Radiation Dose Optimization in Cardiac CT: A Technical Review CHENG Wai Kwong 17 May 2014 Contents 1. Introduction and background 2. Cardiac CT synchronization techniques 3. 4. Conclusion 1
Introduction and background Introduction & background Cardiac CT (mainly CCTA): promising, low-invasive Radiation dose: major concern for CCTA, esp. for young female patients. increase from 16- to 64-slice CT ECG-gated helical acquisitions: dose up 40 msv (Jean- François & Hicham, 2007) Differences in dose due to performance of CT scanners, technical dose-sparing tools. Jean-François, P & Hicham, A. (2007). Strategies for reduction of radiation dose in cardiac multislice CT, European Radiology, 17(8), pp. 2028-2035. 2
Cardiac CT synchronization techniques Cardiac CT synchronization techniques Prospective ECG-triggering Synchronization occurs when HR and sequential CT scanning in harmony x-ray tube in same position for various heart beats Trigger x-ray beam and acquire data in certain cardiac phase preferably diastolic phase (70 80 % of R-R interval) with min. cardiac motion e.g. snapshot of 100 ms in 70 % R-R interval Slow and stable HR, low radiation dose (Courtesy to S. Edyvean) 3
Cardiac CT synchronization techniques Retrospective ECG-gating ECG signals and helical scan data acquired simultaneously. Scan data selected for image reconstruction w.r.t. pre-defined cardiac phase with a certain temporal relation to onset of R-waves. Better temporal resolution with segmented reconstruction Redundancy of data acquisition phase-selective image reconstruction High radiation dose (Courtesy to S. Edyvean) Cardiac CT synchronization techniques Prospectively ECG-triggered high-pitch helical acquisition State-of-art dual-source scanner - a pitch of 3.4 to fill in the gaps of raw data acquired by second X-ray tube Acquire data very quickly in ~ an 1-mile-per hour table translation speed Not prospective ECG-triggering, not retrospective ECG-gating Helical cardiac data acquired by 1 st tube Helical cardiac data acquired by 2 nd tube (Courtesy to Suhny Abbara) 4
Strategy 1: ECG-gated ma modulation Retrospective ECG-gating ma adjusted based on predicted cardiac phases to optimal setting (optimal max. and min. ma) acceptable image quality e.g. 350 ma during mid-diastole (60 80% of R-R interval) and 200 ma for the remainder 30 35% radiation dose (Halliburton, Sola & Kuzmiak 2008; Abbara, Arbab-Zadeh & Callister et al. 2009) 350 ma 200 ma (Courtesy to J. Leipsic & B. Heibron) Halliburton, SS, Sola S & Kuzmiak S (2008). Effect of dual-source cardiac computed tomography on patient radiation dose in a clinical setting: Comparison to single-source imaging. Journal of Cardiovascular Computed Tomography, vol. 2, pp. 392-400. Abbara, CS, Arbab-Zadeh, A & Callister, TQ et al. (2009). SCCT Guidelines for the performance and acquisition of coronary computed tomographic angiography. Journal Cardiovascular Computed Tomography; doi:10.1016/j.jcct. 2009.03.004. 5
Strategy 2: Optimization of kvp Radiation dose with CT ~ α ( kvp) 2, 120 to 100 kv 31% reduction in dose (Bischoff, Hein & Meyer et al. 2009). Intravascular contrast with tube voltage (100 kvp or 80 kvp) because iodine resorption α 1/ kvp due to higher degree of P.E. effect and lower degree of Compton scattering (Gutstein, Dey & Cheng et al 2008). kvp attenuation of vessel lumen and cardiac chambers if iodinated C.M. used. 80 100 kvp protocols in myocardial CT perfusion, CCTA (< 70 kg) Bischoff, B, Hein, F, Meyer, T, Hadamitzky, M, Martinoff, S, Schomig, A, Hausleiter, J (2009). Impact of a reduced tube voltage on CT angiography and radiation dose: results of the PROTECTION I study. JACC Cardiovascular Imaging, Vol. 2, pp. 940 946. Gutstein, A, Dey, D, Cheng, V, Wolak, A, Gransar, H, Suzuki, Y, Friedman, J, Thomson, LE, Hayes, S, Pimentel, R, Paz, W, Slomka, P, Le Meunier, L, Germano, G & Berman, DS (2008). Algorithm for radiation dose reduction with helical dual source coronary computed tomography angiography in clinical practice. Journal of Cardiovascular Computed Tomography, vol. 2, pp. 311 322. Strategy 2: Optimization of kvp Pflederer, Rudofsky & Ropers et al (2009) 100 kvp for patients weighing <= 90 kg or with a BMI <= 30 kg/m 2 120 kvp for patients weighing > 90 kg or with a BMI > 30 kg/m 2 Higher kvp (e.g. 135 kvp) for severely obese patients (> 100 kg). Pflederer, T, Rudofsky, L, Ropers, D, Bachmann, S, Marwan, M, Daniel, WG & Achenbach, S (2009). Image quality in a low radiation exposure protocol for retrospectively ECG-gated coronary CT angiography. American Journal of Roentgenology, vol. 192, pp.1045 1050. 6
Strategy 2: Optimization of kvp Hoe (2013) < 100 kg - 100 kvp (Siemens DSCT FLASH) < 85 kg - 100 kvp (Siemens DSCT) < 25 kg/m 2 BMI 100 kvp (GE 64-slice MDCT) < 30 kg/m 2 BMI 100 kvp (GE 750 HD) < 80 kg 100 kvp, 400 ma > 80 kg 120 kvp, 800 ma (GE 64-slice MDCT) < 60 kg 80 kvp dose by 80 % What body weight or BMI for lower kvp CCTA scan? Remark: (i) < 70 kg 100 kvp, 350 ma, 0.35 s (ii) 120 kvp for post-stenting CCTA (Toshiba Aquilion 64) Hoe, J. (2013). Low dose coronary CTA: What you need to know about medical radiation and how to reduce it, presented in the 9 th CT Coronary Angiography Teaching Course Intermediate-Advanced Level, at HKEC Training Centre for Healthcare Management & Clinical Technology on 30 March 2013. Strategy 2: Optimization of kvp Automatic kvp selection AEC selecting tube potential according to diagnostic task and patient attenuation or size kvp constant for same patient, anatomical region and exam. (Courtesy to Katharine Grant & Bernhard Schmidt) 7
Strategy 3: Optimal scan length Absorbed dose α DLP = CTDIvol x scan length Optimal scan range: minimum scan length clinically necessary With reference to calcium score scan, CCTA scan range from level just above LM to heart base Margins of 10 mm sup. and inf. to most cranial and caudal slices allowed account for breath-hold inconsistencies scan manually stopped if desired anatomy covered Optimal scan length planning each breath-hold at same inspiration depth Strategy 3: Optimal scan length Retrospectively gated helical acquisitions: a scan length reduction of 1 cm approx. dose saving of 1 msv (Hausleiter, Meyer, Hermann et al, 2009) A scan length reduction of 1 cm dose by 10 % (Hoe, 2013) Hausleiter, J, Meyer, T, Hermann, F. et al. (2009). Estimated radiation dose associated with cardiac CT angiography. The Journal of the American Medical Association, 301(5), pp. 500-507. Hoe, J. (2013). Low dose coronary CTA: What you need to know about medical radiation and how to reduce it, presented in the 9 th CT Coronary Angiography Teaching Course Intermediate-Advanced Level, at HKEC Training Centre for Healthcare Management & Clinical Technology on 30 March 2013. 8
Strategy 4: Reconstruction of optimal slice thickness Noise α 1 / (reconstructed slice thickness) radiation dose (through a reduction in kvp or ma) with in slice thickness while maintaining same image noise. 3-mm thick image has 73% less noise than 1-mm thick image (Halliburton, Abbara, Chen, Gentry, Mahesh, Raff, Shaw & Hausleiter, 2011). Images reconstructed with greatest possible slice thickness for given cardiovascular CT indications (e.g. 0.5 mm slices for evaluating coronary artery stenosis, 3.0 mm for calcium scoring or evaluating larger blood vessels such as aorta and PA). Halliburton, SS, Abbara, S, Chen, MY, Gentry, R, Mahesh, M, Raff, GL, Shaw, LJ & Hausleiter, J (2011). SCCT guidelines on radiation dose and dose-optimization strategies in cardiovascular CT. Journal of Cardiovascular Computed Tomography, vol. 5, no. 4, pp. 198-224. Strategy 5: Helical scanning with higher pitch Latest generation of dual-source CT technology: ECG-triggered helical scanning at very high pitch values (High helical pitch flash acquisition) Interleaving data measured from 2 detector systems separated by ~ 90 degrees, pitch 3.4 Near complete elimination of overlap in cone beam amount of redundant data collected (Courtesy to J. Leipsic & B. Heibron) (Courtesy to D.W. Entrikin, J.A. Leipsic & J.J. Carr) (Courtesy to U.J. Schoepf) 9
Strategy 6: Filters and breast shields (Rajiah, Halliburton & Flamm, 2012) Filters placed beneath the x-ray tube selectively attenuate low-energy x-rays not contributing to image formation Flat and cardiac bow-tie filters used with small SFOV radiation exposure to lungs (Courtesy to S. Edyvean) Rajiah, P, Halliburton, SS & Flamm, SD (2012). Strategies for Dose Reduction in Cardiovascular Computed Tomography, Applied Radiology, vol. 41, no. 7-8, pp. 10-15. Strategy 6: Filters and breast shields (Rajiah, Halliburton & Flamm, 2012) Bismuth-impregnated latex shield placed between x-ray beam and breasts (i.e. in AP projection) specifically radiation dose by over 25 % (Courtesy to M. Gunn) degradation of image quality by increased pixel attenuation values Rajiah, P, Halliburton, SS & Flamm, SD (2012). Strategies for Dose Reduction in Cardiovascular Computed Tomography, Applied Radiology, vol. 41, no. 7-8, pp. 10-15. 10
Strategy 7: Organ-based dose modulation ma modulation avoiding direct X-ray in radiosensitive organs (AAPM Workgroup) (Courtesy to Siemens) Strategy 7: Organ-based dose modulation ma over radiosensitive organs on periphery dose to other organs (Dual et al, 2011) 11
Strategy 8: Adaptive collimation Over-ranging in MDCT generally with collimation unnoticed exposure to breasts Adaptive collimation elimination of radiation exposure unnecessary for image reconstruction Helical scanning (Courtesy to Toshiba Medical Systems) (Courtesy to P.D. Deak) Strategy 9: Iterative reconstruction Filtered back-projection (FBP) fast, images prone to high noise in low dose situations clinical situations in which image quality tends to be less acceptable include: (1) CT scans in large patients, (2) images with small voxels, such as the targeted FOV used for cardiac CT, (3) images through the bony pelvis. 12
Strategy 9: Iterative reconstruction Iterative reconstruction algorithm assumes initial μx,y,z for all voxels to predict projection data. predicted projection data compared with actual, measured projection data voxel attenuations modified and compared with measured data until the error between estimated and measured projection data acceptable. significant computational power and additional time improves noise properties (Leipsic, Labounty & Heilbron 2010) (Courtesy to Brian Hutton) (Leipsic, J, Labounty, TM, Min, JK, et al (2010). Adaptive statistical iterative reconstruction: assessment of image noise and image quality in coronary CT angiography. American Journal Roentgenology, 195, 649-654.) Strategy 9: Iterative reconstruction of images IR performed from raw (projection) data or image (slice) data or both (i) IRIS (Iterative Reconstruction in Image Space) (Siemens) IR from image data only (ii) Veo (Model-based Iterative Reconstruction)(GE) IR from raw data only (iii) ASIR (Adaptive Statistical Iterative Reconstruction) (GE) SAFIRE (Sinogram Affirmed Iterative Reconstruction) (Siemens) idose 4 (Philips) AIDR (Adaptive Iterative Dose Reduction) (Toshiba) IR from both raw data and image data (Courtesy to Toshiba Medical) 13
Strategy 10: Garnet gemstone-based detector Garnet gemstone-based scintillator (CT750 HD) configured to emit fluorescence when irradiated with x-rays decay time of 30 ns, 100 times faster than conventional scintillators (e.g. GOS, Bismuth Germinate (Bi4Ge3012) and Cadmium Tungstate (CdWO4)) afterglow level only 25 % of those of conventional scintillator materials spatial resolution: 18.2 lp / cm high-definition imaging at up to 230-μm resolution and delineation of stents enhances S/N improves low contrast resolution optimizes ma and radiation dose without significant loss of image quality (Yanagawa, Tomiyama & Honda et al, 2010) Yanagawa, M, Tomiyama, N & Honda, O et al. (2010). Multidetector CT of the Lung: Image Quality with Garnet-based Detectors. Radiology, 255(3), pp. 944-954. Strategy 11: Patient preparation Proper patient preparation (i) Careful explanation of procedure warm sensation to pelvis during CM injection breathing instructions (esp. for 16- or 64- MDCT) (ii) Administration of oral (50 mg + 50 mg) / IV (5 mg + 5 mg) beta-blockers (iii) TNG (1 puff SL just prior to CCTA scan) High HR, arrhythmia and breathing artefacts degrade image quality Proper HR control prospectively ECG-triggered axial sequential scanning if HR < 60 & stable. 14
Strategy 11: Patient preparation Breath hold practice Patients instructed to take comfortable breath in about 75 % of full inspiration Scan performed in end inspiration not Valsalva Place hand under costal margin to monitor if patient really holding breath Count 1 sec, 2 sec, 3 sec, 4 sec, etc. for each second Strategy 12: Monitoring and recording dose information Two primary purposes: 1. documenting individual dose burden 2. quality control Recording dose descriptors in: 1. A DICOM image with radiation information in a PACS 2. A paper-based logbook 3. epr or RIS 4. A dedicated database or local registry periodic review of radiation doses local data compared with international standards and other references (e.g. Diagnostic Reference Levels) 15
Conclusion Conclusion 1. Cardiac imaging - demanding application of MDCT - improving with technological advances - better temporal resolution 2. Prospective ECG-triggering, retrospective ECG-gating & scan parameters optimize cardiac imaging protocols & radiation dose 3. Recommended radiation dose optimization strategies applied clinical indications characteristics of patient hardware and software constraints on CT scanner lowest possible dose with diagnostic image quality 4. Radiation dose recording & monitoring dose awareness of radiographers 16
Thank you 17