2025 年 19 巻 1 号 論文ID: cr.2024-0122
Objective: Presurgical identification of the artery of Adamkiewicz (AKA) is crucial in thoracic and lumbar spinal lesions. Although CTA is widely employed to identify AKA, the detection rate varies across reports and is often difficult, especially in patients with spinal dural arteriovenous fistulas (SDAVFs). We report 2 cases where sublingual administration of nitroglycerin (NTG), a vasodilator, before CTA enabled good visualization of the AKA in patients with SDAVF.
Case Presentation: (Case 1) A patient in his 60s presented with progressive gait disturbance, and an MRI revealed a spinal vascular lesion. CTA with sublingual NTG administration revealed an SDAVF supplied by the right seventh intercostal artery and an AKA branching from the left ninth intercostal artery. A chronic descending aortic dissection was identified on CTA, but spinal angiography was not performed. The patient was successfully managed with direct surgical disconnection of the SDAVF, and his symptoms improved. (Case 2) A patient in his 60s presented with progressive bilateral lower-limb numbness, and a thoracic SDAVF was diagnosed using MRI. CTA without NTG administration failed to identify the AKA, but after the administration of NTG, a clear depiction of the AKA branching from the left 10th intercostal artery was obtained. Endovascular shunt occlusion was successfully performed via the right 10th intercostal artery, resulting in significant symptom improvement.
Conclusion: Sublingual NTG administration before CTA may improve AKA expression in patients with SDAVF.
The artery of Adamkiewicz (AKA), also referred to as the great anterior radiculomedullary artery, usually supplies the spinal cord at the lower thoracic and lumbar levels.1) The intraprocedural sacrifice of the AKA has a potential risk of spinal cord infarction. Therefore, accurate identification is necessary for surgery or endovascular treatment (EVT) of descending aortic or thoracolumbar spinal cord lesions, including tumors and spinal dural arteriovenous fistulas (SDAVFs). The use of spinal angiography is the gold standard for AKA identification. Although computed CTA and MRA are minimally invasive alternatives, the detection of AKA remains difficult, especially in cases of SDAVFs.2–8) According to previous reports, the detection rate of the AKA using CTA in patients with abdominal aortic aneurysms ranged from 9% to 80%.9,10) Conversely, another study reported a 72.3% detection rate of the AKA in patients with SDAVF.3) However, it is important to note that in the latter study, the AKA was defined as any vessel identified entering the intradural space from a segmental artery. Previous studies have discussed various techniques for a clear depiction of AKA.3,9,11,12)
Sublingual administration of nitroglycerin (NTG) before coronary CTA is reported to facilitate the delineation of small vessels.13) Moreover, the utilization of NTG before CTA or MRA reportedly facilitates AKA delineation.5,10) However, no reports confirm whether NTG administration before CTA improves the visualization of AKA in patients with SDAVF.
In this study, we presented cases of thoracic SDAVF where CTA imaging with NTG administration improved the visualization of AKA and reviewed our experiences with the depiction of AKA with or without NTG. This case report has been approved by the institutional ethics review board of the Institute of Science Tokyo (M2020-102).
Patients who underwent CTA at our facility were administered 0.3 mg of NTG sublingually approximately 5 min prior to the examination. No patients exhibited hypotension before drug administration. Eighty-kV CTA was performed using a 192-section dual-source CT scanner (Somatom Force; Siemens, Erlangen, Germany). The dual-power scan mode was applied with the following parameters: 192 × 0.6 mm detector configuration, 0.5 s per rotation, 0.6 pitch, quality reference 350 mAs (Care Dose 4D; Siemens), and reference 120 kV (CARE kV; Siemens). The scanning was extended from the level of the first thoracic vertebra to that of the fifth sacral vertebra. The imaging method for contrast-enhanced CT is as follows: 135 mL of high-osmolarity iodinated contrast medium (Iomeprol 350; Bracco Japan, Tokyo, Japan) was injected at a rate of 4.5 mL/s, followed by a 50 mL saline chaser via a 20-gauge intravenous catheter placed in an antecubital vein using a double-head power injector. The bolus-tracking method was employed to determine the timing of image acquisition. A circular region of interest was placed in the descending aorta at the level of the first lumbar vertebra. After 10 s, attenuation in the aorta increased by 150 Hounsfield units, and scanning was automatically initiated in the craniocaudal direction. Scanning was performed during a single breath hold. Images were reconstructed with a 0.6-mm slice thickness, a 512 × 512 matrix, a 200-mm display field of view, and kernels Bf36 and Br59 using an iterative reconstruction algorithm (Advanced Modeled Iterative Reconstruction [ADMIRE]; Siemens) in the standard setting. The AKA was identified using a 3D image analysis workstation (Ziostation 2, version 2.3.4.0; Ziosoft, Tokyo, Japan).
The details of the CTA imaging acquisition performed at the previous institution for Case 2 are as follows: Eighty-kV CTA was performed on a 192-section dual-source CT scanner (Aquilion One GS; Canon Medical Systems, Tochigi, Japan). Dual-power scan mode was applied with the following parameters: 80 × 0.5 mm detector configuration, 0.5 s per rotation, 0.813 pitch, and reference 120 kV (CARE kV; Siemens). The imaging method for contrast-enhanced CT is as follows: 100 mL of high-osmolarity iodinated contrast medium (Lopamiron 370; Bayer, Leverkusen, Germany) was injected at a rate of 4.0 mL/s, followed by a 40 mL saline chaser. The bolus-tracking method was the same as the protocol of our institution. Images were reconstructed with a 0.5-mm slice thickness, a 512 × 512 matrix, a 200-mm display field of view (FOV), and kernel Fc13 using an iterative reconstruction algorithm (Adaptive Iterative Dose Reduction 3D; Canon Medical Systems) at the standard setting.
Case 1A patient in his 60s presented to our hospital with a 2-year history of bilateral lower-limb numbness, gait disturbance for the past 3 months, and recent onset of bladder and rectal dysfunction. MRI demonstrated hyperintense signal lesions at the level of Th7–Th12, and multiple vascular flow voids were noted on T2-weighted images (Fig. 1A). The patient had a history of aortic dissection surgery, and preoperative CTA revealed chronic descending aortic dissection. CTA with 0.3 mg NTG sublingual administration revealed SDAVF supplied by the right seventh intercostal artery and AKA branching from the left ninth intercostal artery (Fig. 1B–1E). The feeder of the shunt was presumed to be the left seventh intercostal artery; however, the exact shunt point could not be clearly identified (Fig. 1E). The AKA was not visualized in the volume rendering (VR) or maximum intensity projection (MIP) images of the CTA. The patient underwent resection of the lower half of the spinous process of Th6, laminectomy of Th7 and Th8, and microsurgical disconnection of the SDAVF. Although urinary dysfunction persisted, the symptoms gradually improved. The patient was transferred to a rehabilitation hospital 5 weeks after surgery.
A patient in his 60s presented to another hospital with progressive bilateral lower-limb numbness. MRI revealed a hyperintense signal lesion in the thoracic spinal cord below Th6 on T2-weighted images, with some vascular flow voids observed on constructive interference in steady-state (CISS) images (Fig. 2A and 2B). Although CTA without NTG administration did not depict the AKA, elongated and dilated spinal veins were identified (Fig. 2C). The patient was then referred to our hospital for further treatment. CTA with 0.3 mg NTG sublingual administration clearly depicted AKA branching from the left 10th intercostal artery (Fig. 2D–2F), which was consistent with the findings of spinal angiography (Fig. 2I). A feeder from the right 10th intercostal artery was presumed to be involved in the shunt; however, as in Case 1, the exact shunt point could not be clearly identified (Fig. 2G and 2H). Similar to Case 1, the AKA was not clearly visualized on the VR or MIP images of the CTA. Subsequently, the patient underwent endovascular shunt occlusion via the right 10th intercostal artery. The SDAVF was successfully occluded with N-butyl-2-cyanoacrylate (Fig. 2J and 2K). Numbness in the lower limbs significantly improved postoperatively, and no complications related to embolization were observed. The patient was transferred to a rehabilitation hospital on the eighth day after surgery.
NTG is commonly administered to improve the visualization of the coronary artery and increase its detection rate on CTA.13) However, the vasodilatory effect of NTG extends beyond the coronary vessels, as it is known to affect blood vessels throughout the circulatory system.14) Reports have indicated that the administration of NTG enhances the visualization of AKA in patients with non-SDAVFs, such as aortic aneurysms.10) Conversely, according to a previous report that did not involve the administration of NTG, the AKA in patients with SDAVF was more difficult to visualize on CTA than in patients without SDAVF.3) Eckart Sorte et al. suggested that venous hypertension occurs as a result of shunt reflux from the radicular vein to the perimedullary vein, which impacts the arteriovenous pressure gradient within the spinal cord, making visualization of the AKA difficult.8) Furthermore, the presence of dilated and tortuous shunt vessels makes the visualization of small-caliber AKA challenging. Therefore, we hypothesized that the administration of NTG might lead to clearer visualization of the AKA on CTA in patients with SDAVF.
Our study suggests that NTG administration may help visualize the AKA, even in patients with SDAVF. In particular, in Case 2, the AKA, which could not be recognized on CTA without NTG administration, became recognizable after NTG administration.
Identifying the location of the AKA in patients with thoracic and lumbar spinal lesions including spinal tumors, vertebral body tumors, and descending aortic aneurysms or dissections requiring surgical intervention is important. If the AKA is identified before surgical intervention in patients with the aforementioned conditions, the risk of spinal cord ischemia due to occlusion of the segmental artery originating from the AKA may be reduced. In patients with SDAVF, EVT and direct surgery are viable treatment options. Although EVT has garnered significant attention in recent years, the outcomes of direct surgery also remain favorable. Additionally, neurological complications, such as spinal cord infarction, have been reported to be more frequent in EVT than in direct surgical procedures.15) Preoperative anatomical identification of the AKA and feeding arteries may be useful for selecting relatively safe therapeutic options. Although spinal angiography is the gold standard for SDAVF, preoperative CTA with NTG administration may still be valuable, even in cases where spinal angiography is planned. Visualization of the location of AKA before spinal angiography could optimize the sequence of selective catheterization, potentially reducing procedure time, contrast medium usage, and radiation exposure.
Notably, sublingual NTG administration has potential side effects, including hypotension, headache, and dizziness. Although no adverse events were observed in our cases, previous reports suggest that side effects, when they occur, are typically mild and resolve spontaneously without requiring medical intervention.16) However, given the vasodilatory effects of NTG, careful monitoring during CTA is necessary, especially in older patients and those with cardiovascular conditions.
Future prospective investigations are needed to increase the number of cases and demonstrate the usefulness of CTA with NTG for delineating the AKA. Such studies would further elucidate the role of this technique in the preoperative evaluation of thoracolumbar spinal lesions, including vascular disorders, tumors, and aortic diseases, potentially leading to efficient and safe treatment strategies.
Furthermore, although the effects of NTG on radiculopial artery and posterior spinal artery were not examined in this study, the visualization of these vessels is also essential for determining treatment strategies. Therefore, investigating how NTG affects the visualization rate of small-caliber vessels remains an important issue for future research.
In Case 2, CTA images acquired under different scanning parameters are presented, which constitutes a limitation in terms of direct comparison. At our institution, the decision was made not to perform a second contrast-enhanced CTA without NTG administration on the same patient due to ethical considerations regarding the invasive nature of repeated contrast agent use. For a more rigorous evaluation, it would be necessary to compare NTG-administered and non-administered groups under identical scanning conditions.
Sublingual NTG administration before CTA may improve the AKA depiction in patients with SDAVF.
We would like to thank Editage (www.editage.jp) for English language editing.
Kazutaka Sumita received lecture fees from Japan Stryker. The remaining authors declare that they have no conflicts of interest.
