Intrathecal haematoma after an epidural blood patch

Discussion

The frequency of accidental dural puncture (ADP) ranges from 0.19% to 3.6%.2 The injury to the dura can be due to an obvious puncture of the dura from an epidural needle, or it can be due to an unrecognised dural puncture, either by an epidural needle or from a catheter. If there was a recognised ADP during epidural insertion, then the incidence of PDPH is as high as 16%–86% with a meta-analysis done by Choi et al suggesting an incidence of 50%–55%.3 4

Conservative treatment will not provide complete relief in most patients who have had a large gauge ADP. About two-thirds of them would require an EBP making it not an uncommon procedure in labour wards.5 6 EBP itself has been associated with neurological complications, which, fortunately, are rare. These include paraesthesia, neck ache, facial nerve palsy and lumbovertebral syndrome.7–9 Besides these, seizures, severe headache and transient bradycardia have been reported during and immediately after the procedure.3

Our patient in this case report presented with an intrathecal haematoma after an uneventful blood patch. Intrathecal haematoma is an accumulation of blood around the dura and the vertebrae. Although most patients are asymptomatic, in rare cases, it can cause spinal cord compression, which leads to neurological complications. In such cases, patients can present with neuropathic pain, sensory deficits, bowel or bladder dysfunction, motor weakness, or in severe cases, complete paralysis of the affected limb.10 Kearsley and McCaul reported two cases of subdural haematomas after an uneventful EBP leading to backache and radicular symptoms, both resolved with conservative treatment.11 Intrathecal haematoma can cause meningitis symptoms, epileptic seizures, alterations in consciousness and can be misdiagnosed as cerebral haemorrhage.12 A case report by Seemiller et al describes a serious complication of an EBP causing convulsion, altered mental state, and respiratory distress requiring intubation and critical care stay.13 In their case, the blood had diffusely spread in the cerebrospinal fluid (CSF) with both MRI and CT scans of the brain demonstrating diffuse cortical sulci obscuration due to diluted blood products. Their case demonstrates that if a patient develops any changes in mental state, seizure or respiratory failure, it is vital to consider a diffuse spread of blood rather than just a localised haematoma.

Often patients will present with a severe, constant back pain, with or without a radicular component.10 In our case, there was a localised intrathecal haematoma causing clumping of nerves resulting in generalised back pain and radicular shooting pain down the legs. The patient did not have any sensory deficit, motor weakness or involvement of the bladder or the bowel.

Thorough history taking and clinical examination remain essential in diagnosis of neuropathic pain. If neuropathic pain is suspected, the assessment should include motor, sensory and autonomic system. In cases of sensory deficit, it may be helpful to use a body sensory map to have a visual demonstration of the areas of deficit.14 The Standard Evaluation of Pain (StEP) may be used as a tool to standardise the approach for differentiating the different types of pain. StEP involves 6 interview questions and 10 physical tests.14 In our case, the history and clinical examination directed us towards the suspicion of an intrathecal haematoma, which was subsequently confirmed on MRI.

The mechanism of intrathecal haematoma remains undetermined, but few hypotheses have been suggested. The first one is that the blood was deposited in the epidural space as intended, but, due to the resulting increase in the epidural pressure, some blood might have been pushed inadvertently into the intrathecal space through the existing dural puncture hole.15 The second hypothesis by Verduzco et al suggests that the blood was injected into either the subdural space with subsequent leak into the intrathecal compartment or that it was injected directly into the subarachnoid space.16 In the case of a subdural haematoma following an EBP, a subsequent lumbar spine MRI did not show evidence of blood or fluid in the epidural space. The authors hypothesised that the Tuohy needle was placed subdurally during the EBP and that increased pressure from the injection of blood caused dissection of the dura and arachnoid mater. This case by Verduzco et al also demonstrates that the safety and success of EBP depend on the accurate placement of Tuohy needle in the epidural space. While doing an EBP, it is essential to avoid the same interspace as the original dural puncture level. However, in some cases, it might be challenging to identify the dural space accurately, especially in a lateral position in which most of the EBP is performed. Another possible explanation is that the blood in the subarachnoid space may be due to arterial or venous puncture at the initial placement of the epidural catheter before the EBP.15

Narouze, in his editorial, described EBP as an ‘iatrogenic epidural haematoma’. Injection of blood into the epidural space causes a mass effect, reduction in epidural space compliance and results in cephalad CSF shift. This has been demonstrated by MRI, where the haematoma caused thecal sac and nerve root compression.17

In our case, intrathecal haematoma after the EBP resulted in nerve compression leading to shooting pain down the legs. It highlights the importance of urgent MRI in all patients who present with severe back pain or lower leg radicular pain after an EBP. A mass effect demonstrated on MRI, together with radicular symptoms and bladder and bowel problems, might require an urgent laminectomy. If the patient presents with a seizure or altered cognition, we should consider the potential diffuse spread of the blood in the CSF, in which case supportive treatment in critical care may be required.14 A multidisciplinary approach involving neurology and neurosurgery helps to determine the most appropriate treatment options for such patients.

Our case also illustrates that EBP itself is associated with its own complications profile and, as an anaesthetist, it is vital to include these in consent. This is especially true after the ‘Montgomery vs Lanerkshire Health Board’ judgement, where all the material risks need to be discussed with the patient.18–20

Given the possible complications related to EBP, it is also important to consider alternate options to prevent or treat the PDPH. One option could be inserting the intrathecal catheter during the inadvertent dural puncture, which has some evidence of decreasing the need of EBP.16 21 Other options could be a sphenopalatine block or a suboccipital muscle injection. A bilateral transnasal sphenopalatine block has been shown to be effective in management of moderate PDPH.22 Furthermore, if an EBP is essential in the management of the PDPH, it is important to take steps to minimise the risk of developing an intrathecal haematoma. We can ensure that the EBP is performed by senior anaesthetists, and also use appropriate volume and pressure when injecting the autologous blood.16

In addition to the above strategies, it is also important that all patients treated for PDPH should be regularly followed up for worsening of existing or new symptoms. Careful follow-up will allow prompt diagnosis and treatment leading to a better patient outcome.