Eur Radiol DOI 10.1007/s00330-014-3533-y
HEAD AND NECK
Radiation dose and image quality of 70 kVp cerebral CT angiography with optimized sinogram-affirmed iterative reconstruction: comparison with 120 kVp cerebral CT angiography Guo Zhong Chen & Long Jiang Zhang & U. Joseph Schoepf & Julian L. Wichmann & Cole M. Milliken & Chang Sheng Zhou & Li Qi & Song Luo & Guang Ming Lu
Received: 17 July 2014 / Revised: 15 October 2014 / Accepted: 20 November 2014 # European Society of Radiology 2014
Abstract Objectives To evaluate radiation dose, image quality, and optimal level of sinogram-affirmed iterative reconstruction (SAFIRE) of cerebral CT angiography (CTA) at 70 kVp. Methods One hundred patients were prospectively classified into two groups: Group A (n=50), 70 kVp cerebral CTA with 5 levels of SAFIRE reconstruction (S1-S5); and Group B (n= 50), 120 kVp with filtered back projection (FBP) reconstruction. CT attenuation values, noise, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the internal carotid artery (ICA) and middle cerebral artery (MCA) were measured. Subjective image quality was evaluated. Effective dose (ED) was estimated. Results CT attenuation and noise of the ICA and MCA in Group A were higher than those of Group B (all P0.05) or higher than in Group B (P0.05). ED was 0.2±0.0 mSv for Group A with 85 % ED reduction in comparison to Group B (1.3±0.2 mSv). G. Z. Chen : L. J. Zhang (*) : U. J. Schoepf : C. S. Zhou : L. Qi : S. Luo : G. M. Lu (*) Department of Medical Imaging, Jinling Hospital, Medical School of Nanjing University, Nanjing, Jiangsu 210002, China e-mail: [email protected] e-mail: [email protected] U. J. Schoepf : J. L. Wichmann : C. M. Milliken Department of Radiology and Radiological Science, Medical University of South Carolina, Ashley River Tower, MSC 226, 25 Courtenay Dr, Charleston, SC 29401, USA J. L. Wichmann Department of Diagnostic and Interventional Radiology, University Hospital Frankfurt, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany
Conclusion Cerebral CTA at 70 kVp is feasible, allowing for substantial radiation dose reduction. SAFIRE S4 level is recommended for obtaining optimal image quality. Key points • 70 kVp cerebral CTA is feasible and provides diagnostic image quality. • 70 kVp cerebral CTA resulted in 85 % effective dose reduction. • S4 level of SAFIRE is recommended for 70 kVp cerebral CTA. Keywords Angiography . Low tube voltage . Radiation dose . Tomography X-ray computed . Intracranial aneurysm
Introduction Computed tomography angiography (CTA) has high sensitivity and specificity for the detection of intracranial aneurysms [1–3] and is, thus, increasingly used in the emergency setting to detect intracranial aneurysms in patients with subarachnoid haemorrhage. However, CTA involves a substantial radiation dose to sensitive organs, including the eye lens, and repeated follow-up examinations may result in high cumulative radiation exposure to patients. Thus, strategies to reduce the radiation dose associated with CTA are desirable. Several dose-reduction strategies have been investigated and have proven to be efficient. These include a low tube voltage, an automated tube voltage selection, a low tube current, automatic exposure control, acquisition length optimization, and high pitch techniques [4–12]. Low tube voltage appears to be the most effective radiation dose reduction strategy in cerebral CTA because radiation dose is proportional to the square of tube voltage [5, 6] while the CT attenuation of enhanced vessels can be substantially increased
Eur Radiol
by lowering tube voltage [7]. One recent study demonstrated that 100 kVp and 80 kVp CTA provide comparable cerebral vessel image quality and have similar diagnostic accuracy compared with 120 kVp CTA for the detection of intracranial aneurysms [8]. Recently, 70 kVp settings have been applied in lower extremity CTA [9, 10], coronary CTA [11], and cerebral CT perfusion [12]; however, to our knowledge, the feasibility of 70 kVp cerebral CTA has not yet been investigated. The use of 70 kVp will increase image noise in comparison to 80 kVp or 100 kVp acquisition. Iterative reconstruction (IR) can substantially improve image quality by reducing image noise, compared with traditional filtered back projection (FBP) under the same image acquisition conditions [3, 4]. Sinogram-affirmed iterative reconstruction (SAFIRE) is a hybrid algorithm, based in both the image and raw-data domains, which can be applied to low tube voltage scans to reduce image noise. Of the five SAFIRE levels (S1-5), level S3 is recommended for general use by the manufacturer [13–16]. However, the optimal SAFIRE level for the specific application of noise reduction at low voltage cerebral CTA has not been established. Therefore, the purpose of this study was to evaluate radiation dose, image quality, and optimal SAFIRE level for cerebral CTA at 70 kVp.
Material and methods Patients This prospective study was performed with institutional review board approval. Informed consent was obtained from all patients or their legal guardians. One hundred and twenty consecutive patients with a suspected subarachnoid haemorrhage, intracranial aneurysm, or cerebrovascular disease were enrolled in this study between January 2014 and April 2014. Contraindications for cerebral CTA included known prior reactions to iodinated contrast agents, severe renal impairment, pregnancy, and lactation period. Patients with moyamoya disease, motion artefacts, occlusion of the internal Fig. 1 ROI placement in the internal carotid artery and middle cerebral artery A, B: axial contrast-enhanced CT images show the ROI placements in the cavernous segment of the internal carotid artery (A) and middle cerebral artery (B)
carotid artery (ICA) or middle cerebral artery (MCA), or intracranial clipping or coiling were excluded from the study in order to insure the accuracy of ICA and/or MCA attenuation measurements. Each patient’s body weight and height were measured and recorded prior to the CT examination. Cerebral CT angiography Acquisition protocol All cerebral CTA examinations were performed on a dual source CT system (Somatom Definition Flash, Siemens Healthcare, Forchheim, Germany). The patients were randomly assigned to one of two groups according to CT tube voltage. In Group A, the patients were examined at 70 kVp and at 120 kVp in Group B. Image acquisition parameters included a collimation of 64×2×0.6 mm, a rotation time of 0.5 seconds, and a pitch of 1.2. Automatic anatomic tube current modulation (CAREDose 4D, Siemens) was used in each patient for both CTA protocols. Sixty mL of iodinated contrast medium (iopromide, Ultravist 300 mg I/mL, Bayer Schering Pharma, Berlin, Germany) was intravenously injected at a flow rate of 4.0 mL/s, followed by 30 mL of saline solution through an antecubital vein at the same flow rate via an 18-gauge catheter. CT acquisition was triggered by using a bolus tracking technique with the region of interest (ROI) placed in the right internal carotid artery (ICA). Image acquisition was started 3 seconds after attenuation reached the predefined threshold of 100 HUs. Acquisition time was approximately three to four seconds. Image reconstruction CT images were reconstructed at 120 kVp using a conventional filtered back projection (FBP) algorithm. CT images at 70 kVp were reconstructed at 5 preset SAFIRE strength levels (S1-5). All image series were reconstructed with a 0.75 mm section thickness and 0.5 mm increment. The convolution algorithms (“kernels”) of groups A and B were the
Eur Radiol Fig. 2 Samples of cerebral CTA image quality evaluation on a four-point scale. Panel A: sagittal maximum intensity projection reformatted image shows excellent overall image quality (score 4) with little noise and the sharpest vessel contour; Panel B: sagittal maximum intensity projection reformatted image shows image quality with score 3; Panel C: sagittal maximum intensity projection reformatted image shows image quality of score 2; Panel D: sagittal maximum intensity projection reformatted image shows poor image quality (score 1), intracranial arteries are too noisy and blurry to diagnose
corresponding “J30f” and “H30f” kernels, respectively. The time required for reconstruction was recorded. Each image dataset was coded, patient information was removed, and the sets were randomized to enable double-blinded evaluation. Image quality evaluation Objective image quality evaluation All images were transferred to a dedicated workstation (Multi Modality Workplace; Siemens Healthcare). All measurements Table 1
were performed by one neuroradiologist (S.L. with five years of experience in neuro-imaging). CT attenuation was measured by placing a circular ROI in the centre of the ICA (Fig. 1A) and MCA on both sides of the axial images (Fig. 1B). The ROI area was 0.15-0.2 cm2 for the ICA and 0.03-0.06 cm2 for the MCA. To avoid partial volume effects, ROIs were placed in the cavernous segment of the ICA and the first segment of the MCA [8]. The ROI was drawn within the target vessel without including the walls. Measurements were performed three times. Average CT attenuation values of the left and right ICA/MCA were calculated for each patient,
Objective image quality comparison of 70 kVp and 120 kVp groups
Locations
ICA CT value (HU) SNR CNR MCA CT value (HU) SNR CNR Parenchyma CT value (HU) Noise (HU)
120 kVp
70 kVp
P value
FBP
S1
S2
S3
S4
S5
287.1±45.8 33.9±7.4 30.1±7.0
510.4±88.7 22.7±4.5 21.0±4.4
513.2±87.8 25.1±5.0 23.3±4.9
515.2±89.7 28.2±5.8 26.1±5.6
514.0±91.6 31.3±6.6 29.1±6.4
517.4±87.8 35.6±7.0 33.0±6.7
HEAD AND NECK
Radiation dose and image quality of 70 kVp cerebral CT angiography with optimized sinogram-affirmed iterative reconstruction: comparison with 120 kVp cerebral CT angiography Guo Zhong Chen & Long Jiang Zhang & U. Joseph Schoepf & Julian L. Wichmann & Cole M. Milliken & Chang Sheng Zhou & Li Qi & Song Luo & Guang Ming Lu
Received: 17 July 2014 / Revised: 15 October 2014 / Accepted: 20 November 2014 # European Society of Radiology 2014
Abstract Objectives To evaluate radiation dose, image quality, and optimal level of sinogram-affirmed iterative reconstruction (SAFIRE) of cerebral CT angiography (CTA) at 70 kVp. Methods One hundred patients were prospectively classified into two groups: Group A (n=50), 70 kVp cerebral CTA with 5 levels of SAFIRE reconstruction (S1-S5); and Group B (n= 50), 120 kVp with filtered back projection (FBP) reconstruction. CT attenuation values, noise, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the internal carotid artery (ICA) and middle cerebral artery (MCA) were measured. Subjective image quality was evaluated. Effective dose (ED) was estimated. Results CT attenuation and noise of the ICA and MCA in Group A were higher than those of Group B (all P0.05) or higher than in Group B (P0.05). ED was 0.2±0.0 mSv for Group A with 85 % ED reduction in comparison to Group B (1.3±0.2 mSv). G. Z. Chen : L. J. Zhang (*) : U. J. Schoepf : C. S. Zhou : L. Qi : S. Luo : G. M. Lu (*) Department of Medical Imaging, Jinling Hospital, Medical School of Nanjing University, Nanjing, Jiangsu 210002, China e-mail: [email protected] e-mail: [email protected] U. J. Schoepf : J. L. Wichmann : C. M. Milliken Department of Radiology and Radiological Science, Medical University of South Carolina, Ashley River Tower, MSC 226, 25 Courtenay Dr, Charleston, SC 29401, USA J. L. Wichmann Department of Diagnostic and Interventional Radiology, University Hospital Frankfurt, Theodor-Stern-Kai 7, 60590 Frankfurt, Germany
Conclusion Cerebral CTA at 70 kVp is feasible, allowing for substantial radiation dose reduction. SAFIRE S4 level is recommended for obtaining optimal image quality. Key points • 70 kVp cerebral CTA is feasible and provides diagnostic image quality. • 70 kVp cerebral CTA resulted in 85 % effective dose reduction. • S4 level of SAFIRE is recommended for 70 kVp cerebral CTA. Keywords Angiography . Low tube voltage . Radiation dose . Tomography X-ray computed . Intracranial aneurysm
Introduction Computed tomography angiography (CTA) has high sensitivity and specificity for the detection of intracranial aneurysms [1–3] and is, thus, increasingly used in the emergency setting to detect intracranial aneurysms in patients with subarachnoid haemorrhage. However, CTA involves a substantial radiation dose to sensitive organs, including the eye lens, and repeated follow-up examinations may result in high cumulative radiation exposure to patients. Thus, strategies to reduce the radiation dose associated with CTA are desirable. Several dose-reduction strategies have been investigated and have proven to be efficient. These include a low tube voltage, an automated tube voltage selection, a low tube current, automatic exposure control, acquisition length optimization, and high pitch techniques [4–12]. Low tube voltage appears to be the most effective radiation dose reduction strategy in cerebral CTA because radiation dose is proportional to the square of tube voltage [5, 6] while the CT attenuation of enhanced vessels can be substantially increased
Eur Radiol
by lowering tube voltage [7]. One recent study demonstrated that 100 kVp and 80 kVp CTA provide comparable cerebral vessel image quality and have similar diagnostic accuracy compared with 120 kVp CTA for the detection of intracranial aneurysms [8]. Recently, 70 kVp settings have been applied in lower extremity CTA [9, 10], coronary CTA [11], and cerebral CT perfusion [12]; however, to our knowledge, the feasibility of 70 kVp cerebral CTA has not yet been investigated. The use of 70 kVp will increase image noise in comparison to 80 kVp or 100 kVp acquisition. Iterative reconstruction (IR) can substantially improve image quality by reducing image noise, compared with traditional filtered back projection (FBP) under the same image acquisition conditions [3, 4]. Sinogram-affirmed iterative reconstruction (SAFIRE) is a hybrid algorithm, based in both the image and raw-data domains, which can be applied to low tube voltage scans to reduce image noise. Of the five SAFIRE levels (S1-5), level S3 is recommended for general use by the manufacturer [13–16]. However, the optimal SAFIRE level for the specific application of noise reduction at low voltage cerebral CTA has not been established. Therefore, the purpose of this study was to evaluate radiation dose, image quality, and optimal SAFIRE level for cerebral CTA at 70 kVp.
Material and methods Patients This prospective study was performed with institutional review board approval. Informed consent was obtained from all patients or their legal guardians. One hundred and twenty consecutive patients with a suspected subarachnoid haemorrhage, intracranial aneurysm, or cerebrovascular disease were enrolled in this study between January 2014 and April 2014. Contraindications for cerebral CTA included known prior reactions to iodinated contrast agents, severe renal impairment, pregnancy, and lactation period. Patients with moyamoya disease, motion artefacts, occlusion of the internal Fig. 1 ROI placement in the internal carotid artery and middle cerebral artery A, B: axial contrast-enhanced CT images show the ROI placements in the cavernous segment of the internal carotid artery (A) and middle cerebral artery (B)
carotid artery (ICA) or middle cerebral artery (MCA), or intracranial clipping or coiling were excluded from the study in order to insure the accuracy of ICA and/or MCA attenuation measurements. Each patient’s body weight and height were measured and recorded prior to the CT examination. Cerebral CT angiography Acquisition protocol All cerebral CTA examinations were performed on a dual source CT system (Somatom Definition Flash, Siemens Healthcare, Forchheim, Germany). The patients were randomly assigned to one of two groups according to CT tube voltage. In Group A, the patients were examined at 70 kVp and at 120 kVp in Group B. Image acquisition parameters included a collimation of 64×2×0.6 mm, a rotation time of 0.5 seconds, and a pitch of 1.2. Automatic anatomic tube current modulation (CAREDose 4D, Siemens) was used in each patient for both CTA protocols. Sixty mL of iodinated contrast medium (iopromide, Ultravist 300 mg I/mL, Bayer Schering Pharma, Berlin, Germany) was intravenously injected at a flow rate of 4.0 mL/s, followed by 30 mL of saline solution through an antecubital vein at the same flow rate via an 18-gauge catheter. CT acquisition was triggered by using a bolus tracking technique with the region of interest (ROI) placed in the right internal carotid artery (ICA). Image acquisition was started 3 seconds after attenuation reached the predefined threshold of 100 HUs. Acquisition time was approximately three to four seconds. Image reconstruction CT images were reconstructed at 120 kVp using a conventional filtered back projection (FBP) algorithm. CT images at 70 kVp were reconstructed at 5 preset SAFIRE strength levels (S1-5). All image series were reconstructed with a 0.75 mm section thickness and 0.5 mm increment. The convolution algorithms (“kernels”) of groups A and B were the
Eur Radiol Fig. 2 Samples of cerebral CTA image quality evaluation on a four-point scale. Panel A: sagittal maximum intensity projection reformatted image shows excellent overall image quality (score 4) with little noise and the sharpest vessel contour; Panel B: sagittal maximum intensity projection reformatted image shows image quality with score 3; Panel C: sagittal maximum intensity projection reformatted image shows image quality of score 2; Panel D: sagittal maximum intensity projection reformatted image shows poor image quality (score 1), intracranial arteries are too noisy and blurry to diagnose
corresponding “J30f” and “H30f” kernels, respectively. The time required for reconstruction was recorded. Each image dataset was coded, patient information was removed, and the sets were randomized to enable double-blinded evaluation. Image quality evaluation Objective image quality evaluation All images were transferred to a dedicated workstation (Multi Modality Workplace; Siemens Healthcare). All measurements Table 1
were performed by one neuroradiologist (S.L. with five years of experience in neuro-imaging). CT attenuation was measured by placing a circular ROI in the centre of the ICA (Fig. 1A) and MCA on both sides of the axial images (Fig. 1B). The ROI area was 0.15-0.2 cm2 for the ICA and 0.03-0.06 cm2 for the MCA. To avoid partial volume effects, ROIs were placed in the cavernous segment of the ICA and the first segment of the MCA [8]. The ROI was drawn within the target vessel without including the walls. Measurements were performed three times. Average CT attenuation values of the left and right ICA/MCA were calculated for each patient,
Objective image quality comparison of 70 kVp and 120 kVp groups
Locations
ICA CT value (HU) SNR CNR MCA CT value (HU) SNR CNR Parenchyma CT value (HU) Noise (HU)
120 kVp
70 kVp
P value
FBP
S1
S2
S3
S4
S5
287.1±45.8 33.9±7.4 30.1±7.0
510.4±88.7 22.7±4.5 21.0±4.4
513.2±87.8 25.1±5.0 23.3±4.9
515.2±89.7 28.2±5.8 26.1±5.6
514.0±91.6 31.3±6.6 29.1±6.4
517.4±87.8 35.6±7.0 33.0±6.7
