Myelography and spinal anaesthesia are known to be a cause of intracranial hypotension resulting from liquorrhoea. However, spontaneous intracranial hypotension is a rare event and occasionally it is responsible for obstinate postural headache. We describe two patients with postural headache attributable to spontaneous intracranial hypotension, which was resolved by application of a cervical epidural blood patch with 10 mL autologous blood.
A 28-yr-old male in excellent health suddenly developed a headache accompanied by nausea. The headache started in the occipital region and worsened significantly on standing. Neurological examination revealed no abnormality and a lumbar puncture showed an opening pressure of zero cmH20. Cranial magnetic resonance imaging, with the administration of gadorinium, revealed enhancement of the dura mater. Radionucleotide cisternography after lumbar puncture showed an extradural leak at the C6/C7 level. The patient remained in bed and received 2000 mL intravenous fluids for 7 days. Since the headache was not completely alleviated, an epidural blood patch – with 10 mL aseptic autologous blood at the C6/C7 interspace – was instituted. Nevertheless, the headache was not completely relieved, and further neck stiffness was noted after the blood patch. Because the headache was persistent for the following week, we suspected that previous lumbar punctures were responsible for the leakage of cerebrospinal fluid. Therefore, another epidural blood patch with 30 mL blood, at the L1/L2 interspace, was applied to close the extradural leak. This manoeuvre was effective and the headache and nausea abated. However, neck stiffness persisted for another two weeks, although it was not made worse by the second epidural blood patch. The patient was discharged five days after the second epidural blood patch.
Our second patient was a 41-yr-old male who was admitted to hospital because he experienced a two day history of orthostatic headache. He had no history of chronic headache, craniocervical trauma or dural puncture. His headache had started suddenly in the occipital region when he extended his neck. Orthostatic posture worsened the symptoms, and induced nausea and severe pain confined the patient to bed. Neurological examinations revealed no abnormality and a lumbar puncture showed a depressed opening pressure (zero cmH2O). Cranial computed tomography showed an increased space between the brain and the cranium, and cranial magnetic resonance imaging, with gadorinium administration, revealed enhancement of the tentorium cerebelli. Indium-111 radionucleotide cisternography demonstrated an extradural leak in the upper cervical region. From these results, we made a diagnosis of spontaneous cerebrospinal fluid leakage. The patient stayed in bed and received 3000 mL intravenous fluid per day for the following seven days. However, as the complaint continued for another one week, an epidural patch was planned and nine days after the onset of his headache, the patient received an epidural injection of 10 mL aseptic autologous blood at the C7/T1 interspace. His postural headache resolved completely by 24 h; however, he now complained of mild neck discomfort for several days. The patient was discharged three days after an epidural blood patch and has remained well.
Spontaneous postural headache was reported by Schaltenbrand , where it was called aliquorrhea. This kind of headache usually persists for several months when patients receive only conservative treatment. While the exact aetiology for this syndrome has not yet been clarified, cerebrospinal fluid leakage through a thecal tear, reduced production of cerebrospinal fluid or by hyperabsorption of cerebrospinal fluid are listed as the pathogenesis . An epidural blood patch is highly effective in the management of spontaneous intracranial hypotension  when a cerebrospinal fluid leak can be demonstrated by spinal imaging; the most probable mechanism for the spontaneous intracranial hypotension is cerebrospinal fluid leakage through a thecal tear provoked by an external force or a meningeal diverticulum .
Previous reports described a relationship between a severe headache and postural change (hyperflexion of body)  or a strong blow to the head. In our case, severe headache developed immediately after retroflexion of the neck. What are the underlying systemic disorders? Four patients of spontaneous intracranial hypotension had findings suggestive of connective tissue disorders . In addition, spontaneous intracranial hypotension has complicated Marfan's syndrome, where arachnoid diverticula exist. On the other hand, spontaneous intracranial hypotension can occur in patients without any particular underlying disease. However, our cases are idiopathic since neither of the two patients had any special history of underlying disease.
Baker first reported a case of spontaneous intracranial hypotension where a headache was relieved by an epidural blood patch . Recently, this treatment is recognized as a highly effective treatment for spontaneous intracranial hypotension [2,3]. The effectiveness of this manoeuvre is >80% (26/32 cases reported in English language publications). Other treatments, e.g. epidural saline infusion, oral corticosteroid administration , fluid intake [2,4] or caffeine , are less effective (effective rate 57%, 63%, the latter two treatments gave transient relief, respectively). In 10 patients with a meningeal diverticulum or focal spinal cerebrospinal fluid leakage, surgical intervention was most effective, especially in patients in whom multiple attempts at non-surgical treatment had failed, achieving 100% relief from the headache . However, we should always bear in mind that this is traumatic therapy.
In our patients, a cervical epidural blood patch with 10 mL autologous blood was effective. However, in the first patient, we had to add a lumbar epidural blood patch to patch the fistula generated by the lumbar punctures. Despite the lumbar puncture being indispensable in the diagnosis of a spontaneous intracranial hypotension patient, we should pay attention to the fact that this process per se can trigger headache.
A cervical epidural blood patch is superior to a lumbar epidural blood patch because the quantity of blood required is smaller. Furthermore, this may decrease the possibility of meningitis or septicaemia. Our patients complained of mild neck discomfort only after the cervical injection. As De Rosaryo  has suggested, it is important to follow-up these patients regularly with special reference to the possibility of neurological deficits developing after this procedure. This is chiefly because of the anatomy of the cervical epidural space: a very narrow space between the ligamentum flavum and the dura mater. From our clinical experience, it appears to be safe and effective to control and use <10 mL blood. We recommend the very careful application of a cervical epidural blood patch to patients with persistent headache due to spontaneous intracranial hypotension, although we should always be mindful of possible complications such as a neurological deficit , septicaemia  and postdural puncture headache.
1. Schaltenbrand G. Neure Anschauungen zur Pathophysiologie der Liquorzirkulation. Zentralbl Neurochir
2. Benzon HT, Nemickas R, Molloy RE, et al.
Lumbar and thoracic epidural blood injections to treat spontaneous intracranial hypotension. Anesthesiology
3. Weitz SR, Drasner K. Spontaneous intracranial hypotension. Anesthesiology
4. Schievink WI, Meyer FB, Atkinson JL, Mokri B. Spontaneous spinal cerebrospinal fluid leaks and intracranial hypotension. J Neurosurg
5. Baker CC. Headache due to spontaneous low spinal fluid pressure. Minn Med
6. De Rosaryo AM. Commentary. Anesthesiol Rev
7. Cornwall TD, Dolan WM. Radicular back pain following lumbar epidural blood patch. Anesthesiology
8. Usui T, Saito S, Goto F. Epidural blood patch and spontaneous intracranial hypotension. Pain Clinic