1. Introduction

Follow-up imaging is required after complex surgical procedures involving the ascending aorta to detect postoperative complications and to evaluate for interval growth of aneurysms or dissections in parts of the aorta that have not been surgically replaced. Available imaging modalities include echocardiography, computed tomography (CT) and magnetic resonance imaging (MRI).[1,2]

CT angiography is most frequently used for pre- and postsurgical imaging.[1,3] CT allows multiplanar reformations of excellent image quality[4]Klicken Sie hier, um Text einzugeben. and has been shown to have good accuracy and reproducibility when electrocardiogram (ECG)-gating is used.[5,6] Although the radiation exposure associated with CT has been significantly reduced through new CT techniques such as prospective ECG-triggering, the cumulative radiation dose should not be underestimated particularly in young patients who often need lifelong follow-up.[2,7]

MR angiography is a safe alternative without ionizing radiation, which allows precisely measuring diameter of aneurysms and visualizing location and extent of aortic pathologies[8] As with CT, ECG-gating is recommended in MR angiography for pathologies involving the aortic root or ascending aorta to minimize cardiac motion artifacts. To date, the performance of ECG-gated MR angiography has not been evaluated in the post-operative imaging after surgery involving the ascending aorta.

Therefore, the aim of this study was to evaluate ECG-gated MR angiography in the post-surgical follow-up of patients after aortic surgery involving the ascending aorta regarding technical feasibility, image quality, spectrum of findings and their implications for clinical management.

2. Materials and methods 2.1. Patient population

This study was conducted as a retrospective single-center cohort study. We included all 19 patients who were examined with ECG-gated MR angiography between January and September 2018 for the post-surgical follow-up after aortic surgery involving the ascending aorta. This retrospective study was approved by the institutional review board (Ethics Committee, University Medical Center Rostock) with waiver of informed consent.

2.2. MR imaging protocol

All MR examinations were performed on a 3 Tesla system (Magnetom Verio, Siemens Healthineers, Erlangen, Germany) using a phased-array surface coil. The MR imaging protocol consisted of

localizer sequences fast T2-weigted spin echo sequences in transverse orientation through the entire thorax and abdomen (Half-Fourier Acquisition Single-shot Turbo spin Echo imaging, HASTE) acquired during 2 breath-holds a time-resolved, contrast-enhanced, non-ECG-gated 3D MR angiography of the thorax (Time-resolved angiography With Interleaved Stochastic Trajectories, TWIST sequence) an ECG-gated 3D MR angiography (Fast Low-Angle Shot, FLASH sequence) before and after contrast administration. The ECG-gated MR angiography was acquired in 2 stations (thorax first, immediately followed by the abdomen) to cover the entire aorta using the same contrast bolus. (The abdominal acquisition was omitted in 6 patients with isolated pathology of the thoracic aorta.)

Detailed information about the magnetic resonance angiography (MRA) protocol can be found in the Table S1, Supplemental Digital Content, https://links.lww.com/MD/J23.

For the time-resolved MR angiography, a fixed bolus of 2 mL gadobutrol (Gd-BT-DO3A, Gadovist 1.0 mmol/mL; Bayer Vital, Leverkusen, Germany) was used followed by 30 mL of saline at a flow rate of 2 mL/s. For the time-resolved MRA, image acquisition and contrast injection were started simultaneously. Images were acquired in a sagittal oblique plane with a slab thickness of 84 mm. For the ECG-gated MR angiography, gadobutrol was administered at 0.1 mmol/kg body weight, again followed by a 30 mL saline chaser at a flow rate of 2 mL/s in all patients. Images were acquired in a sagittal oblique plane with a slab thickness of 60 mm. Contrast delay was chosen individually by determining the maximum aortic enhancement in the time-resolved angiography sequence. For both the time-resolved MR angiography and the ECG-gated static MR angiography, patients were instructed to hold their breath at expiration as long as they can and then continue shallow breathing.

2.3. Analysis of clinical data

Clinical data were surveyed by using both digital and paper charts of all patients. Basic patient information as age and BMI were summarized as well as data related to the surgery technique.

2.4. Analysis of image analysis

Image quality was assessed in consensus by 2 radiologists (1 fellow: initials A.B., 1 board-certified attending with sub-specialization in cardiovascular imaging, initials: F.G.M.) for each of the following structures: aortic root, ascending aorta, arch, supra-aortic vessels, descending aorta. This was done separately for ECG-gated MR angiography (Flash 3D sequence) and time-resolved MR angiography (TWIST sequence). Image quality was evaluated in the order in which sequences were acquired, i.e., time-resolved MRA was evaluated first followed by static ECG-gated MRA. Subjective image quality was rated on an established 5-point scale[9]:

5 = excellent, optimal enhancement to allow unambiguous diagnosis of the presence or absence of a pathology;

4 = good, clear enhancement of the structure with slight blurring;

3 = sufficient, reduced enhancement of the structure which still allows confident diagnosis of the presence or absence of a pathology;

2 = poor, inadequate opacification of substantial blurring causing substantial uncertainty regarding the presence or absence of a pathology;

1 = non-diagnostic.

Additionally, we rated motion artifacts (including cardiac motion and breathing artifacts) as “none,” “minor” or “major.” Minor motion artifacts were defined as not relevant for diagnostic confidence, whereas major artifacts were defined as substantial, hampering the diagnostic evaluation of the structures of interest.

2.5. Analysis of diagnostic yield and clinical consequences

All MR angiography examinations and patient charts were retrospectively analyzed for diagnostic findings and their consequences for the further management by consensus reading between 2 radiologists (1 fellow, initials A.B., and 1 board-certified attending with sub-specialization in cardiovascular imaging, F.G.M.) and 2 cardiac surgeons (1 fellow: C.N., 1 board-certified attending, P.D.). We also analyzed whether the MR angiography examinations offered information inaccessible to CT angiography. Since CT angiography was not performed at the time of MR angiography, this analysis refers to a hypothetical CT examination (if CT angiography would have been performed instead of MR angiography).

2.6. Analysis of further testing

Chart review was performed to investigate whether patients underwent additional imaging tests within 1 month of the MR examination. Since this was an observational study, we did not have pre-specified criteria for additional imaging. Rather, additional imaging was performed as clinically indicated.

2.7. Statistical analysis

Descriptive statistical analysis was performed with GraphPad Prism 5. Age and BMI are presented as median and range, since normal distribution cannot be assumed. Image quality scores are presented as mean and standard deviation. Image quality scores were compared between time-resolved MRA and ECG-gated static MRA using t test for paired samples. The frequency of motion artifacts was compared using Fisher’s exact test. A two-tailed P value < .05 was considered to indicate statistical significance.

3. Results 3.1. Patient population

The study population consisted of 12 men and 7 women (median age 60 years, range 38 to 79 years, Table 1). The indication for aortic surgery was aneurysm in 11 patients and dissection in 8 patients. Regarding surgical techniques, 12 patients had undergone a composite replacement of aortic valve, root and ascending aorta (Bentall operation), 3 patients a valve sparing replacement of the aortic root and ascending aorta (David operation), 2 patients an isolated replacement of the ascending aorta and aortic valve and 2 patients a supra-coronary replacement of the ascending aorta. At the same operation, the aortic arch was completely replaced in 5 patients and partially replaced in 3 patients with additional stent-graft insertion into the descending aorta using the frozen elephant technique in 4 patients.

Table 1 - Patient population. Age 59 (38–79) yr Sex  Male 12 (63%)  Female 7 (37%) BMI 28.7 (21.0–45.8) kg/m2 Indication for surgery  Aneurysm 11 (58%)  Dissection 8 (42%) Surgical technique  Bentall operation (composite replacement of aortic valve, root and ascending aorta) 12 (63%)  David operation (valve sparing replacement of aortic root and ascending aorta) 3 (16%)  Replacement of ascending aorta and aortic valve 2 (11%)  Supra-coronary replacement of ascending aorta 2 (11%)  Replacement of aortic arch 5 (26%)  Partial replacement of aortic arch 3 (16%)  Addition repair of descending aorta (Frozen elephant technique) 4 (16%) Time between surgery and MR imaging 11 (0–92) months
1.3.1. MR examinations.

The MR angiography was performed a median of 11 months after surgery (range 0–92 months). The median duration of the examination (in-room time) was 25 minutes (range 11–41 minutes). No complications occurred related to the contrast-enhanced MR examinations.

2.3.1. Image quality.

The results of the subjective image quality evaluation are summarized in Table 2. In time-resolved MR angiography, the aortic root showed poor, the more distal segments of the aorta good image quality. In the ECG-gated MR angiography, image quality was rated as sufficient for the aortic root and excellent for all other levels. Subjective image quality was significantly better for ECG-gated static MRA compared to non-ECG-gated time-resolved MRA for the aortic root (P = .001), the ascending aorta (P = .001), the aortic arch (P = .002), and the descending aorta (P = .009). Motion artifacts were rated as minor for 8 patients (42%) in the time-resolved non-gated MR angiography and 5 patients (26%) in the ECG-gated MR angiography. This difference was not statistically significant (P = .495). There were no examinations with major motion artifacts.

Table 2 - Image quality: results of the subjective image quality evaluation. Non ECG-gated time-resolved MRA ECG-gated static MRA P value Motion artifacts  Minor 8 (42%) 5 (26%) .495  Major None None 1.000 Subjective image quality  Aortic root 2.0 ± 0.8 3.1 ± 1.1 .001  Ascending aorta 3.6 ± 0.7 4.5 ± 0.7 .001  Aortic arch 3.5 ± 0.7 4.5 ± 0.7 .002  Supra-aortic vessels 4.1 ± 0.8 4.5 ± 0.6 .303  Descending aorta 3.8 ± 0.9 4.6 ± 0.7 .009

P values <.05 appear bold.

ECG = electrocardiogram, MRA = magnetic resonance angiography.

3.3.1. Findings and clinical consequences.

In 13 cases (68%) the MR angiography showed normal postoperative findings. In 6 patients (32%) there were abnormalities such as progressive diameter of remaining aneurysm/ dissection or suture aneurysms, summarized in Table 3. In 4 patients (21%) MR angiography offered information inaccessible to CT angiography mainly concerning flow dynamics in a persisting false lumen of a dissection (Video file, Supplemental Digital Content, https://links.lww.com/MD/J24). In all 6 patients with abnormal findings there were consequences for clinical management (Table 4). Figures 1 and 2 show 2 examples of abnormalities found in the patients.

Table 3 - Findings at MR angiography (all patients). Finding Number of patients (%) Normal post-operative findings 13 (68%) Any abnormality 6 (32%) - Progressive diameter of remaining aneurysm or dissection 3 (16%) - Suture pseudoaneurysm 3 (16%) Additional information specific to MR imaging (not accessible to CT) 4 (21%) - Visualization of flow dynamics in remaining dissection 4 (21%)

CT = computer tomography, MR = magnetic resonance.

Table 4 - Clinical consequences in patients with abnormal findings (n = 6). Patient Abnormal finding at MR angiography Consequence for management 1 Persisting dissection, visualization of flow dynamics ruled out type I endoleak Ruled out of need for intervention at this time 2 Enlargement of persisting aneurysm
Suture pseudoaneurysm Re-surgery
More frequent follow-up 3 Enlargement of subvalvular suture pseudoaneurysm More frequent follow-up 4 Suture pseudoaneurysm Follow-up with CT 6 months later, then re-surgery 5 Enlargement of dissection aneurysm 2nd-step surgery with replacement of thoraco-abdominal aorta 6 Enlargement of dissection aneurysm More frequent follow up

CT = computer tomography, MR = magnetic resonance.

Figure 1.:

Subvalvular pseudoaneurysm at the suture after Bentall operation (replacement of the aortic valve, root and ascending aorta). CT angiography 3 years after surgery (A–C) shows a pseudoaneurysm at the suture (*). At MR angiography another year later (D–F) the pseudoaneurysm has increased in size. The patient was scheduled for more frequent follow-up examinations (at 6 months). Ao = aorta, CT = computer tomography, LA = left atrium, LV = left ventricle, RA = right atrium, RV = right ventricle.

Figure 2.:

MR angiography 6 months after ascending aortic replacement (A–C) demonstrates a pseudoaneurysm (*) originating from the aortic root. Follow-up CT angiography 6 months later (D–F) shows progression in diameter. The pseudoaneurysm was successfully repaired by surgery (G). Ao = aorta, CT = computer tomography, LA = left atrium, LV = left ventricle, PA = pulmonary artery, RA = right atrium, RV = right ventricle.

4.3.1. Further testing.

One patient in this study population underwent additional imaging with CT within 1 month after the MR examination to plan second-step endovascular graft extension.

4. Discussion

In this study we investigated the feasibility and image quality of ECG-gated contrast enhanced MR angiography in patients after aortic surgery involving the ascending aorta, recorded abnormal postoperative findings and analyzed their consequences for the further clinical management. We demonstrate that the subjective image quality of ECG-gated MR angiography is excellent for the ascending aorta, aortic arch, supra-aortic vessels and descending aorta. Abnormal findings with consequences for further management were observed in 32% of patients. MR angiography provided additional information inaccessible to CT angiography in 21% of patients concerning flow dynamics in dissections.

Postoperative follow-up imaging is necessary after aortic surgery to identify postoperative complications and for surveillance of persisting pathologies. Long-term follow-up is particularly important if the underlying disease is still present like a persisting dissection or aneurysm.[10] After thoracic aortic surgery, follow-up imaging is typically performed in annual or biannual intervals.[10] Cumulative radiation exposure from repeat follow-up CT angiography may lead to a significantly higher lifetime cancer risk in younger patients.[11] Because of concerns about the radiation exposure associated with repeat CT scans, the European Society of Cardiology guidelines recommend for MR angiography to be used more frequently in the future.[10,12]

MR angiography of the aorta is still not widely used in clinical routine. Reasons may be the limited availability of MR scanners as well as technical challenges. Due to the longer acquisition time, MRI is more susceptible to respiratory and motion artifacts, although these can be mitigated with fast imaging techniques and use of ECG- gating. General contraindications for MRI must be considered, especially after cardiovascular surgery. Sternal wires and vascular clips do not present an additional risk and most mechanical and biological heart valve prostheses are safe at both 1.5T and 3T. Nevertheless, MRI is susceptible to magnetic field inhomogeneity and artifacts caused by metallic implants[13] In our cohort of patients after aortic surgery involving the ascending aorta, we found MR angiography to be safe and diagnostic in all patients.

Of note, our study showed that MR angiography offered information that would not have been available to CT angiography including flow dynamics in true and false lumina of persisting dissections (4 patients). These findings also had an impact on the clinical management of these patients. If quantification of regurgitation over valves or quantification of stenosis severity is desired, the possibility of flow measurement using phase-contrast MR is another important advantage of MRI over CT for risk stratification and choice of right therapy.[14] This may be even improved by using newer techniques such as 4D flow imaging giving the opportunity to measure flow in several levels of the aorta simultaneously. 4D flow MRI can also visualize flow patterns in the thoracic aorta to improve our understanding of flow dynamics in aortic pathologies pre- and postoperatively.[15–17] Nevertheless, this technique is not yet widely available and acquisition times can be as long as 15 to 20 minutes depending of the volume of interest.

Several non-contrast techniques for MR angiography have been evaluated for the aorta and found to offer image quality suitable for clinical use.[18–20] These techniques may be of particular relevance considering the ongoing debate about gadolinium retention in the brain, although no adverse clinical consequences have been proven.[21,22]

Based on our initial experience presented in this manuscript, we think that MR angiography is a reasonable alternative for the repeated imaging surveillance after replacement of the ascending aorta to reduce the cumulative radiation exposure particularly in younger patients. Patients suspected to have additional pathologies such as valve stenosis or regurgitation may benefit from additional flow measurements, which are possible in MR but not CT angiography. In patients with a persisting dissection, the information about flow dynamics in the false lumen from dynamic MR angiography can have an impact on their management. Based on the data of our study, we recommend performing both time-resolved MRA and static ECG-gated MRA in patients after surgery involving the ascending aorta. While ECG-gated static MRA provided superior image quality and spatial resolution for anatomical evaluation, time-resolved MRA offers valuable additional information regarding flow dynamics.

The results of our investigation should be interpreted in light of its limitations. This was a retrospective, single-center study with a relatively small sample size. The study design did not allow for direct comparison to CT angiography as the gold standard to evaluate diagnostic accuracy. In the absence of an external reference standard, the accuracy of findings at ECG-gated MRA and time-resolved MRA cannot be proven.

5. Conclusions

In conclusion, ECG-gated MR angiography at 3T is feasible and yields good image quality for post-operative surveillance after aortic surgery involving the ascending aorta. Abnormal findings at MR angiography are common in this patient cohort and have important implications for clinical management. This technique may serve as alternative to CT particularly in younger patients with repeated follow-up.

Author contributions

Conceptualization: Anke Busse, Felix G. Meinel.

Data curation: Anke Busse, Catharina Neßelmann, Felix G. Meinel.

Formal analysis: Anke Busse, Catharina Neßelmann, Felix G. Meinel.

Investigation: Anke Busse, Catharina Neßelmann, Felix Streckenbach, Ebba Beller, Ann-Christin Klemenz, Pascal Dohmen, Alper Öner, Marc-André Weber, Felix G. Meinel.

Methodology: Anke Busse, Catharina Neßelmann, Felix Streckenbach, Ebba Beller, Ann-Christin Klemenz, Pascal Dohmen, Alper Öner, Marc-André Weber, Felix G. Meinel.

Project administration: Felix G. Meinel.

Resources: Pascal Dohmen, Alper Öner, Marc-André Weber, Felix G. Meinel.

Supervision: Pascal Dohmen, Alper Öner, Marc-André Weber, Felix G. Meinel.

Validation: Anke Busse, Catharina Neßelmann, Felix G. Meinel.

Visualization: Anke Busse, Felix G. Meinel.

Writing – original draft: Anke Busse, Felix G. Meinel.

Writing – review & editing: Anke Busse, Catharina Neßelmann, Felix Streckenbach, Ebba Beller, Ann-Christin Klemenz, Pascal Dohmen, Alper Öner, Marc-André Weber, Felix G. Meinel.

References [1]. Evangelista A. Imaging aortic aneurysmal disease. Heart. 2014;100:909–15. [2]. Litmanovich D, Bankier AA, Cantin L, et al. CT and MRI in diseases of the aorta. AJR Am J Roentgenol. 2009;193:928–40. [3]. Prescott-Focht JA, Martinez-Jimenez S, Hurwitz LM, et al. Ascending thoracic aorta: postoperative imaging evaluation. Radiographics. 2013;33:73–85. [4]. Chung JH, Ghoshhajra BB, Rojas CA, et al. CT angiography of the thoracic aorta. Radiol Clin North Am. 2010;48:249–64, vii. [5]. Blanke P, Bulla S, Baumann T, et al. Thoracic aorta: prospective electrocardiographically triggered CT angiography with dual-source CT – feasibility, image quality, and dose reduction. Radiology. 2010;255:207–17. [6]. Roos JE, Willmann JK, Weishaupt D, et al. Thoracic aorta: motion artifact reduction with retrospective and prospective electrocardiography-assisted multi-detector row CT. Radiology. 2002;222:271–7. [7]. Meinel FG, Nance JW, Harris BS, et al. Radiation risks from cardiovascular imaging tests. Circulation. 2014;130:442–5. [8]. Lichtenberger JP, Franco DF, Kim JS, et al. MR imaging of thoracic aortic disease. Top Magn Reson Imaging. 2018;27:95–102. [9]. Sabel BO, Buric K, Karara N, et al. High-pitch CT pulmonary angiography in third generation dual-source CT: image quality in an unselected patient population. PLoS One. 2016;11:e0146949. [10]. Erbel R, Aboyans V, Boileau C, et al.; ESC Committee for Practice Guidelines. 2014 ESC Guidelines on the diagnosis and treatment of aortic diseases: document covering acute and chronic aortic diseases of the thoracic and abdominal aorta of the adult. The Task Force for the Diagnosis and Treatment of Aortic Diseases of the European Society of Cardiology (ESC). Eur Heart J. 2014;35:2873–926. [11]. Nyheim T, Staxrud LE, Jørgensen JJ, et al. Radiation exposure in patients treated with endovascular aneurysm repair: what is the risk of cancer, and can we justify treating younger patients? Acta Radiol. 2017;58:323–30. [12]. Grabenwöger M, Alfonso F, Bachet J, et al. Thoracic Endovascular Aortic Repair (TEVAR) for the treatment of aortic diseases: a position statement from the European Association for Cardio-Thoracic Surgery (EACTS) and the European Society of Cardiology (ESC), in collaboration with the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Eur Heart J. 2012;33:1558–63. [13]. von Knobelsdorff-Brenkenhoff F, Trauzeddel RF, Schulz-Menger J. Cardiovascular magnetic resonance in adults with previous cardiovascular surgery. Eur Heart J Cardiovasc Imaging. 2014;15:235–48. [14]. Albayati MA, Clough RE. Optimising pre- and postoperative imaging for thoracic aortic pathology. J Cardiovasc Surg (Torino). 2014;55:529–41. [15]. Hope TA, Herfkens RJ. Imaging of the thoracic aorta with time-resolved three-dimensional phase-contrast MRI: a review. Semin Thorac Cardiovasc Surg. 2008;20:358–64. [16]. Markl M, Frydrychowicz A, Kozerke S, et al. 4D flow MRI. J Magn Reson Imaging. 2012;36:1015–36. [17]. Clough RE, Waltham M, Giese D, et al. A new imaging method for assessment of aortic dissection using four-dimensional phase contrast magnetic resonance imaging. J Vasc Surg. 2012;55:914–23. [18]. van Kesteren F, Elattar MA, van Lienden KP, et al. Non-contrast enhanced navigator-gated balanced steady state free precession magnetic resonance angiography as a preferred magnetic resonance technique for assessment of the thoracic aorta. Clin Radiol. 2017;72:695.e1–6. [19]. Zhu C, Haraldsson H, Kallianos K, et al. Gated thoracic magnetic resonance angiography at 3T: noncontrast versus blood pool contrast. Int J Cardiovasc Imaging. 2018;34:475–83. [20]. Kawel N, Jhooti P, Dashti D, et al. MR-imaging of the thoracic aorta: 3D-ECG- and respiratory-gated bSSFP imaging using the CLAWS algorithm versus contrast-enhanced 3D-MRA. Eur J Radiol. 2012;81:239–43. [21]. Tedeschi E, Caranci F, Giordano F, et al. Gadolinium retention in the body: what we know and what we can do. Radiol Med. 2017;122:589–600. [22]. Levine D, McDonald RJ, Kressel HY. Gadolinium retention after contrast-enhanced MRI. JAMA. 2018;320:1853–4.