Acta Neurochir (2003) 145: 151–155 DOI 10.1007/s00701-002-1045-z Acta Neurochirurgica Printed in Austria Case Report Nontraumatic spinal subdural haematoma occurring in a postpartum period K. Yamada1, T. Nakahara2, K. Yamamato2, T. Muranaka3, and Y. Ushio1 1 Department of Neurosurgery, Kumamoto University Medical School, Kumamoto, Japan 2 Department of Neurosurgery, Miyazaki Prefectural Nobeoka Hospital, Nobeoka, Japan 3 Department of Radiology, Miyazaki Prefectural Nobeoka Hospital, Nobeoka, Japan Published online February 10, 2003 6 Springer-Verlag 2003 Summary Spontaneous occurrence of spinal subdural haematoma (SSH) is very rare. While many neurological disorders can develop in pregnant patients in relation to haemostatic imbalance, there have been no reports in the literature suggesting that pregnancy and/or childbirth per se could be a single risk factor for acute SSH. A 38-year-old previously healthy woman gave birth to a baby via transvaginal normal delivery. The patient had no history of receiving anticoagulants. Eight days after delivery, she experienced sudden onset severe interscapular back pain, and the next day she developed dysesthesia in the both legs and mild urinary retention. Severe meningeal signs were noted. Lumbar puncture revealed bloody cerebrospinal fluid. Magnetic resonance images revealed SSH in the ventral position spreading from levels Th1 to Th7. The patient underwent conservative treatment after which the symptoms gradually improved. Serial MRI study at 17 and 69 days after onset showed spontaneous regression of the SSH. Spinal angiography did not show any vascular malformations, but simultaneous cerebral angiography revealed a co-existing cerebral aneurysm on the C2 segment of the left internal carotid artery. It is unlikely that the cerebral aneurysm was the origin of the SSH, based on the clinical and radiographic findings. Moreover, we confirmed the unruptured nature of the aneurysm during the clipping procedure at open surgery. We report the unique case of subacute SSH occurring 8 days after childbirth without other known risk factors. The possible etiology and magnetic resonance imaging (MRI) findings in this case, and the current controversy concerning therapy for SSH are discussed. Keywords: Magnetic resonance imaging; postpartum period; pregnancy; spinal subdural hematoma. Introduction Spinal subdural haematomas (SSHs) are rare, and the majority occur in patients with coagulopathy [7] or receiving anticoagulants [24, 25]. They may also follow major or minor trauma [9, 22, 27], spinal surgery [6], cerebrospinal fluid shunting [21], or lumbar puncture and/or spinal anaesthesia [1, 14]. Vascular malformations in the spinal canal can also be the origins of the SSH [8, 15]. Less frequently, SSH occurs after craniotomy [20], and several studies have reported that SSHs were associated with intracranial or spinal tumours [2, 24]. SSH developing in the absence of the above risk factors is so-called ‘‘spontaneous SSH’’, however, the incidence is very low; the literature reveals, so far, less than 30 reports [3, 5, 13, 16, 23]. We report here a unique case of thoracic SSH in a postpartum woman which seemed to occur spontaneously. Case report History A 38-year-old woman gave birth to her fourth healthy baby at 37 weeks and 3 days gestation via transvaginal normal delivery on February 23, 2001. Lumbosacral subarachnoid and epidural anaesthesia were not used. The patient had previously been healthy, and had not been receiving any medication; no anticoagulants. No complicated status was noted in the postpartum period. She did not develop toxicoremia, such as hypertension or systemic edema, during her pregnancy, but a moderate degree of hypertension (150–160/90– 100 mmHg), probably induced by methylergometrine maleate, occurred from 5 to 7 days after delivery. Blood pressure was taken 3 times a day. Symptoms Sudden onset of severe interscapular back pain occurred spreading forwards to anterior chest while she was lying in bed on March 3, 152 2001 (8 days after the delivery). She experienced moderate nuchal pain, but no neurological deficits were detected. Blood pressure was 169/89 mmHg and heart rate was 79/min. No abnormal pathological findings were noted on cardiological examination. Intracranial K. Yamada et al. subarachnoid haemorrhage (SAH) was not observed on computed tomography (CT). Next day (March 4, 2001), she developed an uncomfortable sensation in both legs, which resulted in gait disturbance. Fig. 1. Spinal magnetic resonance images (MRI ) obtained 2 days after the onset (A; T1 sagittal, B; T2 sagittal, C; T1 transverse at Th3 level, and D; T2 transverse at Th3 level). The haematoma spread from the Th1 (B, upper arrow) to Th7 (B, lower arrow) levels and compresses the spinal cord from the left ventral to a right dorsal direction (C and D), exhibiting iso-to hyperintensity on T1 (A and C) and extreme hypointensity on T2-weighted images (B and D). Flow void signals suggestive of vascular abnormalities are not observed. Note ring-like relative hyperintensity indicating the epidural fat (C and D, small arrows) ventral to the haematoma 153 Spinal subdural haematoma after childbirth Fig. 2. Spinal CT scan at Th3 level. High density haematoma compresses the spinal cord in a ventral to dorsal direction Neurological examinations On examination at our clinic, she was alert, but a mild hyperventilation syndrome was noted. Severe nuchal rigidity and Kernig’s sign were also noted. The center of the interscapular pain was located at the Th5 level. Motor weakness was not apparent while lying down, but she was unable to walk. She presented with bilateral dysesthesia from levels Th12 to L5, but could clearly discriminate pain, temperature, touch, vibration and position sense. Deep tendon reflexes were exaggerated at the knees and ankles. Mild urinary retention was noted. Fig. 3. Lateral projection of the left internal carotid angiogram showing unruptured aneurysm on the C2 segment on March 19, 2001 (16 days after the onset) showed that the haematoma had reduced in size exhibiting partial hyperintensity on T2weighted images (Fig. 4). The spinal cord was decompressed spontaneously. Co-existing cerebral aneurysm was clipped on April 3, 2001. Intracranial CSF was clear and not xanthochromic, and no chronic change suggestive of previous SAH was noted. The distal dome of the aneurysm bulged into the subarachnoid space, but most of it was located out of the distal dural ring. The surface of the aneurysm was smooth without blebs indicating it was unruptured. Follow-up Laboratory data Lumbar puncture at L3/4 revealed bloody cerebrospinal fluid (CSF) with an opening pressure of 160 mmH2 O. Quekenstedt’s test was positive. There were 2,464 cells/cu mm and glucose level was 10 mg/dl, but CSF culture did not reveal bacterial infection [23]. Laboratory examinations ruled out disseminated intravascular coagulation (DIC) and other haemostatic abnormalities. Neuroradiological findings Urgent spinal MRI studies were performed and a hypo-intense longitudinal mass on T2-weighted images was detected in the ventral position of the thoracic spinal cord, although the images were poor because of motion artifact. Repeated MRI on March 5 (day 3) clearly showed SSH in the ventral position (Fig. 1) spreading from levels of Th1 (Fig. 1B, upper arrow) to Th7 (Fig. 1B, lower arrow). Flow void signals suggestive of arteriovenous malformations/fistulas (AVMs/AVFs) were not observed. Spinal CT scan also showed a high density mass compressing the spinal cord (Fig. 2). Spinal angiography performed on day 24 did not show any vascular malformations. However, simultaneous cerebral angiography revealed a cerebral aneurysm on the C2 segment of the left internal carotid artery (ICA) (Fig. 3). Treatment The symptoms gradually improved with conservative management. Mild hypertension (140–150/85–90 mmHg) was normalized to 110–130/70–80 mmHg within the second week. Third MRI study The patient was discharged with no neurological deficits on April 14, 2001. Follow up MRI on May 10, 2001 (day 69) showed further reduction of the SSH. One year after the onset, she was alert and living her usual life without any after e¤ects such as headache, back pain, neurological deficits or hypertension. Discussion According to the recent review by Domenicucci et al. [3], of 106 cases of non traumatic acute SSH, only 15 cases (14%) were considered to be purely spontaneous SSH occurring in the absence of risk factors for acute SSH. While many neurological disorders, such as eclampsia, pseudotumour cerebri, stroke and SAH can develop faster in the pregnant patient [5], there has been no evidence that pregnancy and/or childbirth per se could be a single risk factor for acute SSH. Of the two previous reported cases of SSHs occurring in the perinatal period, one patient was complicated due to lumbosacral epidural anaesthesia [12], and the other developed severe hypertension related to toxicoremia during pregnancy [17]. Our patient developed moderate degree of hypertension 5 days after the delivery which was probably induced by methylergometrine 154 Fig. 4. Follow up T2-weighted MR images obtained on day 17. (A) sagittal, and (B) transverse at Th3 level. The haematoma is reduced in size and exhibits partial hyperintensity. The spinal cord is decompressed spontaneously maleate. This condition is not uncommon during the postpartum period, and it is doubtful that moderate hypertension itself directly contributed to the formation of SSH. While the present case had a cerebral aneurysm, we considered it as a merely coincidental K. Yamada et al. one based on the following facts; first of all, the aneurysmal surface observed intraoperatively was smooth, lacked blebs and was unruptured, secondly, the patient experienced sudden onset back pain localized to the interscapular region rather than headache, and finally, CT findings at onset showed no intracranial SAH. Rader [19] postulated that the mechanisms underlying non traumatic SSH were based on the indirect application of force to the intraspinal vessels: a sudden rapid increase in abdominal and thoracic pressure would increase the intravascular pressure of the spinal vessels as they cross the subdural and subarachnoid spaces, and since the CSF pressure lags momentarily behind the intravascular pressure, this results in rupture of the spinal vessels. This mechanism could possibly be involved even in normal delivery, and the present case may be one of a very rare complication of normal delivery. However, in our patient, the onset of symptoms was delayed until 8 days after the delivery. One possible interpretation for this time lag is that marked damage of the spinal vessels and minor bleeding may have occurred during the childbirth, and that delayed massive re-bleeding occurred 8 days afterwards resulting in symptomatic haematoma. Spinal subdural and epidural haematomas occasionally share very similar imaging appearance on MRI or CT [3, 18]. Post and coworkers [18] reported that the subdural site of the haematoma was demonstrated by the existence of epidural fat, which was also clearly shown in the MRI in the present case (Fig. 1). In agreement with the report by Kulkami et al. [10], serial MRI monitoring of the present case demonstrated that the evolution of SSH is consistent with that of intracranial blood. Since most patients with SSH develop severe spinal symptoms, it was hypothesized that urgent surgical intervention is necessary to relieve spinal compression [11, 26]. Recently, Han et al. [8] suggested that Type I spinal AVMs should be considered in the di¤erential diagnosis of SSH in the thoracic or lumbar region, even if selective spinal angiography failed to demonstrate a pathological process, and recommended exploratory surgery to remove the SSH and to identify its source. However, spontaneous regression of SSH has been reported in a large number of patients [4, 9, 10, 16]: in these patients, the neurological symptoms and signs gradually improved as the haematoma resolved. The findings in the present case agree with the recently reported findings that conservative treatment for SSH 155 Spinal subdural haematoma after childbirth may be possible in selected cases with minimal neurological impairment under strict neurological and neuroradiological monitoring. References 1. Barker GL (1988) Spinal subdural haematoma following spinal anaesthesia. Anaesthesia 43: 664–665 2. Calhoun JM, Boop F (1988) Spontaneous spinal subdural haematoma: case report and review of the literature. Neurosurgery 29: 133–134 3. Domenicucci M, Ramieri A, Ciappetta P, Delfini R (1999) Nontraumaticacute spinal subdural haematoma: report of five cases and review of the literature. J Neurosurg 91 [Suppl] 1: 65– 73 4. Duprez T, Grandin C, Malghem J (1995) MRI monitoring of an acute spinal subdural haematoma with spontaneous resolution. Acta Neurol Bel 95: 101–103 5. Fox MW, Harms RW, Davis DH (1990) Selected neurologic complications of pregnancy. Mayo Clin Proc 65: 1595–1618 6. Gehri R, Zanetti M, Boos N (2000) Subacute subdural haematoma complicating lumbar microdiscectomy. J Bone Joint Surg Br 82: 1042–1045 7. Guthikonda M, Schimidek HH, Wallman LJ, Snyder TM (1979) Spinal subdural hematoma: case report and review of the literature. Neurosurgery 5: 614–616 8. Han PP, Theodore N, Porter RW, Detwiler PW, Lawton MT, Spetzler RF (1999) Subdural hematoma form Type I spinal arteriovenous malformation. Case report. J Neurosurg 90 [Suppl] 4: 255–257 9. Juvonen T, Tervonen O, Ukkola V, Klintrup HE (1994) Widespread posttraumatic spinal subdural hematoma – imaging findings with spontaneous resolution: case report. J Trauma 36: 262–264 10. Kulkami AV, Willinsky RA, Gray T, Cusimano MD (1998) Serial magnetic resonance imaging findings for a spontaneously resolving spinal subdural hematoma: case report. Neurosurgery 42: 398–400 11. Langmayr JJ, Ortler M, Dessl A, Twerdy K, Aichner F, Felber S (1995) Management of spontaneous extramedullary spinal haematomas: results in eight patients after MRI diagnosis and surgical decompression. J Neurol Neurosurg Psychiatry 59: 442–447 12. Lao TT, Halpern SH, MacDonald D, Huh C (1993) Spinal subdural haematoma in a parturient after attempted epidural anaesthesia. Can J Anaesth 40: 340–345 13. Longatti PL, Freschi P, Moro M, Trincia G, Carteri A (1994) Spontaneous spinal subdural hematoma. J Neurosurg Sci 38: 197–199 14. Masdeu JC, Breuer AC, Schoene WC (1979) Spinal subdural hematomas: Clue to a source of bleeding in traumatic lumbar puncture. Neurology 29: 872–876 15. Mastronardi L, Ferrante L, Scarpinati M, Gagliardi M, Celli P, Fortuna A (1991) Intradural extramedullary cavemous angioma: case report. Neurosurgery 29: 924–926 16. Mavroudakis N, Levivier M, Rodesch G (1990) Central cord syndrome due to a spontaneously regressive spinal subdural hematoma. Neurology 40: 1306–1308 17. Nishiyama K, Takahashi M, Tamura K (1982) Spinal subdural hematoma in pregnancy; a case report. Seikeigeka 33: 450–455 18. Post MJ, Becerra JL, Madsen PW, Puckett W, Quencer RM, Bunge RP, Sklar EM (1994) Acute spinal subdural hematoma: MR and CT findings with pathologic correlates. Am J Neuroradiol 15: 1895–1905 19. Rader JP (1955) Chronic subdural hematoma of the spinal cord. Report of a case. N Eng J Med 253: 374–376 20. Shimizu S, Tachibana S, Maezawa H, Fujii K, Kan S (1999) Lumbar spinal subdural hematoma following craniotomy – case report. Neurol Med Chir (Tokyo) 39: 299–301 21. Silver JM, Wilkins RH (1991) Spinal subdural hematoma formation following ventriculoperitoneal shunting for hydrocephalus. Case report. Acta Neurochir (Wien) 108: 159–162 22. Stewart DH, Watkins ES (1969) Spinal cord compression by chronic subdural hematoma. Case report. J Neurosurg 31: 80– 82 23. Swann KW, Ropper AH, New PF, Poletti CE (1984) Spontaneous spinal subarachnoid hemorrhage and subdural hematoma. Report of two cases. J Neurosurg 61: 975–980 24. Toledo E, Shalit MN, Segal R (1981) Spinal subdural hematoma associated with anticoagulant therapy in a patient with spinal meningioma. Neurosurgery 8: 600–603 25. Vinters HV, Barnett HJ, Kaufmann JC (1980) Subdural hematoma of the spinal cord and widespread subarachnoid hemorrhage complicating anticoagulant therapy. Stroke 11: 459–464 26. Yamagami T, Handa H, Higashi K, Takahashi JB (1995) Ventral spinal subdural hematoma – case report. Neurol Med Chir (Tokyo) 35: 896–900 27. Zilkha A, Nicoletti JM (1974) Acute spinal subdural hematoma. Case report. J Neurosurg 41: 627–630 Comment The authors report an interesting and very rare case of spinal subdural haematoma occurring 8 days after delivery without any evident cause. The patient received no medication, nor anticoagulant therapy. No epidural anaesthesia was used. Furthermore, that woman had an unruptured associated intracranial aneurysm. The question of the aetiology remains open. We agree that, according to their operative findings, the aneurysm was not the cause of the spinal haematoma. Moreover, the latter had a typical clinical onset. However, one may question on the presence of a small spinal vascular abnormality. Spinal angiography was reported as normal but were all the pedicles investigated on both sides from lumbar to cervical area? One may also debate on the conservative versus surgical treatment. In this case report, the evolution has been spontaneously favourable and the authors were right to observe rather than to operate since there was a spontaneous regression of clinical signs so as MRI abnormalities but that is not the rule in such pathology. But, that unusual case report is well written and brings original data to our daily practice. J. Brotchi Correspondence: Kazumichi Yamada, M.D., Ph.D., Department of Neurosurgery, Kumamoto University Medical School, 1-1-1 Honjo, Kumamoto 860-8556, Japan.