Warfarin-associated Intraspinal Hematoma

Abstract

Intracerebral hemorrhage is a well-known complication resulting from warfarin use. By contrast, warfarin-associated intraspinal hematoma is very rare. The intraspinal hematoma may exhibit delayed progression and patients may present with atypical symptoms, occasionally resulting in diagnostic delay. A 65-year-old man with non-valvular atrial fibrillation, arterial hypertension and benign prostatic hyperplasia who used warfarin for prevention of systemic embolism, visited our emergency department (ED) with a complaint of acute urinary retention. He was initially diagnosed with worsening of the prostate hyperplasia. After two days, he revisited the ED with a complaint of painless paraparesis. Magnetic resonance imaging of the thoracic spine revealed an intraspinal hematoma at Th 7–8 levels, and blood coagulation test revealed that prothrombin time-international normalized ratio was 3.33. Despite attempts to reverse the effect of warfarin with vitamin K administration, his paraparesis progressed to paraplegia, necessitating urgent surgical removal of the hematoma. There was partial recovery of motor function after surgery. We learned from the present case that intraspinal hematoma needs to be included in the differential diagnosis of patients using warfarin if they present with acute urinary retention. Although there are no evidence-based treatment guidelines for warfarin-associated intraspinal hematoma, surgical treatment may be warranted for those who exhibit neurological deterioration.

Introduction

Warfarin is an oral anti-coagulant which is widely used for prevention of systemic embolism including stroke. Although intracerebral hemorrhage (ICH) is a well-known complication resulting from warfarin use, warfarin-associated intraspinal hematoma has rarely been reported in the literature.¹ Warfarin-associated ICHs are infamous for their delayed expansion,^2,3,4 and patients with warfarin-associated intraspinal hematoma may also present with atypical symptoms associated with delayed expansion of the bleeding, resulting in diagnostic delay.¹ Herein, we report a case of warfarin-associated intraspinal hematoma in which the diagnosis was delayed because of a comorbidity.

Case Report

A 65-year-old man with non-valvular atrial fibrillation (AF), arterial hypertension and benign prostatic hyperplasia (BPH) visited our emergency department (ED) with a complaint of acute urinary retention. His medications included warfarin (5.5 mg/day for more than 10 years) for AF, and imidafenacin for BPH. He had nocturnal polyuria due to BPH, but had never experienced acute urinary retention. He was examined by a urologist on-call: he underwent the insertion of an indwelling urinary catheter, from which more than 1000 mL of cloudy dark-color urine with pH of 6.5 was evacuated. After the procedure, he was discharged home without undergoing thourough neurological examination. Two days after his first visit, he revisited the ED owing to a complaint of numbness and weakness of the lower extremities. Neurological examination revealed mild paraparesis rated as 3/5 on the manual muscle testing scale as well as diminished position and vibration sense below the umbilicus, and hyporeflexia of the knee and ankle joints. Blood coagulation test revealed a prothrombin time-international normalized ratio (PT-INR) of 3.33. Brain computed tomography, performed to rule out warfarin-associated ICH, yielded negative results. However, magnetic resonance imaging (MRI) of the thoracic spine revealed an intra-medullary mass at Th 7–8 level. The mass was depicted as an iso-intensity on T1-weighted image (Fig. 1A) and a mixed-intensity on T2-weighted image (Fig. 1B). Moreover, a highly-intense signal extending in both caudal and rostral directions, which was considered to be perihematomal edema, was noted on T2-weighted image (Fig. 1B). The patient was diagnosed with a warfarin-associated intraspinal hematoma and was admitted to our neurointensive care unit. Despite the administration of an IV bolus of vitamin K (10 mg) to reverse the effect of warfarin and methylprednisone (500 mg) to attenuate spinal edema, his paraparesis progressed to paraplegia within 6 h of admission. Sensation to painful stimuli was partially diminished. He was taken to the operating theatre for the removal of the hematoma. PT-INR decreased to 1.35 immediately before surgery. After performing laminectomy at Th 7–8 levels and subsequent midline myelotomy, the hematoma was identified and was completely removed using an operative microscope (Fig. 2). Under microscopic view, vascular anomaly was considered unlikely as a source of the bleeding. Due to small amount of surgical specimen, however, the results of pathological examination were inconclusive. Postoperative course was uneventful, and warfarin was replaced with dabigatran (150 mg bid) which was started after 7 days of surgery. He was transferred to a rehabilitation facility after 2 months. Follow-up MRI at 6 months showed resolution of the hematoma and edema and subsequent cavity formation, without apparent contrast enhancement (Fig. 3A-C). At 1 year, he was able to walk using four-footed canes, and urinary retention was managed by intermittent self-catheterization. A permission for publication was granted from the patient.

Fig. 1

Magnetic resonance imaging of the thoracic spine showing an intra-medullary mass at the Th 7-8 level. The mass was depicted as isointensity on T1-weighted image (A) and mixed intensity on T2-weighted image (B). Extensive perihematomal edema was also noted on T2-weighted image (B).

Fig. 2

Intraoperative photograph after Th 7-8 laminectomy and midline myelotomy showing an intraspinal hemorrhage (asterisk).

Fig. 3

MRI performed six months after surgery. There was no apparent contrast enhancement on T1-weighted images (A, without contrast; B, with contrast). Shrinkage of the hematoma and cavity formation was seen on T2-weighted image (C)..

Discussion

Until recently, warfarin was the only oral anti-coagulant approved for prevention of systemic embolism in patients with AF. Despite its efficacy, warfarin has an inherent risk of bleeding: ICH is one of the most feared complications resulting from warfarin use. In a recent study on 404 cases of non-traumatic ICH, 69 (17%) were causally associated with warfarin.² In contrast, warfarin-associated intraspinal hematoma is rare: to date, less than 20 cases have been reported in the literature.^{5,6,7,8,9,10,11,12,13,14,15,16,17} Because of its rarity, it is usually difficult to identify warfarin users who are at risk of intraspinal hematoma. Although advanced age and high PT-INR value (>3.0) are known risk factors for warfarin-associated ICH,³ these may not necessarily be associated with intraspinal hematoma. Similar to warfarin-associated ICH, warfarin-associated intraspinal hematoma may be slowly progressive. Ideally, early detection and intervention, including the administration of vitamin K and fresh frozen plasma to reverse the effect of warfarin, may minimize the hematoma expansion and subsequent neurologic sequelae. In reality, however, diagnosis has often been delayed with only limited neurological recovery.¹ Typically, patients with warfarin-associated intraspinal hematoma present with a triad of back or neck pain, progressive para- or quadriparesis, and acute urinary retention.⁵ Diagnosis was more likely to be delayed in patients who do not present with back or neck pain.⁵ In the present case, pre-existing BPH was obviously responsible for the diagnostic delay. The evolution of symptoms, i.e., acute urinary retention followed by numbness and paraparesis, suggested that the hematoma originated around the spinal canal, and that it gradually spread outward. Although there were no sequential imaging studies of the spinal cord to provide evidence for delayed growth of the hematoma, presence of mixed intensity on T2-weighted images (Fig. 1B) suggests heterogeneity of the composition and age within the hematoma.

We learned from the present case that intraspinal hematoma needs to be included in the differential diagnosis of anti-coagulant users who visit the ED with a complaint of acute urinary retention. Because of its rarity, there have been no evidence-based guidelines for the treatment of warfarin-associated intraspinal hematoma. Conservative management may be justified in patients with mild symptoms without progression¹⁶: however, surgical treatment may be warranted for those with neurological deterioration. In the present case, postoperative neurological improvement may have been because of both neural decompression owing to hematoma removal and resolution of perihematomal edema. After the hemorrhagic event, warfarin was replaced with dabigatran, a novel oral anti-coagulant approved for use in Japan in 2011.¹⁸ Although hemorrhagic complications are less common in patients using dabigatran than those using warfarin,^18,19 there is a lack of data on the frequency of dabigatran-associated hemorrhage in patients who have already sustained warfarin-associated hemorrhage, and careful follow-up is warranted.

Regarding the cause of bleeding, we assumed that underlying arterial hypertension may have caused disruption of an arteriole within the spinal cord, and coagulation disturbances associated with warfarin use may have prompted hematoma expansion. We need to be careful about the possibility of intramedullary tumor as the cause of bleeding; there was no contrast enhancement on postoperative MRI (Fig. 3AB), and high-grade malignancy such as glioblastoma was unlikely. However, periodical surveillance MRI will be required to rule out the presence of low-grade malignancy such as cystic ependymas.²⁰

Conflict of Interest

The authors declare that there is no conflict of interest regarding this manuscript.

References

• 1.  Leep Hunderfund AN, Wijdicks EF: Intramedullary spinal cord hemorrhage (hematomyelia). Rev Neurol Dis 2009; 6: E54–E61.
• 2.  Ma M, Meretoja A, Churilov L, Sharma GJ, Christensen S, Liu X, Weir L, Davis SM, Yan B: Warfarin-associated intracerebral hemorrhage: volume, anticoagulation intensity and location. J Neurol Sci 2013; 332: 75–79.
• 3.  Fang MC, Chang Y, Hylek EM, Rosand J, Greenberg SM, Go AS, Singer DE: Advanced age, anticoagulation intensity, and risk for intracranial hemorrhage among patients taking warfarin for atrial fibrillation. Ann Intern Med 2004; 141: 745–752.
• 4.  Inamasu J, Nakatsukasa M, Miyatake S, Hirose Y: Influence of warfarin and low-dose aspirin on the outcomes of geriatric patients with traumatic intracranial hemorrhage resulting from ground-level fall. Geriatr Gerontol Int 2012; 12: 667–672.
• 5.  Famularo G, Sajeva MR, Gasbarrone L: Warfarin-associated hematomyelia. Intern Med 2014; 53: 623–626.
• 6.  Constantini S, Ashkenazi E, Shoshan Y, Israel Z, Umansky F: Thoracic hematomyelia secondary to coumadin anticoagulant therapy: a case report. Eur Neurol 1992; 32: 109–111.
• 7.  Rosier MP, Neau JP, Vandermarcq P, Agbo C, Gil R: [Hematomyelia under the effect of anticoagulants]. Presse Med 1993; 22: 1747.
• 8.  Liebeschuetz S, Baddeley H, Thomas DJ: Haematomyelia following warfarin therapy. Br J Clin Pract 1994; 48: 105–106.
• 9.  Pullarkat VA, Kalapura T, Pincus M, Baskharoun R: Intraspinal hemorrhage complicating oral anticoagulant therapy: an unusual case of cervical hematomyelia and a review of the literature. Arch Intern Med 2000; 160: 237–240.
• 10.  Pisani R, Carta F, Guiducci G, Silvestro C, Davini MD: Hematomyelia during anticoagulant therapy. Surg Neurol 1985; 24: 578–580.
• 11.  Cakirer S, Basak M, Galip GM: Cervical hematomyelia secondary to oral anticoagulant therapy: case report. Neuroradiology 2001; 43: 1087–1088.
• 12.  Dahlin PA, George J: Intraspinal hematoma as a complication of anticoagulant therapy. Clin Pharm 1984; 3: 656–661.
• 13.  Murphy MA, Nye DH: Thoracic intramedullary haematoma as a complication of warfarin: case report and literature review. Aust N Z J Surg 1991; 61: 789–792.
• 14.  Zeidman SM, Olivi A: Cervical intramedullary hemorrhage as a result of anticoagulant therapy. J Spinal Disord 1993; 6: 456–457, discussion 457–458.
• 15.  Brandt M: Spontaneous intramedullary haematoma as a complication of anticoagulant therapy. Acta Neurochir (Wien) 1980; 52: 73–77.
• 16.  Suzuki M, Ando T, Kawakami O, Sugiura M, Kato H, Inagaki T: [A case of atraumatic hematomyelia extending from the C1 to T11 segments of the spinal cord]. Rinsho Shinkeigaku 2013; 53: 536–542.
• 17.  Robbins M, Verghese J: Acute painless progressive quadriplegia associated with warfarin use. J Am Geriatr Soc 2007; 55: 799–800.
• 18.  Ogawa S, Koretsune Y, Yasaka M, Aizawa Y, Atarashi H, Inoue H, Kamakura S, Kumagai K, Mitamura H, Okumura K, Sugi K, Yamashita T: Antithrombotic therapy in atrial fibrillation: evaluation and positioning of new oral anticoagulant agents. Circ J 2011; 75: 1539–1547.
• 19.  Tanoue S, Inamasu J, Yamada M, Toyama H, Hirose Y: Does dabigatran increase the risk of delayed hematoma expansion in a rat model of collagenase-induced intracerebral hemorrhage? J Stroke Cerebrovasc Dis 2015; 24: 374–380.
• 20.  Tait MJ, Chelvarajah R, Garvan N, Bavetta S: Spontaneous hemorrhage of a spinal ependymoma: a rare cause of acute cauda equina syndrome: a case report. Spine 2004; 29: E502–E505.