Extra-axial haemorrhage in a patient with Alport syndrome after epidural anaesthesia
- 1 Department of Obstetrics & Gynaecology, Ipswich Hospital, Ipswich, Queensland, Australia
- 2 Department of Anaesthesia, Ipswich Hospital, Ipswich, Queensland, Australia
- Correspondence to Dr Shanika Wijayanayaka; shanika.wijayanayaka@uq.net.au
Abstract
Extra-axial haemorrhage following epidural anaesthesia is extremely rare. We present the case of an 18-year-old G1P0 woman with Alport syndrome who had a ventouse delivery for failure to progress that was complicated by a postpartum tonic–clonic seizure. Clinically, and confirmed radiologically, the patient was found to have experienced an extra-axial haemorrhage (extradural and subdural haemorrhage) secondary to a cerebrospinal fluid leak caused by a dural puncture during epidural anaesthesia. Differentiating between postdural puncture headache, subdural haemorrhage and extradural haemorrhage can be extremely challenging, but it is important to consider these rare conditions when evaluating patients presenting with postpartum headache and seizure.
Background
Inadvertent dural puncture during epidural anaesthesia in obstetrics has an estimated incidence of less than 1%.1 Subsequent postdural puncture headache (PDPH) is a well-recognised complication, whereas intracranial subdural haematoma (ISH) is rare but can have potentially serious sequelae. The incidence of ISH caused specifically by epidural anaesthesia used in obstetric practice has been estimated at 1:500 000 to 1:250 000.2 Given the rarity of this complication, most of the current evidence arises from case reports and a few small case series.2 Differentiating between PDPH, subdural haemorrhage (SDH) and extradural haemorrhage (EDH) can be difficult because they often share similar clinical presentations. Postpartum headache can have a wide range of differential diagnoses from normal postpartum headache to pre-eclampsia, cortical vein thrombosis or meningitis.2 Hence, to differentiate extra-axial haemorrhage from other clinically innocuous conditions, it is important to identify risk factors, to assess symptoms carefully and to perform appropriate imaging studies.
Case presentation
We present the case of an 18-year-old G1P0 with a background of Alport syndrome (heterozygous COL4A5 mutation, X-linked inheritance) and a family history of hypertension. Because of her medical history, she underwent shared antenatal care involving midwifery, high-risk obstetric care and Obstetric Medicine. Antenatally, she presented five times to the maternity day assessment unit having problem of headaches and was admitted twice for persisting headaches. On all occasions, the patient was normotensive and had normal examination and investigation results, such as full blood count and serum chemistry. She had a baseline proteinuria of 33 g/mol with stable renal function and was taking prophylactic aspirin and calcium.
The patient presented to the birthing suite in spontaneous labour at 39+4 weeks. An epidural was initiated on her request. Two attempts were made with an 18 G Tuohy needle into the L3/L4 vertebral space. There was loss of resistance to saline at 5 cm, and the catheter was fixed at 10 cm from the skin surface. There was no noted cerebrospinal fluid (CSF) or blood leak at the time of insertion. After the test and block doses had been administered, the patient developed motor blockade with hypotension at 97/50 mm Hg (despite her being normotensive before the procedure). The hypotension resolved promptly with a fluid bolus and head elevation. The epidural infusion was stopped and then restarted 2 hours later and had good analgesic effect without inducing motor blockade.
Given the patient’s slow progress, she was started on oxytocin for augmentation at 5 cm dilation. Following full dilation, she proceeded to 1 hour of passive descent, followed by 2 hours of active pushing. However, because of failure to progress in the second stage and the foetal position in the left occiput posterior position at +1 station, a ventouse delivery was required. The patient had a total estimated blood loss of 360 mL and received a single dose of intravenous antibiotics.
On postpartum day 1, the patient reported problem of generalised headache and pain at the epidural site but displayed no neurological deficits. Her vital signs were within the normal limits. She was reviewed by the anaesthetic pain service who noted symptomatic improvement as the patient moved from the erect to supine position, which suggested possible PDPH. She was treated with a sphenopalatine ganglion (SPG) block along with analgesia. Her symptoms improved markedly following the block and only a minor headache persisted.
On postpartum day 3, a medical emergency team call was activated because the patient had a generalised tonic–clonic seizure lasting 5 min. This had been preceded 1 hour before by problems of left-sided paraesthesia, severe headache and nausea. The patient was resuscitated acutely. After the seizure, she was found to have an incidental fever of 38.1°C, tachycardia (132 beats/min) and tachypnoea (32 breaths/min) but was normotensive with an oxygen saturation of 98% on 6 L of O2 via a Hudson mask. After the seizure, she displayed no persisting neurological deficits except for brisk reflexes.
Investigations
Post-seizure investigations revealed a haemoglobin level of 115 g/L, platelet count of 352×109/L, white cell count of 23.8×109/L, C reactive protein (CRP) concentration of 41 mg/L, lactate concentration of 4.5 mmol/L, albumin concentration of 29 g/L, mildly deranged liver function test results (alanine aminotransferase concentration of 38 U/L, aspartate aminotransferase concentration of 50 U/L), glucose concentration of 5.6 mmol/L, and normal renal function tests and electrolytes. Her activated partial thromboplastin time was 30 s, and the international normalised ratio was 0.9. Sepsis screening was conducted because of the incidental fever, but the blood culture results were normal.
Brain and spine CT was performed immediately and showed a right extra-axial acute haemorrhage overlying the right frontal lobe extending along the falx cerebri and likely along the tentorium for a maximum of 12 mm. CT also showed a small left acute SDH, measuring 3 mm maximally, overlying the frontal lobe. No midline shift was seen, and the cisternal spaces were normal (figure 1).
Head and spine CT. Right extra-axial acute haemorrhage and a possible small left acute subdural haemorrhage are evident.
Differential diagnosis
There are many possible aetiologies for a postpartum seizure in the context of an epidural catheter. The main differential diagnosis was eclampsia. However, the presentation was atypical as the patient had normal blood pressure and pre-existing proteinuria. She had been seen antenatally on multiple occasions for severe headaches, but pre-eclampsia had been excluded, and the headaches were managed with analgesia alone. A case report by Stotland et al suggested that intractable headache with no evidence of acute of chronic pathology can occur in patients with Alport syndrome.3
Stroke was considered as an alternative differential diagnosis, but the patient had no focal deficits. Intracranial haemorrhage (SDH) was considered, and head and spine CT was performed. Infection (eg, meningitis, epidural abscess, bacteraemia after instrumental delivery) was also considered because the patient had a fever at 38°C and elevated WCC count (23.8×109/L) and CRP concentration (41 mg/L). Metabolic causes were also explored but were discounted because of normal electrolyte and blood glucose levels, and renal function test results.
Treatment
Because of the initial suspicion of eclampsia, the patient was initially started on a magnesium sulphate infusion (MgSo4) and transferred to the birthing suite for 1:1 care before the CT scan. Multidisciplinary care was implemented once the CT scan revealed the extra-axial haemorrhage, and an immediate consultation was sought with a neurosurgery specialist at a tertiary hospital, along with local review by anaesthetics and intensive care specialists. On advice from the neurosurgery specialist, the MgSO4 was stopped, levetiracetam was started, and the patient was immediately transferred to the care of the tertiary neurosurgical team.
Under neurosurgical care in the high-dependency unit, the patient underwent a CT angiogram, which confirmed a stable, yet similar situation as that revealed by the brain and spine CT, and repeated full sepsis screening for her isolated fever. The sepsis screening suggested a urinary tract infection, and the patient was started on intravenous ceftriaxone.
On postpartum day 5, brain and spine MRI and MR venogram were performed and demonstrated a subtle T2 hyperintensity within the L4/L5 interspinous space, which suggested the location of a CSF leak. There were no changes in the size of the extra-axial haemorrhage on the right frontal and parietal lobes or in the small subdural component lying adjacent to the right anterior cerebral falx. Diffuse leptomeningeal enhancement or venous distension was noted. Clinically, her symptoms improved, and the patient exhibited no further headaches or neurological deficits.
On postpartum day 9, the patient was prescribed levetiracetam and discharged home after being reviewed by physiotherapy, occupational therapy, and speech pathology confirming near baseline function.
Outcome and follow-up
Based on the clinical findings and imaging results, the diagnosis was extra-axial haemorrhage secondary to intracranial hypotension resulting from a CSF leak, which was considered to be a complication of the labour epidural. The patient was reviewed by a neurosurgery specialist 1 month after delivery and was noted to have mild headaches but no neurological symptoms. She is scheduled for a repeat CT scan in 6 months, with plans to wean her from the antiepileptic medication.
Discussion
Postpartum headache is common and develops in 39% of women.4 Given the widespread use of neuraxial blocks in the obstetric population, these women are at higher risk of developing PDPH, which presents as a complication of inadvertent dural puncture during neuraxial anaesthesia. The incidence of PDPH following dural puncture with epidural anaesthesia is 76%–85%.4 However, SDH after dural puncture is extremely rare, with an incidence of 1:500 000 in obstetric practice.2
Differentiating between PDPH, SDH and EDH can be challenging given the similarity of symptoms (table 1). The incidence of SDH following dural puncture tends to be under-reported because patients are often treated for PDPH and show gradual improvement in symptoms.4 On postpartum day 1, the patient reported relief of her headache while in the supine position, and this postural component is typically associated with PDPH.4 However, on postpartum day 3, she had a tonic–clonic seizure, which indicated a more serious pathology.
Symptoms and signs of postdural puncture headache, subdural haematoma and extradural haematoma
| Postdural puncture headache | Subdural haematoma | Extradural haematoma |
| Neck pain | Non-postural headache | Severe headache |
| Nausea | Persistent headache | Nausea and vomiting |
| Tinnitus | Vomiting | Confusion |
| Photophobia | Convulsions | Dizziness |
| Hypoacusis | Hemiplegia | Drowsiness |
| Dizziness | Disorientation | Weakness in parts of the body |
| Interscapular pain | More serious neurological symptoms | Loss of consciousness |
Initially, the provisional diagnosis was PDPH, and the patient was treated with a SPG block and oral analgesics. SPG has gained interest as an effective treatment modality for PDPH but is not offered routinely given the limited evidence on the efficacy of SPG block for PDPH.5 However, the definitive treatment for PDPH is an epidural blood patch; with a success rate of approximately 85%.6 On postpartum day 3, brain imaging studies were requested urgently because of the onset of serious neurological symptoms (seizure, severe headache, left-sided paraesthesia), which confirmed the extra-axial haemorrhage. Determining when to request brain imaging studies can be challenging. Amorim et al suggested that imaging (CT or MRI) is mandated for the following conditions: a change in the headache’s response to posture (eg, postural to non-postural component), worsening of symptoms after an epidural blood patch, postural headache lasting >1 week, and development of neurological signs with the headache; as in our patient.7 8 Once the patient was stabilised acutely, she was immediately transferred to the neurosurgery department. An MRI was requested on postpartum day 5 and revealed a subtle T2 hyperintensity within the L4/L5 interspinous space, which indicated a CSF leak site.
Like PDPH, SDH shares a similar mechanism. Leakage of CSF from inadvertent dural puncture leads to reduced volume of CSF, which decreases intraspinal and intracranial pressure. These changes induce caudal movement of the spinal cord and brain, which results in stretching of the dura, pain-sensitive structures, cranial nerves and subdural bridging veins.2 9 10 Bridging veins, which are thinnest and most fragile in the subdural portion, can rupture and subsequently result in an ISH; intracranial hypotension is a prodrome of ISH.
The risk factors associated with SDH in a parturient are peripartum dehydration, postpartum diuresis, the second stage of labour (which causes abrupt release of intra-abdominal pressure and vena cava compression at delivery), increased maternal pushing (which can increase CSF leakage), hormone-induced ligamentous changes and venous congestion.11 Other risk factors include cerebral atrophy, anticoagulant use, cerebral vascular abnormalities and excessive CSF leakage from multiple punctures with a large-bore needle.9 10 The factors that may have led to the development of SDH in this case include the multiple dural punctures, ventouse delivery (which could have increased CSF leakage), antithrombotic use (aspirin until 12 hours before epidural insertion and prophylactic low-molecular-weight heparin after delivery), and type IV collagen defects associated with her background of Alport syndrome.
Extradural haematoma is the extra-axial collection of blood within the potential space between the outer layer of the dura mater and the inner surface of the skull. Although epidural haematoma (EDH) is commonly associated with arterial bleeding, up to 10% occur with venous bleeding secondary to traumatic or non-traumatic mechanisms, such as underlying coagulopathy and vascular malformations.12 Venous bleeding may result from a combination of increased CSF pressure (during active pushing and ventouse delivery), and compensatory vasodilation with a loss of CSF. In contrast, arterial bleeding may be a consequence of increased mean arterial pressure with loss of CSF; effectively increasing the transmural gradient across the blood vessel wall.12 To our knowledge, no cases of intracranial EDH have been reported following inadvertent dural puncture, which suggests further significance of this case. The exact mechanism responsible for EDH in this case is uncertain. However, an increase in transmural pressure gradient across the dura, in the setting of a type IV collagen defect, may have played a role in the development of EDH in this case. Additionally, undiagnosed aneurysmal rupture or spontaneous obstetric EDH, although rare, is still a possibility.13 Antenatal imaging would have been helpful in ruling out any pre-existing aneurysm or arteriovenous malformation.
Clinically, EDH is generally associated with acute changes in mental status, whereas SDH typically has a more gradual onset and is often seen in older populations.14 There is a lower threshold for surgical management of EDH, as uncontrolled bleeding may lead to progressive expansion and permanent neurological injury.15 Triggers for surgery in SDH include a midline shift >5 mm, haematoma thickness >10 mm or neurological deterioration.2 In this case, the patient was treated conservatively on the basis of a small acute SDH with no midline shift; as is supported by the literature.2 Prognostically, full recovery is expected in over 80% of patients, but death has been reported in 7%–10% of patients.2
Clinically, differentiating between PDPH, SDH and EDH can be challenging. When treating the obstetric population, clinicians should be vigilant with women who have undergone multiple punctures and/or are taking antithrombotic drugs because these patients are at higher risk of SDH following neuraxial blockade.7 Some reports have suggested that, given the haemostatic changes and differences in dural elasticity, pregnant patients are more susceptible to SDH.16 17 Delayed initiation of anticoagulants in patients with suspected PDPH may be considered.
Learning points
-
The incidence of intracranial subdural haematoma specifically caused by epidural anaesthesia used in obstetric practice has been estimated at 1:500 000 to 1:250 000.
-
Brain medical imaging should be requested without delay if the patient exhibits neurological symptoms and signs with a headache.
-
Extradural haemorrhage has not been reported as a complication of epidural anaesthesia, although in this case, the type IV collagenopathy of Alport syndrome may have contributed to its genesis.
-
Clinicians should consider delaying initiation of deep vein thrombosis prophylaxis in patients with suspected postdural puncture headache.
-
Regular monitoring for the development of neurological complications following epidural anaesthesia should be considered because early diagnosis and prompt treatment usually lead to complete resolution.
Footnotes
-
Contributors SW drafted the manuscript. AG contributed to review and editing of the manuscript. MV contributed to review and editing of the manuscript. KS contributed to review and editing of the manuscript. All authors contributed equally to creation of this manuscript.
-
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
-
Competing interests None declared.
-
Provenance and peer review Not commissioned; externally peer reviewed.
- © BMJ Publishing Group Limited 2021. No commercial re-use. See rights and permissions. Published by BMJ.
References
Use of this content is subject to our disclaimer