Can multiphase dynamic CT angiography provide a better assessment of aortic dissection compared with the standard triphasic protocol?

Abstract

Background

Acute aortic dissection (AD) is a life-threatening medical emergency. It has been debated whether the multiphase dynamic computed tomography angiography (CTA) protocol is superior to the standard triphasic protocol for revealing the characteristics of AD.

Purpose

To examine two multiphase dynamic protocols, Dynamic four-dimensional (4D) CTA using the shuttle mode and Flash 4D CTA using the high-pitch mode for the assessment of AD and to compare them with the standard triphasic protocol.

Material and Methods

A total of 54 consecutive patients were randomly and equally assigned to three groups and scanned with a second-generation DSCT scanner. Groups A, B, and C were assessed with the Dynamic 4D CTA in the shuttle mode, the Flash 4D CTA in the high-pitch mode, and the standard triphasic acquisition protocol, respectively. Image quality of all patients was evaluated. The effective radiation dose (ED) was recorded.

Results

In 54 patients, CTA images could display the true and false lumens, the intimal flap, the entry tear, and branch vessel involvement in the AD. Compared with group C, additional diagnostic information was obtained in groups A and B, including the dynamic enhancement delay between the true and false lumens (A = 18, B = 18); the presence of membrane oscillation (A = 8, B = 14); dynamic ejection of the contrast material from the true lumen into the false lumen (A = 6, B = 7); and the dynamic obstruction of the left renal artery (B = 2). The ED in these three groups was significantly different (P < 0.05).

Conclusion

Compared to the standard triphasic protocol, the multiphase dynamic CTA protocol is feasible and is able to reveal additional diagnostic information. Therefore, we recommend using the high-pitch, dual-source multiphase dynamic CTA to assess ADs.

Keywords

Aortic dissection dual-source computed tomography angiography multiphase dynamic imaging lower radiation dose

Introduction

Acute aortic dissection (AD) is a life-threatening medical emergency associated with high rates of morbidity and mortality. It is characterized by a tear in the intimal layer of the aorta, causing an inflow of blood into the media that forms a true and a false lumen that are separated by an intimedial flap (1 –3). With the widespread clinical application of ultrasonic echocardiogram, magnetic resonance imaging (MRI) (4), and multidetector computed tomography (CT), the definition of AD has become increasingly fast and accurate. CT is the most commonly used imaging modality due to its wide availability, accuracy, and large field of view for imaging ADs; its sensitivity and specificity approach 100% (5 –7). CT allows for the early recognition and characterization of AD as well as the determination of the presence of any associated complications, which are findings that are essential for optimizing treatment and improving clinical outcomes (8).

Clinical outcome is determined by the type and extent of AD and the presence of associated complications (e.g. cerebral sequelae, aortic branch involvement, pericardial involvement, and visceral involvement). Presently, percutaneous treatment methods are maturing and have become more prevalent, and AD treatment methods are decided based on the luminal origins of branch vessels and branch obstruction mechanisms (9,10), which determines the level of organ perfusion. Branch obstructions are classified as either static or dynamic with anatomical features (11). Furthermore, the aorta is a dynamic structure, and the intimal flap often oscillates as a result of changing flow conditions (12,13). Therefore, diagnostic information regarding intimal flap dynamics is important. Additionally, false lumen thrombus and hemodynamics in the false lumen are important clinical measurements and have been used as major predictors of prognosis (14). However, these dynamic changes are frequently missed when using standard imaging techniques such as conventional computed tomography angiography (CTA) due to an inability to adequately assess dynamic changes in the AD, such as blood flow patterns, dynamic ejection of the contrast material from the true lumen into the false lumen, and oscillation of the dissection membrane (1).

Recent advances in CT technology have allowed for multiphase dynamic CTA using the shuttle mode or the high-pitch mode (1,15,16). Although magnetic resonance angiography (MRA) (14) and ultrasound have been used for time-resolved imaging of ADs, multiphase dynamic CTA imaging is more readily available and able to acquire images in less time, particularly in an emergency setting; it offers higher spatial resolution than MRA and is less time-consuming and dependent on experience than ultrasound imaging. In this pilot study, we assessed the feasibility and additional diagnostic value of multiphase dynamic CTA protocols (including Dynamic four-dimensional [4D] CTA and Flash 4D CTA) in patients with AD and compared them with the standard triphasic protocol on a dual-source CT (DSCT) scanner.

Material and Methods

Patients

This study was conducted according to the ethical standards of our institution and was approved by our review board. A total of 54 consecutive patients with known or suspected AD (45 men, 9 women; age range, 30–76 years) referred for aortic CTA in our hospital between March 2013 and May 2014 were randomly and equally assigned into three groups, and CTA imaging of all patients were collected and analyzed. The exclusion criterion was a lack of AD as determined by CTA imaging. Out of 54 patients, 15 were referred for aortic CTA to follow up on a previously diagnosed AD and 39 were referred for aortic CTA due to high clinical suspicion of acute AD, which were all confirmed by the CT examination; additionally, 15 out of the 39 patients suspected of AD were further confirmed by surgery or endovascular aortic repair (EVAR).

CT protocols

All 54 patients were scanned on a DSCT scanner (SOMATOM Definition Flash; Siemens Healthcare, Forchheim, Germany). They were randomly and equally divided into groups A, B, and C. All patients first underwent unenhanced CT. After the unenhanced CT scan, patients in group A underwent a CTA exam with the shuttle mode (the Dynamic 4D CT technology allowed for time-resolved CT imaging with bidirectional table movement during image acquisition, thus providing dynamic image acquisitions (1)) for multiphase dynamic image acquisition (range, 48 cm; time resolution, 6 s, 4 phases; 80 kV; 125 mAs/rot). Group B underwent CTA with the high-pitch mode (Flash 4D) for multiphase dynamic image acquisition (range from the plane of entrance to bony thorax to the plane of symphysis pubis; time resolution: 12 s; 4 phases; pitch, 3.0; and CARE kV with ref. 80 kV and ref. 100 mAs, Siemens Healthcare). Group C underwent standard triphasic acquisition of unenhanced, arterial, and portal vein phases (range from the plane of entrance to bony thorax to the plane of symphysis pubis; 100 kV; and 210 m As/rot).

In each patient, 95 mL of iopromide (Ultravist 370, 370 mg I/mL, Bayer Schering Pharma, Berlin, Germany) was injected at a flow rate of 5 mL/s followed by 20 mL saline solution at a flow rate of 6 mL/s, and finally, 30 mL of saline solution at a flow rate of 4 mL/s using a 20-gauge needle and an automatic injector. All groups used the threshold-triggered mode with an attenuation threshold value of 100 HU. The region of interest (ROI) of groups A and B was in the center of the left atrium, while the ROI of group C was in the ascending aortic root.

Image analysis

Subjective assessment: All CT images of each patient were reviewed in consensus by two radiologists with nine and 15 years of experience, respectively, in vascular radiology. The reviewers gave the examinations an overall grade based on a five-point scale (1: non-diagnostic to 5: excellent).

Objective assessment: Image quality was also evaluated objectively by the signal-to-noise ratio (SNR). Mean Hounsfield unit values of the aorta were determined in all phases with a standardized 1 cm² ROI within the true lumen at the place of pulmonary trunk. Background noise was determined by measuring the CT density of a standardized 1 cm² ROI in the air just above the body surface and calculating the standard deviation. SNR was calculated using the following formula: maximum enhancement of the true lumen (HU)/background noise (HU).

Radiation dose: The CT dose indices (CTDIs) and the dose length products (DLPs) were recorded. To calculate the effective radiation dose (ED) from the DLP, a conversion factor of 0.015 mSv/mGy was used (15).

CT imaging: The multiphase dynamic CT images were displayed in movie mode with dynamic visualizations of the axial, coronal, and sagittal slices as well as three-dimensional (3D) volume-rendering reconstructions, while the standard triphasic protocol CT images were observed as conventional axial, coronal, and sagittal slices as well as a 3D reconstruction.

Characteristics of the AD, including the true and false lumens, the intimal flap, the entry tear, and branch vessel involvement as well as additional diagnostic information regarding the AD, including the enhancement delay between the true and false lumens, the presence of membrane oscillation, and the dynamic ejection of contrast material from the true lumen into the false lumen were evaluated.

Statistical analysis

All analyses were performed using SPSS19.0. (SPSS Inc., Chicago, IL, USA). Quantitative data including ED, SNR, and subjective image quality were expressed as the mean ± SD and analyzed by one-way ANOVA among the three groups. If the variance was homogeneous, pair-wise comparisons were performed using Tukey’s test; if the variance was inhomogeneous, Dunnett’s T3 method was used for pair-wise comparisons. A P value < 0.05 was considered statistically significant.

Results

AD was eventually confirmed in all 54 patients included in the study by CTA (Table 1). In all 54 patients, CTA was able to discern the true and false lumens, the intimal flap, the entry tear, and branch vessel involvement in the AD. Compared with the standard triphasic protocol (unenhanced, arterial, and portal vein phase scans), additional diagnostic information could also be obtained by the multiphase dynamic CTA protocol, including the visualization of membrane oscillation (A = 8, B = 14), dynamic ejection of the contrast material from the true lumen into the false lumen (A = 6, B = 7), and dynamic obstruction of the left renal artery (B = 2) (Figs. 1 –3).

Table 1.

Imaging findings in a total of 54 AD patients with the shuttle mode (Group A, n = 18), Flash 4D mode (Group B, n = 18), and traditional scanning mode (Group C, n = 18).

	Group A	Group B	Group C
Stanford classification
Stanford type A	5	12	2
Stanford type B	13	6	16
True and false lumens, intimal flap	18	18	18
Entry tear	18	18	18
Branch involvement
Static obstruction	3	6	8
Dynamic obstruction	0	2	0
Renal perfusion asymmetry	7	9	7
Additional diagnostic information
Enhancement delay between the true and false lumens	18	18	13
Membrane oscillation	8	14	0
Dynamic ejection of the contrast material from the true lumen into the false lumen	6	7	0
Dynamic obstruction of the left renal artery		2

Stanford type A dissection involves the ascending aorta regardless of the origin of the intimal tear or the extent of the dissection; Stanford type B dissection affects only the descending aorta.

AD, aortic dissection.

Fig. 1.

In a 60-year-old man with type A AD, Flash 4D CTA showed dynamic changes of blood flow in the true and false lumens and membrane oscillation (a–d). When the intimal flap oscillated to the left, the intimal entry was closed (c); when the intimal flap oscillated to the right, the intimal entry was open (d).

Fig. 2.

In a 43-year-old woman with type B AD, Flash 4D CTA showed dynamic renal perfusion asymmetry and dynamic obstruction of the left renal artery (a–d). When the intimal flap oscillated to the posterior, the intimal flap covered the opening of the left renal artery, and left renal perfusion was lower than right renal perfusion (a). When the intimal flap oscillated to the anterior, the intimal flap did not cover the opening of the left renal artery; therefore, left renal perfusion was comparable to right renal perfusion (c).

Fig. 3.

In a 60-year-old woman with type B AD, shuttle mode 4D CTA showed prominent enhancement delay in the false lumen, showed that the right renal artery originated from the false lumen, and showed that the superior portion of the right kidney appeared to be infarcted (a–d).

The radiation dose of each group was recorded and shown in Table 2. The CTDIs (mGy) of groups A, B, and C were 11.63 ± 0.16, 11.00 ± 0.35, and 22.35 ± 1.13, respectively. The effective radiation dose (ED) among these three groups were statistically different (P < 0.05). The ED of group A (8.80 ± 0.12 mSv) was significantly lower than that of groups B and C (11.60 ± 0.38 mSv and 23.86 ± 1.30 mSv, respectively).

Table 2.

Radiation dose of group A (shuttle mode), group B (Flash 4D mode), and group C (standard triphasic protocol).

Group	CTDI_vol (mGy)	DLP (mGy.cm)	ED (mSv)
A	11.63 ± 0.16	586.28 ± 8.04	8.80 ± 0.12
B	11.00 ± 0.35	773.22 ± 25.36	11.60 ± 0.38
C	22.35 ± 1.13	1473.15 ± 118.41	23.86 ± 1.30
P (A/C)	0.000	0.000	0.000
P (B/C)	0.000	0.000	0.000
P (A/B)	0.287	0.000	0.000

CTDIvol, volume CT dose index; DLPs, length products; ED, effective dose.

The aortic CT enhancement values are shown in Table 3. The SNR between the A/C groups and B/C groups was significantly different (P < 0.05). There was no significant difference of SNR between groups A and B.

Table 3.

Aortic CT enhancement values in group A (shuttle mode), group B (Flash 4D mode), and group C (standard triphasic protocol).

Group	SNR	CNR	Maximum of aorta
A	27.98 ± 3.80	25.19 ± 3.63	609.28 ± 34.57
B	17.19 ± 1.84	15.09 ± 1.66	407.61 ± 21.01
C	7.30 ± 0.46	20.90 ± 2.52	372.93 ± 26.45
P (A/C)	0.000	0.509	0.000
P (B/C)	0.000	0.296	0.065
P (A/B)	0.050	0.030	0.000

CNR, contrast-to-noise ratio; SNR, signal-to-noise ratio.

The subjective image quality (SIQ) scores of the two radiologists by consensus is shown in Table 4. All images were of diagnostic quality and none had a score of 1 (unacceptable) or 2. The SIQ of groups A, B, and C was 4.61 ± 0.61, 4.28 ± 0.67, and 4.00 ± 0.59, respectively. The SIQ in group A was statistically higher than that in group C (P < 0.05). However, there were no significant differences between groups A and B as well as groups B and C.

Table 4.

Subjective image quality of groups A, B, and C.

Score	5	4	3
Group A	12	5	1
Group B	7	9	2
Group C	3	12	3

Discussion

AD is the most common acute emergency condition of the aorta and often has a fatal outcome. Prompt and correct diagnosis of AD is critical for rapid and timely treatment, better prognosis, decreasing mortality, and increasing the survival rate. CT remains the technique of choice for evaluating patients with suspected AD due to its speed, availability, and sensitivity. It allows for detailed morphological and functional assessment of the aorta. Conventional standard CT imaging techniques are unable to adequately assess dynamic changes in Ads’. In this study, we evaluated the feasibility and the additional diagnostic value of multiphase dynamic CTA imaging of the aorta in patients with AD.

The multiphase dynamic protocols (including the shuttle mode and the Flash 4D mode) and the standard triphasic protocol could both discern the true and false lumens, the intimal flap, the entry tear, and the branch vessels involvement in the AD as well as significant dynamic renal perfusion asymmetry. Additionally, the SIQ was mostly rated as excellent or good for both protocols in our study. However, dynamic obstruction of the left renal artery in two patients was observed only with the multiphase dynamic protocol (Flash 4D mode), which suggested that the renal artery did not require immediate treatment due to the absence of renal ischemia. In addition, other additional diagnostic information, such as the dynamic enhancement delay between the true and false lumens, membrane oscillation, and dynamic ejection of the contrast material from the true lumen into the false lumen, could only be obtained using the multiphase dynamic CTA protocols. It might be because AD is a dynamic disease that undergoes changes during the cardiac cycle; therefore, dynamic imaging modalities might be more suitable for the assessment of AD than the standard triphasic CT protocol.

Compared with the standard triphasic CT protocol (unenhanced, arterial, and venous scans), which has an ED of approximately 23.86 ± 1.30 mSv, the multiphase dynamic CTA protocols used in this study had statistically lower EDs; the shuttle mode had an ED of 8.80 ± 0.12 mSv, and the Flash 4D mode had an ED of 11.60 ± 0.38 mSv. Meinel et al. (1) reported an average dose of 27.7 ± 3.5 mSv in their multiphase dynamic CTA study, which was much higher than that reported in this study. This difference might be because our multiphase dynamic CTA protocol included four phases, while theirs included six phases. However, the average mean radiation doses for each phase in this study, which were 2.2 mSv (8.80 ± 0.12 mSv/4) with the shuttle mode and 2.9 mSv (11.60 ± 0.38 mSv/4) with the Flash 4D mode, were still lower than that in Meinel’s study, which was 4.6 mSv (27.7 ± 3.5 mSv). Another reason could be that we had set the tube voltage at 80 kVp whereas they set it at 100 kVp. As suggested by Lovy et al. (17), using decreased tube voltage settings, when appropriate, could significantly decrease radiation exposure. In this study, the ED of group A (8.80 ± 0.12 mSv) was significantly lower than that of group B (11.60 ± 0.38 mSv). However, the CTDIvol shown in Table 2 indicates that there were small differences between groups A and B. Ultimately, group B had the lowest radiation exposure. The DLP, which had a higher scan range in group B, showed that scan-length was the main cause of the differences observed in radiation exposure between groups A and B.

Some recent studies have shown that imaging of the thoraco-abdominal aorta with ECG-triggered CTA provides higher quality images, which can compensate for motion artifacts in the aortic root complex that may mimic the appearance of a dissection in the ascending aorta and thus lead to the misdiagnosis of dissection (18 –20). However, shuttle mode imaging in this study was not combined with ECG triggering. For a better comparison, ECG triggering in Flash 4D mode was not used. In this study, cardiac motion artifacts did not pose a major challenge, since the information from multiple phases allowed for an accurate assessment of the aortic root or ascending aorta even when the images from a single phase were blurred by motion artifacts. Beeres et al. concluded that high-pitch DSCTA of the whole aorta is a robust and dose-efficient examination strategy for the evaluation of aortic pathologies with or without ECG gaiting (21).

In this study, the scan length of the shuttle mode for multiphase dynamic image acquisition was fixed to 48 cm, which was the maximum setting for the second-generation DSCT. While imaging in Flash 4D mode, the scan range could be adjusted for different patients, which was set from the plane of the entrance to the bony thorax to the plane of the symphysis pubis. The scan range of 48 cm in the shuttle mode is likely sufficient for most clinical applications. However, if the AD extends into the iliac arteries, especially during the preoperative evaluation for endovascular treatment and in the emergency setting when the full extent of the AD needs to be determined, it would be better to perform multiphase dynamic CTA in the Flash 4D mode or use a third-generation DSCT in which the scan length in shuttle mode could be extended to 80 cm.

To our knowledge, our study was the first to compare the multiphase dynamic CTA protocols (including the shuttle mode and the Flash 4D mode) with the standard triphasic protocol in patients with AD.

The study also had some limitations. First, the patient number in each group was relatively small. Second, although additional diagnostic information was obtained by the multiphase dynamic CTA protocol, the correlations between these findings and clinical symptoms or clinical outcomes could be further elucidated, such as the correlation between renal perfusion asymmetry and clinical laboratory tests or membrane oscillation and endovascular stent repair. To enable the generalization of these findings, larger trials are required to further confirm the clinical benefits of the additional diagnostic information in the treatment of AD.

In conclusion, multiphase dynamic CTA imaging of the entire aorta is feasible. Compared with the standard triphasic protocol, multiphase dynamic CTA imaging can accurately reveal the pathological and anatomical features of AD with relatively low radiation dose and may lead to a change in therapeutic strategy for some patients. Since every case of AD is unique, multiphase CT dynamic scanning techniques offer the possibility of understanding the precise morphology and hemodynamic environment inside the aorta, which can help radiologists and clinicians gain a more-accurate depiction of the dynamics in an AD. Thus, it may be the optimal choice for AD patients.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Meinel

Nikolaou

Weidenhagen

et al.

Time-resolved CT angiography in aortic dissection. Eur J Radiol 2012; 81: 3254–3261.

Chiu

Lakshminarayan

Ettles

. Acute aortic syndrome: CT findings. Clin Radiol 2013; 68: 741–748.

Hagan

Nienaber

Isselbacher

et al.

The International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. JAMA 2000; 283: 897–903.

Raptis

Fowler

Narra

et al.

Emergency thoracic vascular magnetic resonance imaging: protocols and clinical considerations. Semin Roentgenol 2014; 49: 157–168.

Manghat

Morgan-Hughes

Roobottom

. Multi-detector row computed tomography: imaging in acute aortic syndrome. Clin Radiol 2005; 60: 1256–1267.

Evangelista

Carro

Moral

et al.

Imaging modalities for the early diagnosis of acute aortic syndrome. Nat Rev Cardiol 2013; 10: 477–486.

Shiga

Wajima

Apfel

et al.

Diagnostic accuracy of transesophageal echocardiography, helical computed tomography, and magnetic resonance imaging for suspected thoracic aortic dissection: systematic review and meta-analysis. Arch Intern Med 2006; 166: 1350–1356.

McMahon

Squirrell

. Multidetector CT of aortic dissection: A pictorial review. Radiographics 2010; 30: 445–460.

Midulla

Fattori

Beregi

et al.

Aortic dissection and malperfusion syndrome: a when, what and how-to guide. Radiol Med 2013; 118: 74–88.

10.

LePage

Quint

Sonnad

et al.

Aortic dissection: CT features that distinguish true lumen from false lumen. Am J Roentgenol 2001; 177: 207–211.

11.

Williams

Lee

Hamilton

et al.

The dissected aorta: part III. Anatomy and radiologic diagnosis of branch-vessel compromise. Radiology 1997; 203: 37–44.

12.

Ganten

Weber

von Tengg-Kobligk

et al.

Motion characterization of aortic wall and intimal flap by ECG-gated CT in patients with chronic B-dissection. Eur J Radiol 2009; 72: 146–153.

13.

Yang

Chao

et al.

Abdominal aortic intimal flap motion characterization in acute aortic dissection: assessed with retrospective ECG-gated thoracoabdominal aorta dual-source CT angiography. PLoS One 2014; 9: e87664–e87664.

14.

Clough

Zymvragoudakis

Biasi

et al.

Usefulness of new imaging methods for assessment of type B aortic dissection. Ann Cardiothorac Surg 2014; 3: 314–318.

15.

Liu

et al.

The ascending aortic image quality and the whole aortic radiation dose of high-pitch dual-source CT angiography. J Cardiothorac Surg 2013; 8: 228–228.

16.

Sommer

Clevert

Bamberg

et al.

Time-resolved computed tomography imaging of the aorta: a feasibility study. J Thorac Imaging 2010; 25: 161–167.

17.

Lovy

Rosenblum

Levsky

et al.

Acute aortic syndromes: a second look at dual-phase CT. Am J Roentgenol 2013; 200: 805–811.

18.

Bolen

Popovic

Tandon

et al.

Image quality, contrast enhancement, and radiation dose of ECG-triggered high-pitch CT versus non-ECG-triggered standard-pitch CT of the thoracoabdominal aorta. Am J Roentgenol 2012; 198: 931–938.

19.

Schernthaner

Stadler

Beitzke

et al.

Dose modulated retrospective ECG-gated versus non-gated 64-row CT angiography of the aorta at the same radiation dose: comparison of motion artifacts, diagnostic confidence and signal-to-noise-ratios. Eur J Radiol 2012; 81: e585–590.

20.

Roos

Willmann

Weishaupt

et al.

Thoracic aorta: motion artifact reduction with retrospective and prospective electrocardiography-assisted multi-detector row CT. Radiology 2002; 222: 271–277.

21.

Beeres

Wichmann

Frellesen

et al.

ECG-gated versus non-ECG-gated high-pitch dual-source CT for whole body CT angiography (CTA). Acad Radiol 2016; 23: 163–167.