Free
Case Reports  |   November 2007
Subacute Spinal Subarachnoid Hematoma after Spinal Anesthesia That Causes Mild Neurologic Deterioration
Author Affiliations & Notes
  • Joon-Hee Park, M.D.
    *
  • Keun-Mann Shin, M.D., Ph.D.
  • Sung-Joon Hong, M.D.
    *
  • Il-Suk Kim, M.D.
    *
  • Sung-Keun Nam, M.D.
  • * Lecturer, † Professor, ‡ Resident.
Article Information
Case Reports / Central and Peripheral Nervous Systems / Coagulation and Transfusion / Regional Anesthesia
Case Reports   |   November 2007
Subacute Spinal Subarachnoid Hematoma after Spinal Anesthesia That Causes Mild Neurologic Deterioration
Anesthesiology 11 2007, Vol.107, 846-848. doi:10.1097/01.anes.0000287195.65822.cb
Anesthesiology 11 2007, Vol.107, 846-848. doi:10.1097/01.anes.0000287195.65822.cb
SPINAL subarachnoid hematoma after lumbar puncture has been reported to occur in the presence of preexisting coagulopathy or anticoagulant therapy.1–4 Although occult spinal subarachnoid hematoma may be possible, symptom onset generally is acute and dramatic, and it also represents serious complications for hemostatically compromised patients.5–7 
We report a patient who developed spinal subarachnoid hematoma after lumbar puncture. However, she had normal blood coagulation profiles and no spinal disorders such as spinal stenosis or spinal fractures. The onset of her symptoms was gradual, and neurologic deterioration was mild. To our knowledge, this is the first report about spinal subarachnoid hematoma with mild and vague symptoms in a hemostatically normal patient.
Case Report
A previously healthy 47-yr-old woman underwent an elective sling operation for stress incontinence by our urologic department. At our hospital, coagulation profiles, including platelet counts, prothrombin time, activated prothrombin time, and bleeding time used to be routinely checked in all patients, 1 week before surgery. Laboratory examinations such as platelet counts (254,000/mm3), prothrombin time (119.4%), activated prothrombin time (27.9 s), and bleeding time by the method of Duke (3 min) were within normal limits. The patient had no clinical history of bleeding tendency, and she also had not taken any medication.
The patient’s height and weight were 1.57 m and 46.6 kg, respectively. Before spinal anesthesia, the anesthetist explained the procedure and its possible complications. The patient wanted spinal anesthesia.
The patient was placed in the left lateral position. Under sterile conditions, a 26-gauge Quincke spinal needle (viola) (Sato, Kitamoto, Japan) was inserted into the L3–L4 interspaces, but accidental paresthesia occurred through the patient’s left leg. The needle was removed, and a second attempt at L4–L5 with a 26-gauge Quincke spinal needle (viola) successfully yielded clear cerebrospinal fluid. Two milliliters heavy bupivacaine, 0.5%, was injected slowly through the spinal needle. In the aspiration after injection, there was no bleeding, and the cerebrospinal fluid was clear. The spinal anesthetics produced a sensory block to T10, with motor block to L1 or more. After the operation, the patient’s estimated blood loss was 50 ml, and the operation time was 60 min. Therefore, no blood product or colloid solution was infused. Three hundred milliliters Hartmann solution was infused during the operation.
Five days after the spinal anesthesia, the patient noted intermittent low back pain and bilateral lower extremity pain in both thighs and the gluteal areas. She also reported gait discomfort because of her pain. However, she had no voiding difficulty. Although these symptoms increased, muscle power and sensory functions were intact. Plain radiographs of the spine did not reveal any abnormalities. A sagittal T2-weighted magnetic resonance image of the spine disclosed a mass-like lesion of high signal intensity in the intradural space from the L4 lower endplate level to the S1 lower endplate. It was located between the nerve roots of the cauda equina (fig. 1). The official magnetic resonance image reading indicated a suspected spinal subarachnoid hematoma in the L4–S1 area.
Fig. 1. Magnetic resonance imaging demonstrating spinal subarachnoid hematoma. T1-weighted sagittal image showing heterogenous increased signal intensity consistent with subarachnoid hematoma from L4 to S1. 
Fig. 1. Magnetic resonance imaging demonstrating spinal subarachnoid hematoma. T1-weighted sagittal image showing heterogenous increased signal intensity consistent with subarachnoid hematoma from L4 to S1. 
Fig. 1. Magnetic resonance imaging demonstrating spinal subarachnoid hematoma. T1-weighted sagittal image showing heterogenous increased signal intensity consistent with subarachnoid hematoma from L4 to S1. 
×
Before the patient was transferred to the orthopaedic department, we asked her again meticulously about previous bleeding tendency, but no history could be found out. Investigations at this time included a normal coagulation screen. Platelet counts (252,000/mm3), prothrombin time (118.4%), activated prothrombin time (27.2 s), and bleeding time by the method of Duke (3 min) were within normal limits. In addition, we performed thromboelastography. Normal findings of the thromboelastography were obtained (reaction time, 18.3 min; k time, 7.2 min; α angle, 27.6°; maximum amplitude, 50.0 mm). Despite conservative treatments such as medication and physical therapy, the patient’s radiating pain was getting worse. Therefore, she decided to undergo surgery. A laminectomy was performed at L4 and L5. There was no evidence of bleeding in the epidural space. On opening the dural sac, there was clotted blood between the cauda equina in the subarachnoid space. The hematoma was removed completely from the dural sac. After no hematoma in the subdural space was confirmed, the dura was sutured. After surgical decompression, the patient’s bilateral sciatica improved considerably, and only minimal low back pain remained.
Discussion
Spinal subarachnoid hematoma that arises from a spinal tap is usually associated with anticoagulants and underlying coagulopathy.1,2 To our knowledge, only one case has reported that spinal subarachnoid hematoma may be the result of technical difficulties encountered in spinal anesthesia.3 
Within the subarachnoid space at the level of the caudal equine, the only vessels of substantial size are the radiculomedullary artery of Adamkiewicz and its corresponding vein.4 On occasion, this vessel may arise low and accompany L3, L4, or L5 nerve roots, where it could be jeopardized during lumbar puncture. Most authors believe such hemorrhages are caused by injury to these structures or to the smaller radicular vessels entering the subarachnoid space with each segmental nerve root.8,9 Masdeu et al  .9 confirmed this mechanism at autopsy. Breuer et al  .8 estimated that the frequency of brushing a nerve root, with the associated risk of lacerating the radicular artery or vein on its surface, was more than 25%. They suggested that the frequency of occult spinal subarachnoid hemorrhage is much higher than currently suspected, but is usually clinically significant only in hemostatically compromised patients. Ruff and Dougherty2 reported the results of the only large prospective study on the risks of lumbar puncture followed by anticoagulation versus  lumbar puncture alone. They found a significantly higher incidence of complications from lumbar puncture in the group receiving anticoagulation therapy. Spinal subarachnoid hemorrhage rarely occurs in patients who have not been anticoagulated, and although spinal subarachnoid hemorrhage occurs, the development of a discrete hematoma causing symptoms of spinal cord or root compression is unusual. First, blood in the cerebrospinal fluid is rapidly diluted by diffusion, which is facilitated by spinal motion. It rarely reaches the concentration necessary for clot formation.10 Second, the cerebrospinal fluid has an intrinsic fibrinolytic activity that increases after hemorrhage.11 Third, the pulsatile motion of the dural sac aids this process. Kirkpatrick and Goodman12 presumed that an anatomical block or relatively anatomical block to normal cerebrospinal fluid flow due to preexisting disease might contribute to hematoma formation.
In our case, there were significantly different points compared with previous reports. Our patient had no associated coagulopathy and took no anticoagulants. We confirmed this not only by the traditional clotting tests but also by thromboelastography, which provides a kinetic analysis of the entire clot formation and stabilization as well as clot dissolution by the fibrinolytic system. In the only previous report in which spinal subarachnoid hematoma was caused by technical difficulties of lumbar puncture, the patient was 81 yr old, so she was thought to have such a stenotic spinal canal due to natural degenerative changes that it might be a causative factor of spinal subarachnoid hematoma.13 In our case, however, the patient had no preexisting anatomic block of cerebrospinal fluid flow such as spinal stenosis, fractures, or previous surgery, as her lumbar spine magnetic resonance image showed.
It has been reported that spinal subarachnoid hematoma causes compromise of the neurologic state and rapid deterioration.1,2,5–8,13,14 
Furthermore, Kreppel et al  .15 reviewed all kinds of spinal hematoma, including epidural, subdural, and subarachnoid hematoma. In his review, the subarachnoid hemorrhage after lumbar puncture was also rare, and the patients’ coagulation systems had been impaired. Although some cases had subacute clinical manifestation (5.6%), clinical symptoms were almost severe, including neurologic deficit. In our case, a hematoma was formed between the fourth and fifth lumbar segments. However, the patient had only mild back pain and discomfort of the thighs and gluteal areas 4 days after lumbar puncture. There was no neurologic deterioration, and it can be explained as follows. First, our patient did not have any disorders related to blood coagulation and had taken no anticoagulants; therefore, subarachnoid hemorrhage by lumbar puncture might be limited by the natural coagulation system. Second, although two attempts had been made during the spinal anesthesia and the patient felt paresthesia on the first attempt, it was never thought to be traumatic or a difficult lumbar puncture case. Because we used a 26-gauge spinal needle and no blood was found through the spinal needle, vessel injury by the lumbar puncture was not considered to be much.
Our case shows that spinal subarachnoid hematoma may occur in a patient who has a normal coagulation system and has taken no anticoagulants, through a nondifficult, nontraumatic lumbar puncture. Most of all, the clinical manifestation of our patient is different from previous reports in that she reported only mild and vague sciatica with no neurologic deficit several days after the lumbar puncture. In conclusion, although clinical features seem to be doubtful for spinal subarachnoid hematoma, early computed tomographic myelography or magnetic resonance imaging should be considered in case surgical decompression may be indicated for improving clinical symptoms and preventing complications of the bowel or bladder that are significant sequelae of cauda equina syndrome.
References
Owens EL, Kasten GW, Hessell EA: Spinal subarachnoid hematoma after lumbar puncture and heparinization. Anesth Analg 1986; 65:1201–7Owens, EL Kasten, GW Hessell, EA
Ruff RL, Dougherty JH: Complications of lumbar puncture followed by anticoagulation. Stroke 1981; 12:879–81Ruff, RL Dougherty, JH
Bills DC, Blumbergs P, North JB: Iatrogenic spinal subdural hematoma. Aust N Z J Surg 1991; 61:703–6Bills, DC Blumbergs, P North, JB
Edelson RN, Chernik NL, Posner JB: Spinal subdural hematomas complicating lumbar puncture: Occurrence in thrombocytopenia patients. Arch Neurol 1974; 31:134–7Edelson, RN Chernik, NL Posner, JB
Brem SS, Hafler DA, Van Uitert RL, Ruff RL, Reichert WH: Spinal subarachnoid hematoma: A hazard of lumbar puncture resulting in reversible paraplegia. N Engl J Med 1981; 303:1020–1Brem, SS Hafler, DA Van Uitert, RL Ruff, RL Reichert, WH
Blade J, Gaston F, Montserrat E, Marin P, Granena A, Bachs A, Rozman C: Spinal subarachnoid hematoma after lumbar puncture causing reversible paraplegia in acute leukemia. J Neurosurg 1983; 58:438–9Blade, J Gaston, F Montserrat, E Marin, P Granena, A Bachs, A Rozman, C
Diaz FG, Yock DH, Rockswold GL: Spinal subdural hematoma after lumbar puncture producing acute thoracic myelopathy: Case report. Neurosurgery 1978; 3:404–6Diaz, FG Yock, DH Rockswold, GL
Breuer AC, Tyler HR, Marzewski DJ, Rosenthal DS: Radicular vessels are the most probable source of needle-induced blood in lumbar puncture: Significance for the thrombocytopenic cancer patient. Cancer 1982; 49:2168–72Breuer, AC Tyler, HR Marzewski, DJ Rosenthal, DS
Masdeu JC, Breuer AC, Schoene WC: Spinal subarachnoid hematomas: Clue to a source of bleeding in traumatic lumbar puncture. Neurology 1979; 29:872–6Masdeu, JC Breuer, AC Schoene, WC
Scott EW, Cazenave CR, Virapongse C: Spinal subarachnoid hematoma complicating lumbar puncture: Diagnosis and management. Neurosurgery 1989; 25:287–93Scott, EW Cazenave, CR Virapongse, C
Tovi D, Nilsson IM: Increased fibrinolytic activity and fibrin degradation products after experimental intracerebral haemorrhage. Acta Neurol Scand 1972; 48:403–15Tovi, D Nilsson, IM
Kirkpatrick D, Goodman SJ: Combined subarachnoid and subdural hematoma following spinal puncture. Surg Neurol 1975; 3:109–11Kirkpatrick, D Goodman, SJ
Mayumi T, Dohi S: Spinal subarachnoid hematoma after lumbar puncture in a patient receiving antiplatelet therapy. Anesth Analg 1983; 65:777–9Mayumi, T Dohi, S
Rengachary SS, Murphy D: Subarachnoid hematoma following lumbar puncture causing compression of the cauda equina. J Neurosurg 1974; 41:252–4Rengachary, SS Murphy, D
Kreppel D, Antoniadis G, Seeling W: Spinal hematoma: A literature survey with meta-analysis of 613 patients. Neurosurg Rev 2003; 26:1–49Kreppel, D Antoniadis, G Seeling, W
Fig. 1. Magnetic resonance imaging demonstrating spinal subarachnoid hematoma. T1-weighted sagittal image showing heterogenous increased signal intensity consistent with subarachnoid hematoma from L4 to S1. 
Fig. 1. Magnetic resonance imaging demonstrating spinal subarachnoid hematoma. T1-weighted sagittal image showing heterogenous increased signal intensity consistent with subarachnoid hematoma from L4 to S1. 
Fig. 1. Magnetic resonance imaging demonstrating spinal subarachnoid hematoma. T1-weighted sagittal image showing heterogenous increased signal intensity consistent with subarachnoid hematoma from L4 to S1. 
×