Application of multi-planar reconstruction technique in endovascular repair of aortic dissection

Abstract

BACKGROUND

Endovascular repair of aortic dissection is an effective method commonly used in the treatment of Stanford type B aortic dissection. Stent placement during the operation was one-time and could not be repeatedly adjusted during the operation. Therefore, it is of great significance for cardiovascular physicians to fully understand the branch status, position, angle, and other information regarding aortic arch dissection before surgery.

AIM

To provide more references for clinical cardiovascular physicians to develop treatment plans.

METHODS

Data from 153 patients who underwent endovascular repair of aortic dissection at our hospital between January 2021 and December 2022 were retrospectively collected. All patients underwent multi-slice spiral computed tomography angiography. Based on distinct post-image processing techniques, the patients were categorized into three groups: Multiplanar reconstruction (MPR) (n = 55), volume reconstruction (VR) (n = 46), and maximum intensity projection (MIP) (n = 52). The detection rate of aortic rupture, accuracy of the DeBakey classification, rotation, and tilt angles of the C-arm during the procedure, dispersion after stent release, and the incidence of late complications were recorded and compared.

RESULTS

The detection rates of interlayer rupture in the MPR and VR groups were significantly higher than that in the MIP group (P < 0.05). The detection rates of DeBakey subtypes I, II, and III in the MPR group were higher than those in the MIP group, and the detection rate of type III in the MPR group was significantly higher than that in the VR group (P < 0.05). There was no statistically significant difference in the detection rates of types I and II compared to the VR group (P > 0.05). The scatter rate of markers and the incidence of complications in the MPR group were significantly lower than those in the VR and MIP groups (P < 0.05).

CONCLUSION

The application of MPR in the endovascular repair of aortic dissection has improved the detection rate of dissection rupture, the accuracy of anatomical classification, and safety.

Key Words: Multiplanar reconstruction, Endovascular repair of aortic dissection, Image-processing technology, Rate of aortic rupture, Volume reconstruction

Core Tip: Stent placement for the endovascular repair of aortic dissection is performed once and cannot be adjusted repeatedly during surgery. Therefore, it is necessary to fully understand the branch status, position, angle, and other information of aortic arch dissection before the operation. Tulti planar reconstruction is a technique that can achieve omnidirectional rotational imaging by adjusting the slice thickness and can help doctors observe the details of blood vessels.

INTRODUCTION

Endovascular repair is a highly effective method that is frequently employed in the treatment of Stanford type B aortic dissection. However, one-time stent placement is limited by the inability to repeat adjustments during the procedure. Consequently, a comprehensive understanding of the status, position, and angle of the aortic arch is necessary before operation[1]. Although computed tomography (CT) angiography is considered the gold standard for diagnosing aortic dissection, its invasive nature poses challenges such as low acceptance, limited repeatability, and high costs. Consequently, the clinical applications of this procedure are restricted[2]. In addition, as a non-invasive and widely adopted imaging modality, CT angiography is pivotal for the diagnosis and postoperative evaluation of cardiovascular and cerebrovascular diseases. The high sensitivity and specificity of this tool for diagnosing vascular lesions can be attributed to its powerful post-processing functions[3,4]. Multi planar reconstruction (MPR) is a technique that facilitates omnidirectional rotational imaging by adjusting the layer thickness and enhancing the observation of vascular details[5]. In this study, we applied MPR, volume reconstruction (VR), and maximum intensity projection (MIP) to the endovascular repair of aortic dissection. A comparative analysis of these three techniques was performed to provide valuable insights for cardiovascular physicians when formulating treatment plans.

MATERIALS AND METHODS

Design, setting, and participants

First, this was a retrospective study. A total of 153 patients who underwent endovascular repair of aortic dissection in our hospital between January 2021 and December 2022 were selected as research subjects. The inclusion criteria encompassed the following: (1) Patients with a diagnosis of Stanford type B aortic dissection based on imaging or intraoperative diagnosis, following the 2017 Chinese expert consensus of standardized diagnosis and treatment for aortic dissection[6]; (2) Patients in the acute or subacute stage, meeting the indications for endovascular repair of aortic dissection; (3) Those aged ≥ 18 years; and (4) Patients who provided their informed consent. The exclusion criteria were as follows: (1) Undergoing surgery in the chronic phase; (2) Absence of standard conservative treatment during the perioperative period; and (3) Incomplete clinical or imaging data. All patients underwent multi-slice spiral CT angiography and were categorized into three groups based on different post-image processing techniques: MPR (n = 55, multi-planar reconstruction technique), VR (n = 46, VR), and MIP (n = 52, maximum density projection).

Endovascular repair of aortic dissection

All patients underwent preoperative CT angiography using a Siemens Somatom Definitin Flash spiral CT scanner. The scanning parameters were set as follows: Voltage, 120 kV; current, 210 mA; field of view, 360 mm; layer thickness, 0.6 mm, reconstruction layer thickness, 1 mm; and layer interval, 0.6 mm. Patients were instructed to hold their breath during the scan, which covered the range from the top of the aortic arch to the bottom of the pelvic cavity. For enhanced scans, the contrast media-tracking technique was employed, with an injection of 60–80 mL of 370 mg/mL iohexol under high pressure. The scanning process was triggered by artificial intelligence, and when the CT threshold of the region of interest reached 100 HU, the scan was delayed by 6 s. Subsequently, the original image data were transmitted to the workstation for image reconstruction, including the MPR, VR, and MIP techniques, for detailed observation of the vascular structure. During the operation, the patients were maintained in the supine position, with the C-arm detector positioned in front and the ball tube of the digital subtraction angiography (DSA) placed at the rear. Aortic DSA angiography was performed to visualize the rupture of the aortic dissection and to distinguish between the true and false lumens. The surgeon evaluated the true lumen of the aorta and determined the need for rotation and tilt according to the results obtained from the preoperative CT MPR, VR, and MIP reconstruction images. Subsequently, the stent was removed. Following release, angiography was repeated to assess its impact on aortic dissection. The femoral artery was then sutured, punctured, and managed to control bleeding, followed by bandaging.

Data collection

The electronic clinical records of our institution were reviewed by one investigator, who abstracted the data for each time point, as described previously. The information collected included patient demographics and pre-existing comorbidities, clinical presentation, laboratory findings, imaging findings, microbiological investigations, treatment, and outcomes; the detection rate of dissection breach, the accuracy rate of DeBakey classification, the rotation and tilt angle of the C-arm obtained in each group were counted, and the dispersion after stent release and late complications were recorded and analyzed.

Main observation indexes

Aortic rupture detection: Using DSA examination results as the gold standard[7], we analyzed different post-processing techniques for the detection of aortic rupture.

Anatomical classification: The DeBakey classification was used to observe the corresponding reconstructed images[8]. Type I involved an intimal tear at the proximal end of the ascending aorta, with the lesion extending to the abdominal aorta; Type II exhibited rupture at the proximal end of the ascending aorta, involving only the ascending aorta; and Type III was characterized by a rupture in the buccal area of the descending aorta, with the lesion extending to either the descending or abdominal aorta.

Rotation and tilt angle of the C-arm: The rotation and tilt angles in each group were determined according to specific reconstruction methods. The sagittal section line was laterally shifted on the cross-sectional image, with a focus on the apex of the aortic arch. To determine the rotational angle, the coronal section was manipulated until the maximum coronal plane was observed at the apex. Subsequently, the tilt in the coronal plane was fine-tuned until alignment occurred between the root of the essential branch artery and the midpoint of the cross-sectional line, along with the aortic wall. The sagittal plane tilt angle was determined by placing the root of the branch artery at its apex.

Scaffold dispersion: The evaluation of scaffold dispersion involved evaluating the alignment of the proximal marker points of the stent and determining whether they formed a straight line or an oval shape.

Complications: Post-stent placement complications, including stent displacement, internal leakage, artery dilatation, and aortic rupture, were recorded.

Statistical analysis

The SPSS software package (version 23.0) was used to analyze the data. Measurement data conforming to the normal distribution were expressed as (mean ± SD), and the independent sample t-test was used for comparisons between groups. Count data were expressed as percentages (%), and the four-fold table χ² test was used to perform the comparison among groups. Statistical significance was set at a P-value of < 0.05.

RESULTS

Baseline characteristics

In the MPR group, there were 30 males and 25 females, with an age range of 33 years-79 years (mean: 52.75 ± 12.82). Of the 55 patients in the MPR group, 20 were classified as type I according to the DeBakey classification, 11 as type II, and 24 as type III. The VR group consisted of 25 males and 21 females, with an age range of 32 years-78 years (mean: 51.83 ± 12.29), of whom 16 were classified under type I DeBakey classification, 10 under type II, and 20 under type III. Lastly, the MIP group included 30 males and 22 females, with the age range of 33 years-80 years (mean: 52.08 ± 12.13), and according to DeBakey classification, 17 were under type I, 13 were under type II, and 22 were under type III. No significant differences in sex, DeBakey classification, age, or other general data were observed among the three groups (P > 0.05).

Outcomes

The MIP group exhibited an aortic rupture detection rate of 0 rupture, whereas higher detection rates were observed in the MPR and VR groups. In particular, the detection rate in the MPR group was higher than that in the VR group (P < 0.05) (Table 1).

Table 1 Comparison of detection rate of dissection crevasse in each group.

Groups	MPR group	VR group	MIP group	χ2	P value
Number of aortic rupture cases1	40	24	0	-	-
Proportion (%)1	72.73	52.17	0	60.994	0.000

¹Data presented in frequency and percentage.

MPR: Multi-planar reconstruction; VR: Volume reconstruction; MIP: Maximum intensity projection.

Compared with the MIP group, the MPR group exhibited elevated detection rates for DeBakey types I, II, and III blood vessels. Conversely, when compared with the VR group, only the detection rate of DeBakey type III blood vessels in the MPR group demonstrated a significant increase (P < 0.05) (Table 2).

Table 2 Comparison of detection rate of DeBakey typing in each group.

Groups	MPR group (n = 55; 36%)	VR group (n = 46; 30%)	MIP group (n = 52; 34%)	χ2	P value
Type Ⅰ1	19/20	14/16	0/17	41.509	0.000
Type Ⅱ1	4/11	3/10	0/13	7.937	0.019
Type Ⅲ1	23/24	13/20	0/22	43.783	0.000

¹Data presented in frequency and percentage.

MPR: Multi-planar reconstruction; VR: Volume reconstruction; MIP: Maximum intensity projection.

The MPR group demonstrated significantly higher C-arm rotation and tilt angles than the MIP and VR groups (P < 0.05) (Table 3).

Table 3 Comparison of C-arm rotation and tilt angle in each group.

Groups	MPR group (n = 55; 36%)	VR group (n = 46; 30%)	MIP group (n = 52; 34%)	F	P value
Rotation angle (°)1	57.26 ± 6.18	52.72 ± 5.09	49.30 ± 3.15	34.197	0.000
Tilt angle (°)1	-2.09 ± 7.18	-0.16 ± 0.93	-0.13 ± 0.87	3.495	0.033

¹Data presented in mean ± SD.

MPR: Multi-planar reconstruction; VR: Volume reconstruction; MIP: Maximum intensity projection.

In contrast to the VR and MIP groups, the MPR group exhibited a lower rate of marker point dispersion. Furthermore, the MPR group demonstrated a lower probability of overall complications, such as stent displacement, endoleak, arterial dilatation, and aortic rupture, than the MIP group (P < 0.05) (Table 4).

Table 4 Comparison of marker point dispersion rate and complication rate in each group, n (%).

Groups		MPR group (n = 55; 36%)	VR group (n = 46; 30%)	MIP group (n = 52; 34%)	χ2	P value
Marker point dispersion rate		2 (3.64)	8 (17.39)	14 (26.92)	11.104	0.004
Complication	Stent displacement	1 (1.82)	1 (2.17)	3 (5.77)	-	-
	Internal leakage	1 (1.82)	2 (4.35)	5 (9.62)
	Artery dilatation	0	1 (2.17)	2 (3.85)
	Aortic rupture	0	0	1 (1.92)
	Overall complications	2 (3.64)	4 (8.70)	11 (21.15)	8.693	0.013

MPR: Multi-planar reconstruction; VR: Volume reconstruction; MIP: Maximum intensity projection.

DISCUSSION

Aortic dissection is a relatively significant cardiovascular disease that relies on imaging examinations for disease diagnosis and prognostic evaluation[9]. Among the techniques used for this purpose, computed tomographic angiography is typically employed. This method reveals typical manifestations, such as the intimal film of the lumen and the distinction between true and false lumens. Conversely, conventional plain scans exhibit a low diagnostic efficacy for aortic dissection[10]. Studies have indicated that transverse axial images obtained through contrast-enhanced CT provide clear visualization of intimal displacement and the presence of double cavities. The identification of a continuous low-density shadow between these cavities, whether straight or curved, is of diagnostic importance. The detection of a discontinuous linear low-density shadow at a certain level or contrast medium leakage within the lumen signified aortic rupture[11,12]. Owing to the occurrence of intimal tears, blood extends into the aortic wall, giving rise to a dissecting hematoma that can expand longitudinally along the blood vessels, encompassing the entire aorta and associated organ arteries. Therefore, precise knowledge of the location of aortic tears is important for clinical cardiovascular physicians and serves as a crucial foundation for formulating treatment plans[13].

The findings of this study revealed that the MPR group exhibited a significantly higher aortic rupture detection rate than the MIP and VR groups, suggesting the effectiveness of MPR in detecting this critical condition. In addition, MPR is a technical method for reconstructing two-dimensional images by presenting volume data in the coronal and sagittal positions, offering a simplified and faster alternative to other 3D imaging techniques[14]. This method can obtain two-dimensional image of any plane. In practice, this method allows the display of vascular morphology and anatomical structures from multiple planes by adjusting the layer thickness to a minimum. Several studies have demonstrated the efficacy of MPR technology in observing aortic dissections, intimal flaps, true and false lumens, and intraluminal thrombus[15]. Scholars have adjusted the layer thickness to less than 1 mm and employed omnidirectional rotation imaging, resulting in significantly superior outcomes compared to curved planar reconstruction, VR, and MIP in visualizing intimal flaps and ruptures. MIP is a technical method that re-projects the maximum CT value pixel onto a projection ray, which can accurately reflect the difference in vascular density and display vascular shapes. Although it has a good clinical application value for lumen stenosis and wall calcification, its spatial resolution for overlapping three-dimensional blood vessels is limited[16]. VR is a technical method with a three-dimensional sense and rich display structure established by different transparency and color coding; however, these results can be attributed to the fact that MIP can retain more CT values of the original image; thus, its density resolution is consistent with that of the axial image[17]. In MIP, the maximum density pixel is displayed; hence, it is unable to depict areas with a lower density of the intima or those with little density difference between the true and false cavities. Moreover, VR struggles with ruptured displays due to threshold adjustment limitations. In addition, aortic classification is crucial for guiding clinical plans, considering variations in lesions and branch involvement and influencing different treatment strategies[18]. The findings of this study demonstrate that the detection rates of DeBakey types I, II, and III in the MPR group were significantly higher than those in the MIP group, with a higher detection rate for type III blood vessels compared to than in the VR group. This indicates that the detection rate of MPR for anatomical classification was better than that of MIP and VR, in agreement with previous studies[19]. Moreover, this study identified significant differences in the C-arm rotation and tilt angles obtained in the MPR group compared with the MIP and VR groups. The aortic arch is a three-dimensional structure, and MPR allows automatic updates of cross-sectional views, facilitating simultaneous observation from three planes. Both the DSA and C-arm were adjusted to align with the level of the corresponding thoracic vertebra on the cross-section. Subsequently, the longitudinal positioning of the aortic arch was performed using this as the central reference point. The C-arm rotated in both the anterior and posterior directions, facilitating comprehensive expansion in the left and right directions. Simultaneously, the detector is tilted along the left and right directions to provide complete exposure in the anterior and posterior directions. This adjustment was aimed at optimizing effective positioning during the operation[20]. This study further discovered that the MPR group exhibited lower rates of marker point dispersion and complications than the VR and MIP groups, suggesting that the application of MPR technology contributes to enhanced positioning accuracy. Furthermore, this precision may also be a reason for the observed lower incidence of complications. Because of the geographical limitations of the patients’ residence, most of this study followed the principle of nearest treatment and did not involve hospitals of different regions or levels, resulting in certain selection biases. Additionally, the sample size of this study was only 153 cases, and it is advisable to obtain sufficient samples for in-depth and detailed analyses.

CONCLUSION

In summary, the application of MPR in the endovascular repair of aortic dissection demonstrated an improvement in the detection rate of aortic rupture and the accuracy of anatomical classification. The following is an alternative phrasing to this part “In addition, the integration of supplementary technologies, similar to VR and MIP, contributes to achieving more satisfactory imaging results. This, in turn, has the potential to enhance patient outcomes and improve overall healthcare.