Successfully Treated Isolated Posterior Spinal Artery Aneurysm Causing Intracranial Subarachnoid Hemorrhage: Case Report

Introduction

Subarachnoid hemorrhage (SAH) of spinal origin is rare. Approximately, only 1% of cases of intracranial SAH are related to spinal pathologies.¹⁾ The most common cause is spinal cord arteriovenous malformations, followed by spinal tumors such as ependymomas and neurinomas.²⁾ Rupture of spinal artery aneurysms is rare and in most cases associated with vascular anomalies that cause hemodynamic stress such as spinal arteriovenous malformations or coarctation of the aorta.³⁾ Connective tissue disease or vasculitis of infectious or autoimmune etiology are thought to be other causes.³⁾ Isolated spinal aneurysms that are not associated with vascular malformations or above-mentioned predisposing factors are extremely rare, and majority of them are located at the anterior spinal artery.⁴⁾ We here report a case of SAH caused by rupture of an isolated posterior spinal artery (PSA) aneurysm that was successfully treated surgically. Because there are few reports of isolated PSA aneurysms, the optimal therapeutic strategy is debatable.

Case Report

An 84-year-old man who was being treated for right thalamic infarction in another hospital presented with restlessness and disorientation of sudden onset. Computed tomography (CT) demonstrated an intracranial SAH (Fig. 1A); however, magnetic resonance angiography (MRA) failed to identify an intracranial aneurysm. Spinal magnetic resonance imaging (MRI) detected a localized subarachnoid hematoma at Th10–11 (Fig. 1B). He was transferred to our hospital on the same day for further examination and treatment.

On admission, he was disorientated and had left hemiplegia and sensory disturbance, which were attributed to the known right thalamic infarction. He had no symptoms or signs of spinal cord or root lesions. Computed tomography angiography of the head showed no aneurysms or vascular malformations. Contrast-enhanced spinal CT revealed an area of enhancement on the left dorsal aspect of spinal cord at Th10–11 (Fig. 2A). Enhanced MRI and MRA 5 days after the onset of symptoms also showed enhancement in the center of the localized hematoma (Fig. 2B). Nine days after onset, digital subtraction angiography with superselective angiography of the left Th12 intercostal artery demonstrated an area of faint enhancement in the venous phase. However, the connections with the surrounding vessels were unclear (Fig. 3).

The imaging findings were thus highly suggestive of SAH caused by spinal artery aneurysm. Dural arteriovenous fistula and cavernomas were among the different diagnosis at this time. To prevent rebleeding, surgery was performed 13 days after the onset.

A Th10–11 laminectomy was performed to approach the lesion. When the dura mater and arachnoid were opened, old yellowish hematoma was exposed on the left dorsal aspect of the spinal cord. Careful removal of the hematoma revealed a small fusiform aneurysm situated on the radiculopial artery, which ran along the dorsal root. Indocyanine green video-angiography failed to visualize the aneurysm, suggesting that the lumen was mostly thrombosed. The parent artery was coagulated and cut both distal and proximal to the aneurysm and the aneurysm resected en bloc (Fig. 4).

The patient's postoperative course was uneventful and rehabilitation was soon commenced. His activities of daily living gradually improved and he was transferred to another hospital for further rehabilitation, finally being discharged in an ambulatory condition 5 months after the onset of symptoms. Histopathological analysis revealed a disrupted elastic lamina in part and slightly organized thrombus in and around the aneurysmal wall.

Discussion

The ventral aspect of spinal cord is fed by the radiculomedullary artery, which is connected to the anterior spinal artery. The dorsal aspect is fed by the radiculopial artery, which has a communication with vasa corona, the latter mainly consisting of a pair of PSAs.⁵⁾ Our patient's SAH was caused by rupture of an isolated radiculopial artery aneurysm. Aneurysms of this artery are often reported as isolated PSA aneurysms. To our knowledge, this is the 21st reported isolated PSA aneurysm (Table 1).

Most of the isolated PSA aneurysms are considered to have been caused by dissection.^6,7) In our case, histopathological examination of the resected specimen failed to reveal a pseudolumen, probably because resection was performed 13 days after the onset of symptoms. However, its fusiform shape and location in a site unrelated to arterial branching were consistent with dissecting aneurysm.

Ruptured isolated PSA aneurysms have been treated by surgery, endovascular embolization, or conservatively. Because PSA aneurysms are located on the dorsolateral aspect of the spinal cord, they can easily be approached surgically. Most reported cases have therefore been managed surgically.^1,2,6–15) Endovascular embolization was performed in four of the reported cases.^1,4,16,17) However, when the parent artery is small, it is difficult to access the lesion. In our case, the parent artery was very small and its connection with the aneurysm was not apparent angiographically. We did not use intraoperative electrophysiological monitoring. PSAs have an abundant pial network and there had been no reports showing neurological deficit attributed to sacrificing the parent artery. However, Takata et al. recently presented a case of isolated PSA aneurysm in which motor evoked potential was deteriorated by temporary clipping of the PSA.¹⁴⁾

Some authors have reported success with conservative management, which has resulted in thrombosis of the aneurysm and spontaneous healing of the dissection.^10,18) Berlis et al.²⁾ reported a case in which spontaneous healing of an aneurysm was found when surgery was performed 26 days after the onset; they concluded that conservative management may be appropriate. In our case, intraoperative indocyanine green video-angiography did not show blood flow in the aneurysm, and pathological examination of the resected aneurysm showed it contained organized thrombus. Therefore, the aneurysm might have been in the process of healing spontaneously. Conversely, Henson et al.¹⁹⁾ reported a case of death that resulted from rebleeding 8 months after the original onset. Thus, it is difficult to distinguish which aneurysms truly require active treatment.

Koçak et al.²⁰⁾ reported a patient who died of rebleeding within 24 hours of onset though surgery had been planned, and emphasized that treatment to prevent rebleeding should be performed as soon as possible. However, therapeutic interventions were delayed in some reports, suggesting difficulty with diagnosis. Similarly, surgical resection was performed 13 days after the onset in our case.

There are few relevant published reports. Furthermore, the publication of case reports may be influenced by a positive outcome bias, especially for cases undergoing surgery or endovascular treatment. Therefore, it is dangerous to draw conclusions about optimal therapy based on a few sporadic case reports. Our 84-year-old patient is the oldest of the 21 reported cases of isolated PSA aneurysm. Nevertheless, he tolerated surgery well and was thereafter able to participate in aggressive rehabilitation without the risk of rebleeding. Our case illustrates both the validity and efficacy of surgical treatment for ruptured isolated PSA aneurysms.