per CT was 6.0 4.8 mSv. For studies in which DLP was available, the mean (SD) ED for standard dose thoracic CTs was 4.3 0.4 mSv (n ¼ 118) compared with 1.9 0.4 mSv for thoracic CTs performed with a low-dose protocol (n ¼ 135).
Clinical Factors Factors associated with high cumulative exposure were epistaxis (OR 2.7 [95% confidence interval, 1.1-6.3]; P ¼ .02]) and a history of a previous GI bleed (OR 2.0 [95% confidence interval, 1.0-3.8]; P ¼ .04) (Table 3). The number of patient-years is significantly related to CED (P < .0001). The CED per patient-year was not significantly associated with genotype (P ¼ .55).The mean CED per patient-year was 7.1 mSv (n ¼ 132), 6.9 mSv (n ¼ 27), and 15.0 mSv (n ¼ 7) for genotypes ENG (mutation of the endoglin gene), ALK1 (mutation of the activin receptor-like kinase 1 gene), and SMAD4, respectively. The genotype was not available in 33% of patients (n ¼ 80).
Discussion There is increasing acceptance of the linear, no-threshold hypothesis of radiation exposure, which raises concerns about the potential long-term effects of repeated exposure to radiation due to medical imaging [5]. The ED is defined by the International Commission on Radiological Protection as the sum of the absorbed doses in all tissues and organs of the body, each weighted according to its radiation sensitivity. ED allows for population-level comparisons across different types of radiation exposures. Interventional procedures were the highest contributor to total CED in the study cohort. Current guidelines recommend preventative treatment, with coil embolization therapy, of PAVMs with feeding artery diameters larger than 3 mm [2]. The average dose associated with pulmonary angiography and embolization is estimated at 5-15 mSv [9]. The mean number of embolization procedures per patient was 2.2 (range, 0-16). Patients may have multiple PAVMs that require more than 1 treatment session. Patients may also require repeated embolization because reperfusion occurs in up to 8%-19% of successfully occluded lesions [2e4]. The modality of CT was the second greatest contributor to total CED. Thoracic CT is indicated to confirm the presence of a PAVM, to evaluate for growth of documented lesions, and to assess for reperfusion of previously coiled lesions [2]. The mean number of CTs per patient was 4.0 (range, 0-20).
Table 2 CED per patient-year according to dose categories Dose category Low Moderate High Very high
Dose range (mSv/y) 3 > 3-20 > 20-50 > 50
CED ¼ cumulative effective dose.
Comparison Background level of radiation from natural sources[23]; estimated annual per-capita effective dose from medical procedures in the United States[24] Upper annual limit for occupational exposure for at-risk workers, averaged over 5 y[25] Upper annual limit for occupational exposure for at-risk workers in any given year[25] Reasonable evidence for an increase in cancer risk at protracted doses higher than 50 mSv[24]
No. (%) CED per patient-year 47 (19.1) 189 (76.8) 10 (4.1) 0 (0)
Radiation dose in patients with HHT / Canadian Association of Radiologists Journal xx (2013) 1e6
5
Figure 3. (A) Percentage contribution of each imaging modality to the number of procedures performed for the study cohort. (B) Percentage contribution of each imaging modality to the total study cohort cumulative effective dose. CT ¼ computed tomography; MRI ¼ magnetic resonance imaging.
Clinical factors that were associated with high cumulative exposure were epistaxis and a history of a previous GI bleed. Patients with clinically severe symptoms, including epistaxis and GI bleeding, may be more likely to present to their physician and to follow-up on imaging recommendations, which results in increased imaging and cumulative dose when compared with patients with milder symptoms. Clinicians may order additional imaging investigations to rule out hemoptysis as a known complication of PAVMs in patients who present with epistaxis or a GI bleed, which may also contribute to increased radiation exposure in this subset. The majority of patients in the study cohort experienced epistaxis (84%). The minority of patients without epistaxis may be young, such that they have not yet developed this manifestation of HHT. Younger patients generally have been followed for a shorter amount of time and, as a result, have received less imaging. The number of patient-years was significantly associated with a higher CED. Although patient genotype was not a statistically significant determinant of high cumulative dose, genotype data were only available in a proportion of patients (67%), which limited analysis. A limitation of the study is that PAVM characteristics, including the number of PAVMs, number of feeding vessels, and feeding artery diameter were not assessed. It is possible that more complex lesions and patients with more than 1 PAVM might receive more radiation. Of the study cohort, 11% (26 patients) received estimated CEDs of >100 mSv, which is the level at which there is considered to be good evidence for a risk of significant harm [5]. The Biological Effects of Ionizing Radiation VII report Table 3 Association between patient factors and total cumulative effective dose higher than 50 mSv Patient factors
No. (%)
OR (95% CI)
P value
Cerebral AVM positive Liver AVM positive Epistaxis Gastrointestinal bleed
16 31 205 53
0.5 1.9 2.7 2.0
.24 .12 .02a .04a
(7.7) (24.6) (84.4) (28.0)
(0.2-1.6) (0.8-4.4) (1.1-6.3) (1.0-3.8)
AVM ¼ arteriovenous malformation; CI ¼ confidence interval ;OR ¼ odds ratio. a P < .05.
estimated an excess lifetime cancer risk of 1 case in 1000 population for a standard population that receives a 10-mSv exposure [5]. If we consider a risk factor of 1 case per 1000 population per 10 mSv, then exposure to the mean CED of our study cohort (51.7 mSv) might result in an excess lifetime cancer risk of 0.5%. Individuals who receive >100 mSv (11% of our study cohort) might have an excess risk of 1%, and those who receive >200 mSv (2% of the study cohort) might have an excess risk of 2%. The results of our study are comparable with a large case series of patients who underwent imaging with CT at a tertiary referral center [8]. The investigators reported that 15% of patients received an estimated CED of more than 100 mSv when patients’ imaging histories were reviewed over a period of 22 years. In our cohort of patients, only 11% received estimated CED of more than 100 mSv from all imaging modalities, and only 3 patients (1.2%) received an estimated CED of more than 100 mSv from CTs alone. Sodickson et al [8] reported that 33% of patients underwent 5 or more CT examinations. Similarly, we report that 70 patients (28.5%) in our cohort underwent 5 or more CTs. Patients with HHT may receive radiation doses comparable with other patient populations with chronic conditions. Donadieu et al [12] reviewed the CT histories of adult patients with cystic fibrosis. They reported an average lifetime ED of 19.5 mSv (range, 2.2-75.8 mSv) from CTs alone [12]. The mean CED in patients with Crohn disease, patients with chronic renal failure on hemodialysis, and adult cardiology patients, has been reported at 36.1 mSv, 57.7 mSv, and 60.6 mSv, respectively [13e15]. Dose reduction strategies in patients with HHT may include Doppler ultrasound or magnetic resonance imaging rather than abdominal CT, magnetic resonance imaging to screen for and assess PAVM patency, and low-dose techniques for thoracic CTs [16,17]. At our institution, the thoracic CT protocol for patients with HHT was changed to a low-dose protocol in 2009. In the study cohort, the mean ED for standard dose CTs was 4.3 mSv compared with 1.9 mSv for the low-dose protocol, which is a 2.4 mSv difference per scan. Given that the average number of CT studies per patient is 4.0, this change alone will result in
6
K. Hanneman et al. / Canadian Association of Radiologists Journal xx (2013) 1e6
a reduction in cumulative dose by 9.5 mSv per patient. Multiple studies have confirmed that, even with a 50% reduction in radiation dose, chest CT image quality is acceptable for diagnostic evaluation, with no significant difference in detection of abnormalities [18e22]. There are several limitations regarding our study, including the fact that it was retrospective and nonrandomized. Also, the study was performed at a single tertiary center in North America, and, therefore, results may not necessarily be generalizable to other populations with HHT. The study was for a finite period (22 years) and, therefore, is not a true estimate of total lifetime exposure. Patients may have undergone imaging studies at other institutions, and, if these were not imported into the system at our hospital, then they would not have been taken into account. DLP was not available for all CTs; therefore, doses were estimated for a significant number of studies. Similarly, actual doses from interventional procedures were not available, and, therefore, estimates were used. Conclusion Patients with HHT who have PAVMs undergo repeated imaging procedures that result in increased exposure to ionizing radiation. A significant proportion of patients received estimated radiation doses that may put them at increased risk of cancer. Strategies to reduce dose and to optimize imaging indications and frequency should be considered in all patients with HHT. References [1] Faughnan ME, Granton JT, Young LH. The pulmonary vascular complications of hereditary haemorrhagic telangiectasia. Eur Respir J 2009;33:1186e94. [2] Faughnan ME, Palda VA, Garcia-Tsao G, et al. International guidelines for the diagnosis and management of hereditary haemorrhagic telangiectasia. J Med Genet 2011;48:73e87. [3] Remy-Jardin M, Dumont P, Brillet P, et al. Pulmonary arteriovenous malformations treated with embolotherapy: helical CT evaluation of long-term effectiveness after 2-21-year follow-up. Radiology 2006; 239:576e85. [4] Mager J, Overtoom TTC, Blauw H, et al. Embolotherapy of pulmonary arteriovenous malformations: long-term results in 112 patients. J Vasc Interv Radiol 2004;15:451e6. [5] National Research Council of the National Academies, Committee to Assess Health Risks From Exposure to Low Levels of Ionizing Radiation. Health Risks From Exposure to Low Levels of Ionizing Radiation: BEIR VII Phase 2. Washington, DC: The National Academies Press, 2006.
[6] Pierce DA, Preston DL. Radiation-related cancer risks at low doses among atomic bomb survivors. Radiat Res 2000;154:178e86. [7] Griffey RT, Sodickson A. Cumulative radiation exposure and cancer risk estimates in emergency department patients undergoing repeat or multiple CT. AJR Am J Roentgenol 2009;192:887e92. [8] Sodickson A, Baeyens PF, Andriole KP, et al. Recurrent CT, cumulative radiation exposure, and associated radiation-induced cancer risks from CT of adults. Radiology 2009;251:175e84. [9] Mettler FA, Huda W, Yoshizumi TT, et al. Effective doses in radiology and diagnostic nuclear medicine: a catalog. Radiology 2008;248: 254e63. [10] McCollough CH, Christner JA, Kofler JM. How effective is effective dose as a predictor of radiation risk? AJR Am J Roentgenol 2010;194: 890e6. [11] Storm ES, Miller DL, Hoover LJ, et al. Radiation doses from venous access procedures. Radiology 2006;238:1044e50. [12] Donadieu J, Roudier C, Saguintaah M, et al. Estimation of the radiation dose from thoracic CT scans in a cystic fibrosis population. Chest 2007; 132:1233e8. [13] Desmond AN, O’Regan K, Curran C, et al. Crohn’s disease: factors associated with exposure to high levels of diagnostic radiation. Gut 2008;57:1524e9. [14] Bedetti G, Botto N, Andreassi MG, et al. Cumulative patient effective dose in cardiology. Br J Radiol 2008;969:699e705. [15] De Mauri A, Brambilla M, Chiarinotti D, et al. Estimated radiation exposure from medical imaging in hemodialysis patients. J Am Soc Nephrol 2011;22:571e8. [16] Schneider G, Uder M, Koehler M, et al. MR angiography for detection of pulmonary arteriovenous malformations in patients with hereditary hemorrhagic telangiectasia. AJR Am J Roentgenol 2008; 190:892e901. [17] Boussel L, Cernicanu A, Geerts L, et al. 4D time-resolved magnetic resonance angiography for noninvasive assessment of pulmonary arteriovenous malformations patency. J Magn Reson Imaging 2010;5: 1110e6. [18] Prasad SR, Wittram C, Shepard JA, et al. Standard-dose and 50%-reduced-dose chest CT: comparing the effect on image quality. AJR Am J Roentgenol 2002;179:461e5. [19] Rusinek H, Naidich DP, McGuinness G, et al. Pulmonary nodule detection: low-dose versus conventional CT. Radiology 1998;209:243e9. [20] Mayo JR, Hartman TE, Lee KS, et al. CT of the chest: minimal tube current required for good image quality with the least radiation dose. AJR Am J Roentgenol 1995;164:603e7. [21] Takahashi M, Maguire WM, Ashtari M, et al. Low-dose spiral computed tomography of the thorax: comparison with the standarddose technique. Invest Radiol 1998;33:68e73. [22] Kubo T, Lin PJ, Stiller W, et al. Radiation dose reduction in chest CT: a review. AJR Am J Roentgenol 2008;190:335e43. [23] Mettler F, Bhargavan M, Faulkner K, et al. Radiologic and nuclear medicine studies in the United States and worldwide: frequency, radiation dose, and comparison with other radiation sourcesd19502007. Radiology 2009;3:520e31. [24] Brenner D, Doll R, Goodhead D, et al. Cancer risks attributable to low doses of ionizing radiation: assessing what we really know. Proc Natl Acad Sci U S A 2003;100:13761e6. [25] The 2007 Recommendations of the International Commission on Radiological Protection. ICRP Publication 103. Ann ICRP 2007;37: 1e332.
Thoracic and Cardiac Imaging / Imagerie cardiaque et imagerie thoracique
Cumulative Radiation Dose in Patients With Hereditary Hemorrhagic Telangiectasia and Pulmonary Arteriovenous Malformations Kate Hanneman, MDa,*, Marie E. Faughnan, MD, MScb,c,d, Vikramaditya Prabhudesai, MBBS, MS, FRCRa a
Department of Medical Imaging, St Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada b Division of Respirology, Department of Medicine, University of Toronto, Toronto, Ontario, Canada c Toronto Hereditary Hemorrhagic Telangiectasia Program, Division of Respirology, Department of Medicine, St Michael’s Hospital, Toronto, Ontario, Canada d Keenan Research Centre and the Li Ka Shing Knowledge Institute, St Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada
Abstract Purpose: To determine the cumulative effective dose (CED) of radiation from medical imaging and intervention in patients with hereditary hemorrhagic telangiectasia (HHT) who have pulmonary arteriovenous malformations and to identify clinical factors associated with exposure to high levels of radiation. Methods: All patients with at least 1 pulmonary arteriovenous malformation were identified from the dedicated patient database of a tertiary HHT referral center. Computerized imaging and electronic patient records were systematically examined to identify all imaging studies performed from 1989-2010. The effective dose was determined for each study, and CED was calculated retrospectively. Results: Among 246 patients (mean age, 53 years; 62.2% women) with a total of 2065 patient-years, 3309 procedures that involved ionizing radiation were performed. CED ranged from 0.2-307.6 mSv, with a mean of 51.7 mSv. CED exceeded 100 mSv in 26 patients (11%). Interventional procedures and computed tomography (CT) were the greatest contributors, which accounted for 51% and 46% of the total CED, respectively. Factors associated with high cumulative exposure were epistaxis (odds ratio 2.7 [95% confidence interval, 1.1-6.3]; P ¼ .02), HHT-related gastrointestinal bleeding (odds ratio 2.0 [95% confidence interval, 1.0-3.8]; P ¼ .04) and number of patient-years (P < .0001). Conclusions: Patients with HHT are exposed to a significant cumulative radiation dose from diagnostic and therapeutic interventions. Identifiable subsets of patients are at increased risk. A proportion of patients receive doses at levels that are associated with harm. Imaging indications and doses should be optimized to reduce radiation exposure in this population. Resume Objet: Determiner la dose efficace cumulative de radiation emanant des examens d’imagerie medicale et des interventions chez les patients atteints de telangiectasie hemorragique hereditaire et presentant des malformations pulmonaires arterio-veineuses, et definir les facteurs cliniques associes a une forte radioexposition. Methodes: Tous les patients presentant au moins une malformation pulmonaire arterio-veineuse qui figuraient dans la base de donnees sur les patients d’un centre de soins tertiaires specialise en telangiectasie hemorragique hereditaire ont ete identifies. Une revue systematique des images informatisees et des dossiers electroniques de ces patients a ensuite permis de relever tous les examens d’imagerie ayant ete realises entre 1989 et 2010. Enfin, la dose efficace a ete determinee pour chaque examen, puis a mene au calcul retrospectif de la dose efficace cumulative. Resultats: 3 309 interventions comportant de la radiation ionisante ont ete realisees chez 246 patients (^age moyen de 53 ans; 62,2 % de sexe feminin) representant un total de 2 065 annees-patients. Les calculs ont revele une dose efficace cumulative variant de 0,2 a 307,6 mSv et une moyenne de 51,7 mSv. Une dose efficace cumulative superieure a 100 mSv a ete constatee chez 26 patients (11 %). Par ailleurs, les interventions et les examens de tomodensitometrie (TDM) ont eu une incidence decisive sur les resultats, representant respectivement 51 % et 46 % de la dose efficace cumulative. Enfin, les facteurs suivants ont ete associes a une radioexposition cumulative elevee : l’epistaxis (rapport
* Address for correspondence: Kate Hanneman, MD, Department of Medical Imaging, St Michael’s Hospital, 30 Bond Street, Toronto, Ontario M5B 1W8, Canada.
E-mail address: [email protected] (K. Hanneman).
0846-5371/$ - see front matter Ó 2013 Canadian Association of Radiologists. All rights reserved. http://dx.doi.org/10.1016/j.carj.2013.02.007
2
K. Hanneman et al. / Canadian Association of Radiologists Journal xx (2013) 1e6
des cotes de 2,7 [intervalle de confiance de 95 %, de 1,1 a 6,3]; P ¼ 0,02), les saignements gastro-intestinaux lies a la telangiectasie hemorragique hereditaire (rapport des cotes de 2,0 [intervalle de confiance de 95 %, de 1,0 a 3,8]; P ¼ 0,04) et le nombre d’annees-patients (P < 0,0001). Conclusions: Les patients atteints de telangiectasie hemorragique hereditaire sont exposes a une dose cumulative significative emanant des interventions a visee diagnostique et therapeutique. Certains sous-groupes de patients presentent un risque accru. Une certaine proportion de patients a rec¸u des doses dont le niveau est associe a des effets nocifs. Les indications et les doses en matiere d’imagerie doivent ^etre optimisees afin de reduire la radioexposition que subit ce type de patients. Ó 2013 Canadian Association of Radiologists. All rights reserved. Key Words: Telangiectasia; Hereditary hemorrhagic; Arteriovenous malformations; Radiation; Ionizing; Radiation dosage; Multidetector computed tomography
Hereditary hemorrhagic telangiectasia (HHT), historically known as Rendu-Osler-Weber syndrome, is a rare hereditary disorder that may be complicated by arteriovenous malformations (AVM) in the brain, lung, gastrointestinal (GI) tract, and liver. Pulmonary AVMs (PAVM) are frequently present in HHT, occurring in 15%-45% of patients (Figure 1) [1]. Adults with untreated PAVMs are at risk of life-threatening hemorrhage and neurologic complications [1]. Routine screening for PAVMs in all patients with HHT is recommended, given the serious nature of complications that may occur and the availability of preventative treatment with transcatheter embolotherapy [2]. Given the potential for multiorgan involvement, as well as the role of interventional radiology, standard care for patients with HHT results in repeated imaging procedures, and patients may be exposed to a cumulative dose of radiation much higher than with the average patient. Unenhanced thoracic computed tomography (CT) is recommended to confirm the presence of PAVMs after an initial positive screening with transthoracic contrast echocardiography. Small PAVMs may be followed with repeated serial CTs because growth is demonstrated in up to 18% of small PAVMs, and these may eventually warrant treatment [2e4]. Significant PAVMs are treated with transcatheter embolotherapy, then follow-up CT is recommended at
12 months after treatment, and then every 1-3 years (depending on the size of residual PAVMs). Long-term follow-up is recommended to detect growth of previously small, insignificant PAVMs but also to detect reperfusion of embolized PAVMs [2]. However, high radiation exposure is an established cause of cancer, and there is evidence from observational studies of excess risks from fractionated exposures in the dose range that would be received from repeated CTs [5e8]. The purpose of this study was to quantify the cumulative effective dose (CED) associated with diagnostic, interventional, and follow-up imaging in a cohort of patients with HHT and with at least 1 PAVM, and to identify factors associated with exposure to high levels of diagnostic radiation. The lifetime radiation dose received by patients with HHT has not been studied previously. Materials and Methods Overview To characterize radiation exposure in this population, all patients with at least 1 PAVM were identified from the dedicated patient database of a tertiary HHT referral center. Demographic and clinical data were obtained from the
Figure 1. A 30-year-old woman with hereditary hemorrhagic telangiectasia. (A) Frontal chest radiograph and (B) coronal computed tomography image, demonstrating a right upper lobe pulmonary arteriovenous malformations (arrows).
Radiation dose in patients with HHT / Canadian Association of Radiologists Journal xx (2013) 1e6
database. The number of patient-years was calculated from the time of the first recorded imaging procedure to the study cutoff date. Patient genotype was recorded for patients for whom these data were available. High cumulative exposure to radiation was defined as CED that exceeded 50 mSv, which is an effective dose (ED) equivalent to approximately 500 standard chest radiographs [9]. This study was approved by the institution’s research ethics board. Subjects A total of 251 patients were identified as potentially eligible. Patients who had no recorded diagnostic or therapeutic imaging procedures were excluded (n ¼ 4). A patient who, on review of his clinical data, did not have a PAVM was also excluded (n ¼ 1). A total of 246 patients were included, 93 men (37.8%) and 153 women (62.2%). The mean age was 53 years (range, 17-92 years). The mean number of patientyears was 8.4 years (range, 1-22 patient-years). Methods Computerized imaging records (electronic radiology information system) were systematically examined in April 2011 to identify all imaging studies performed from January 1, 1989, to December 31, 2010, including fluoroscopyguided interventional procedures. Studies performed for non-HHTerelated indications and studies performed at other institutions that had been imported into the picture archiving and communication system were not excluded. The date of each study and the anatomic region examined were recorded. CTs were read to determine the following technical parameters when available: displayed kilovolt value, anodic current (in milliamperes), slice thickness, and dose length product (DLP). For CT studies in which DLP was available (338/982 CTs), ED was calculated by using published conversion coefficients [10]. For thoracic CTs in which DLP was the only imaging parameter not available, studies performed by using the same protocol (including kilovolt peak, anodic currant, and slice thickness) were assigned the calculated average ED of all other studies performed when using that same protocol for which DLP was available (66/982 CT studies). To estimate radiation exposure for CTs for which technical parameters were not available, and for all other imaging modalities that involved ionizing radiation, including interventional and fluoroscopic procedures and nuclear medicine studies, we obtained
3
estimates of EDs from the published literature [9]. Mean ED estimates for interventional procedures ranged from 1-15 mSv [9,11]. Statistical Analysis Data compilation and statistical analyses were performed by using Microsoft Excel 2007 (Microsoft Corporation, Redmond, WA), and Stata version 10.0 (StataCorp, College Station, TX). All continuous data are expressed as mean (standard deviation [SD]) unless noted otherwise. In addition to standard descriptive statistics, a generalized linear model was used to identify determinants of CED. To normalize the distribution of the residuals, CED was first log-transformed. Analyses of the difference between groups of nonparametric data, including genotype, were performed by using the KruskalleWallace test. Univariate analysis of the associations among patient clinical factors (including sex, the presence of cerebral and liver AVMs, and a history of epistaxis and GI bleeding), and high exposure (defined as CED higher than 50 mSv) were performed by using the Mantel-Haenszel test (expressed as odds ratios [OR]). Results Number of Studies Among 246 patients with a total of 2065 patient-years, 3309 procedures that involved ionizing radiation were performed, with a mean of 13.5 (range, 1-78) per patient. The distribution of the number of studies and CED according to imaging modality is presented in Table 1. Cumulative Dose The mean CED was 51.7 mSv, with an asymmetrical distribution (range, 0.2-307.6 mSv) (Figure 2). CED exceeded 50 mSv and 100 mSv in 84 (34%) and 26 (11%) patients, respectively. The mean CED per patient-year was 7.3 mSv (range, 0.03- 44.6 mSv). The ED per patient-year according to dose categories is presented in Table 2, with reference values for comparison. Imaging Interventional procedures and CT were the greatest contributors to CED, which accounted for 51% and 46%
Table 1 The number of procedures performed, total number of procedures and patients, and CED according to imaging modality Imaging modality
No. procedures, mean (range) Total no. procedures Total no. patients who received imaging Mean (range) CED, mSv CED ¼ cumulative effective dose.
Radiography
Computed tomography
Fluoroscopy and/or interventional
Nuclear medicine
All modalities
7.0 (0-56) 1734 243 0.9 (0-12.1)
4.0 (0-20) 982 243 23.8 (0-232.0)
2.2 (0-16) 552 213 26.2 (0-210)
0.2 (0-6) 43 27 0.8 (0-33.9)
13.5 (1-78) 3311 246 51.7 (0.2-307.6)
4
K. Hanneman et al. / Canadian Association of Radiologists Journal xx (2013) 1e6
Figure 2. Distribution of individual cumulative effective dose (CED) estimates for total study cohort (n ¼ 246). CED exceeded 50 mSv and 100 mSv in 84 (34%) and 26 (11%) patients, respectively.
of total CED, respectively (Figure 3). Plain radiographs represented 43% of the studies but accounted for only 2% of CEDs. In total, 982 CTs were performed with 243 patients, of which the majority were thoracic CTs (825 [84.0%]). The mean number of CTs per patient was 4.0 (range, 0-20). Only 3 patients had not undergone a CT. The mean (SD) CED from CTs was 23.8 23.3 mSv per patient. The mean (SD) ED per CT was 6.0 4.8 mSv. For studies in which DLP was available, the mean (SD) ED for standard dose thoracic CTs was 4.3 0.4 mSv (n ¼ 118) compared with 1.9 0.4 mSv for thoracic CTs performed with a low-dose protocol (n ¼ 135).
Clinical Factors Factors associated with high cumulative exposure were epistaxis (OR 2.7 [95% confidence interval, 1.1-6.3]; P ¼ .02]) and a history of a previous GI bleed (OR 2.0 [95% confidence interval, 1.0-3.8]; P ¼ .04) (Table 3). The number of patient-years is significantly related to CED (P < .0001). The CED per patient-year was not significantly associated with genotype (P ¼ .55).The mean CED per patient-year was 7.1 mSv (n ¼ 132), 6.9 mSv (n ¼ 27), and 15.0 mSv (n ¼ 7) for genotypes ENG (mutation of the endoglin gene), ALK1 (mutation of the activin receptor-like kinase 1 gene), and SMAD4, respectively. The genotype was not available in 33% of patients (n ¼ 80).
Discussion There is increasing acceptance of the linear, no-threshold hypothesis of radiation exposure, which raises concerns about the potential long-term effects of repeated exposure to radiation due to medical imaging [5]. The ED is defined by the International Commission on Radiological Protection as the sum of the absorbed doses in all tissues and organs of the body, each weighted according to its radiation sensitivity. ED allows for population-level comparisons across different types of radiation exposures. Interventional procedures were the highest contributor to total CED in the study cohort. Current guidelines recommend preventative treatment, with coil embolization therapy, of PAVMs with feeding artery diameters larger than 3 mm [2]. The average dose associated with pulmonary angiography and embolization is estimated at 5-15 mSv [9]. The mean number of embolization procedures per patient was 2.2 (range, 0-16). Patients may have multiple PAVMs that require more than 1 treatment session. Patients may also require repeated embolization because reperfusion occurs in up to 8%-19% of successfully occluded lesions [2e4]. The modality of CT was the second greatest contributor to total CED. Thoracic CT is indicated to confirm the presence of a PAVM, to evaluate for growth of documented lesions, and to assess for reperfusion of previously coiled lesions [2]. The mean number of CTs per patient was 4.0 (range, 0-20).
Table 2 CED per patient-year according to dose categories Dose category Low Moderate High Very high
Dose range (mSv/y) 3 > 3-20 > 20-50 > 50
CED ¼ cumulative effective dose.
Comparison Background level of radiation from natural sources[23]; estimated annual per-capita effective dose from medical procedures in the United States[24] Upper annual limit for occupational exposure for at-risk workers, averaged over 5 y[25] Upper annual limit for occupational exposure for at-risk workers in any given year[25] Reasonable evidence for an increase in cancer risk at protracted doses higher than 50 mSv[24]
No. (%) CED per patient-year 47 (19.1) 189 (76.8) 10 (4.1) 0 (0)
Radiation dose in patients with HHT / Canadian Association of Radiologists Journal xx (2013) 1e6
5
Figure 3. (A) Percentage contribution of each imaging modality to the number of procedures performed for the study cohort. (B) Percentage contribution of each imaging modality to the total study cohort cumulative effective dose. CT ¼ computed tomography; MRI ¼ magnetic resonance imaging.
Clinical factors that were associated with high cumulative exposure were epistaxis and a history of a previous GI bleed. Patients with clinically severe symptoms, including epistaxis and GI bleeding, may be more likely to present to their physician and to follow-up on imaging recommendations, which results in increased imaging and cumulative dose when compared with patients with milder symptoms. Clinicians may order additional imaging investigations to rule out hemoptysis as a known complication of PAVMs in patients who present with epistaxis or a GI bleed, which may also contribute to increased radiation exposure in this subset. The majority of patients in the study cohort experienced epistaxis (84%). The minority of patients without epistaxis may be young, such that they have not yet developed this manifestation of HHT. Younger patients generally have been followed for a shorter amount of time and, as a result, have received less imaging. The number of patient-years was significantly associated with a higher CED. Although patient genotype was not a statistically significant determinant of high cumulative dose, genotype data were only available in a proportion of patients (67%), which limited analysis. A limitation of the study is that PAVM characteristics, including the number of PAVMs, number of feeding vessels, and feeding artery diameter were not assessed. It is possible that more complex lesions and patients with more than 1 PAVM might receive more radiation. Of the study cohort, 11% (26 patients) received estimated CEDs of >100 mSv, which is the level at which there is considered to be good evidence for a risk of significant harm [5]. The Biological Effects of Ionizing Radiation VII report Table 3 Association between patient factors and total cumulative effective dose higher than 50 mSv Patient factors
No. (%)
OR (95% CI)
P value
Cerebral AVM positive Liver AVM positive Epistaxis Gastrointestinal bleed
16 31 205 53
0.5 1.9 2.7 2.0
.24 .12 .02a .04a
(7.7) (24.6) (84.4) (28.0)
(0.2-1.6) (0.8-4.4) (1.1-6.3) (1.0-3.8)
AVM ¼ arteriovenous malformation; CI ¼ confidence interval ;OR ¼ odds ratio. a P < .05.
estimated an excess lifetime cancer risk of 1 case in 1000 population for a standard population that receives a 10-mSv exposure [5]. If we consider a risk factor of 1 case per 1000 population per 10 mSv, then exposure to the mean CED of our study cohort (51.7 mSv) might result in an excess lifetime cancer risk of 0.5%. Individuals who receive >100 mSv (11% of our study cohort) might have an excess risk of 1%, and those who receive >200 mSv (2% of the study cohort) might have an excess risk of 2%. The results of our study are comparable with a large case series of patients who underwent imaging with CT at a tertiary referral center [8]. The investigators reported that 15% of patients received an estimated CED of more than 100 mSv when patients’ imaging histories were reviewed over a period of 22 years. In our cohort of patients, only 11% received estimated CED of more than 100 mSv from all imaging modalities, and only 3 patients (1.2%) received an estimated CED of more than 100 mSv from CTs alone. Sodickson et al [8] reported that 33% of patients underwent 5 or more CT examinations. Similarly, we report that 70 patients (28.5%) in our cohort underwent 5 or more CTs. Patients with HHT may receive radiation doses comparable with other patient populations with chronic conditions. Donadieu et al [12] reviewed the CT histories of adult patients with cystic fibrosis. They reported an average lifetime ED of 19.5 mSv (range, 2.2-75.8 mSv) from CTs alone [12]. The mean CED in patients with Crohn disease, patients with chronic renal failure on hemodialysis, and adult cardiology patients, has been reported at 36.1 mSv, 57.7 mSv, and 60.6 mSv, respectively [13e15]. Dose reduction strategies in patients with HHT may include Doppler ultrasound or magnetic resonance imaging rather than abdominal CT, magnetic resonance imaging to screen for and assess PAVM patency, and low-dose techniques for thoracic CTs [16,17]. At our institution, the thoracic CT protocol for patients with HHT was changed to a low-dose protocol in 2009. In the study cohort, the mean ED for standard dose CTs was 4.3 mSv compared with 1.9 mSv for the low-dose protocol, which is a 2.4 mSv difference per scan. Given that the average number of CT studies per patient is 4.0, this change alone will result in
6
K. Hanneman et al. / Canadian Association of Radiologists Journal xx (2013) 1e6
a reduction in cumulative dose by 9.5 mSv per patient. Multiple studies have confirmed that, even with a 50% reduction in radiation dose, chest CT image quality is acceptable for diagnostic evaluation, with no significant difference in detection of abnormalities [18e22]. There are several limitations regarding our study, including the fact that it was retrospective and nonrandomized. Also, the study was performed at a single tertiary center in North America, and, therefore, results may not necessarily be generalizable to other populations with HHT. The study was for a finite period (22 years) and, therefore, is not a true estimate of total lifetime exposure. Patients may have undergone imaging studies at other institutions, and, if these were not imported into the system at our hospital, then they would not have been taken into account. DLP was not available for all CTs; therefore, doses were estimated for a significant number of studies. Similarly, actual doses from interventional procedures were not available, and, therefore, estimates were used. Conclusion Patients with HHT who have PAVMs undergo repeated imaging procedures that result in increased exposure to ionizing radiation. A significant proportion of patients received estimated radiation doses that may put them at increased risk of cancer. Strategies to reduce dose and to optimize imaging indications and frequency should be considered in all patients with HHT. References [1] Faughnan ME, Granton JT, Young LH. The pulmonary vascular complications of hereditary haemorrhagic telangiectasia. Eur Respir J 2009;33:1186e94. [2] Faughnan ME, Palda VA, Garcia-Tsao G, et al. International guidelines for the diagnosis and management of hereditary haemorrhagic telangiectasia. J Med Genet 2011;48:73e87. [3] Remy-Jardin M, Dumont P, Brillet P, et al. Pulmonary arteriovenous malformations treated with embolotherapy: helical CT evaluation of long-term effectiveness after 2-21-year follow-up. Radiology 2006; 239:576e85. [4] Mager J, Overtoom TTC, Blauw H, et al. Embolotherapy of pulmonary arteriovenous malformations: long-term results in 112 patients. J Vasc Interv Radiol 2004;15:451e6. [5] National Research Council of the National Academies, Committee to Assess Health Risks From Exposure to Low Levels of Ionizing Radiation. Health Risks From Exposure to Low Levels of Ionizing Radiation: BEIR VII Phase 2. Washington, DC: The National Academies Press, 2006.
[6] Pierce DA, Preston DL. Radiation-related cancer risks at low doses among atomic bomb survivors. Radiat Res 2000;154:178e86. [7] Griffey RT, Sodickson A. Cumulative radiation exposure and cancer risk estimates in emergency department patients undergoing repeat or multiple CT. AJR Am J Roentgenol 2009;192:887e92. [8] Sodickson A, Baeyens PF, Andriole KP, et al. Recurrent CT, cumulative radiation exposure, and associated radiation-induced cancer risks from CT of adults. Radiology 2009;251:175e84. [9] Mettler FA, Huda W, Yoshizumi TT, et al. Effective doses in radiology and diagnostic nuclear medicine: a catalog. Radiology 2008;248: 254e63. [10] McCollough CH, Christner JA, Kofler JM. How effective is effective dose as a predictor of radiation risk? AJR Am J Roentgenol 2010;194: 890e6. [11] Storm ES, Miller DL, Hoover LJ, et al. Radiation doses from venous access procedures. Radiology 2006;238:1044e50. [12] Donadieu J, Roudier C, Saguintaah M, et al. Estimation of the radiation dose from thoracic CT scans in a cystic fibrosis population. Chest 2007; 132:1233e8. [13] Desmond AN, O’Regan K, Curran C, et al. Crohn’s disease: factors associated with exposure to high levels of diagnostic radiation. Gut 2008;57:1524e9. [14] Bedetti G, Botto N, Andreassi MG, et al. Cumulative patient effective dose in cardiology. Br J Radiol 2008;969:699e705. [15] De Mauri A, Brambilla M, Chiarinotti D, et al. Estimated radiation exposure from medical imaging in hemodialysis patients. J Am Soc Nephrol 2011;22:571e8. [16] Schneider G, Uder M, Koehler M, et al. MR angiography for detection of pulmonary arteriovenous malformations in patients with hereditary hemorrhagic telangiectasia. AJR Am J Roentgenol 2008; 190:892e901. [17] Boussel L, Cernicanu A, Geerts L, et al. 4D time-resolved magnetic resonance angiography for noninvasive assessment of pulmonary arteriovenous malformations patency. J Magn Reson Imaging 2010;5: 1110e6. [18] Prasad SR, Wittram C, Shepard JA, et al. Standard-dose and 50%-reduced-dose chest CT: comparing the effect on image quality. AJR Am J Roentgenol 2002;179:461e5. [19] Rusinek H, Naidich DP, McGuinness G, et al. Pulmonary nodule detection: low-dose versus conventional CT. Radiology 1998;209:243e9. [20] Mayo JR, Hartman TE, Lee KS, et al. CT of the chest: minimal tube current required for good image quality with the least radiation dose. AJR Am J Roentgenol 1995;164:603e7. [21] Takahashi M, Maguire WM, Ashtari M, et al. Low-dose spiral computed tomography of the thorax: comparison with the standarddose technique. Invest Radiol 1998;33:68e73. [22] Kubo T, Lin PJ, Stiller W, et al. Radiation dose reduction in chest CT: a review. AJR Am J Roentgenol 2008;190:335e43. [23] Mettler F, Bhargavan M, Faulkner K, et al. Radiologic and nuclear medicine studies in the United States and worldwide: frequency, radiation dose, and comparison with other radiation sourcesd19502007. Radiology 2009;3:520e31. [24] Brenner D, Doll R, Goodhead D, et al. Cancer risks attributable to low doses of ionizing radiation: assessing what we really know. Proc Natl Acad Sci U S A 2003;100:13761e6. [25] The 2007 Recommendations of the International Commission on Radiological Protection. ICRP Publication 103. Ann ICRP 2007;37: 1e332.