Spontaneous intracranial hypotension can resolve either spontaneously or with conservative treatment that includes bed rest, oral hydration, and caffeine intake. For patients who do not adequately respond to a conservative approach, injection of autologous blood into the spinal epidural space is the preferred invasive treatment. The success rate for each EBP varies from 30–70% for non-targeted patching, and up to 87% for targeted patching [2, 5, 6, 9].
The first EBPs has been reported to have shown successful clinical improvement in 64% of patients without the need for further intervention. However, a single EBP may not produce a permanent response in some patients and one or two additional patches at least seven days apart may be required to produce complete relief of symptoms [2, 5, 6, 8, 10, 11]. The mechanism of action of an EBP is not clear, but proposed mechanisms include an increase in intracranial CSF volume and pressure, plugging of the dural leak, and facilitation of rapid healing of the tear. The volumes of blood used in different studies have ranged from small (10 ml) to large (50 ml) amounts [5, 8, 11, 12].
RIH is an infrequently reported complication of EBP in patients with SIH. It results from postprocedural elevation of the CSF pressure. Up to 27% of patients can develop intracranial hypertension following a blood patch for treatment of SIH [1, 6, 13].
The Monro-Kellie hypothesis asserts that the total volume within the skull is constant and volume loss in one compartment will be compensated for by an increase in the other compartments. These compensatory mechanisms in SIH can lead to subdural fluid collection, venous system engorgement, and a possible increase in CSF production. Early RIH that occurs immediately after EBP can be explained by the failure of intracranial venous distension to reverse immediately after repair of the CSF leak. Delayed RIH can develop days to weeks after the EBP. The proposed mechanisms for this include upregulation of CSF production and disrupted CSF reabsorption during the period of the CSF leak. [6, 14–17].
Our report is an interesting case of a patient with SIH treated with EBP who developed both early RIH after 24 hours and delayed RIH at three weeks post-EBP. A failure to correctly diagnosis RIH and its differentiation from SIH can cause complications for a patient because the treatments are different. For a rebound headache, the strategy for management is to lower the CSF pressure . Our patient had a CSF pressure of 50 cm H2O, which was immediately treated with head elevation and intravenous mannitol, followed by oral acetazolamide for three to four months.
Kranz et al.  reported nine cases of confirmed RIH after EBP; however, most of their patients developed symptoms less than 48 hours post-EBP that included a change in headache location, new nausea, and new blurred vision. The average opening pressure reported was 30 cm H2O (range of 22–55) and all cases improved following lumbar puncture and administration of acetazolamide for durations varying from five days to three years. In most RIH cases reported, the onset of symptoms has been rapid (from immediate to three days) and papilledema has been rarely reported, possibly due to the appearance of the usual fundoscopic changes over a period of days to weeks. The CSF opening pressure measurement in these patients ranged from 20 to 55 cm H2O [7, 18, 19].
There are few reports of delayed RIH in the literature. In a study of 113 patients with SIH who underwent percutaneous or microsurgical treatment, 27.4% of patients developed RIH. In 23 patients, it occurred within 72 hours, in seven patients in 3–7 days, and in two patients in 7–30 days. Funduscopic examination showed papilledema in only two patients. A lumbar puncture was performed in four patients and found elevated pressure in only one patient (30 cm H2O) .
CSF pressure is measured infrequently in RIH because of the increased likelihood of a post-dural puncture headache. Therefore, the diagnosis of rebound intracranial hypertension primarily has been clinical. Lumbar puncture is done when the symptoms of elevated CSF pressure become intolerable or if there is doubt about the diagnosis of a rebound high-pressure headache.
Our patient developed both types of rebound headache after the EBP. The location of the headache changed from the occipital to frontal areas and the orthostatic feature reversed. The early positional headache was relieved after two days, whereas the delayed one did not respond to conservative management and a lumbar puncture was done after the symptoms and signs of increased intracranial pressure appeared. The opening CSF pressures recorded at the different times were 50 and 38 cm H2O.
As in cases like that reported by Kranz et al. , our patient responded very well to acetazolamide over a duration of about three months. Patients with RIH generally are reported to be younger, primarily female, and more often showed extradural CSF collection on the spinal imaging. These were all are applicable to our patient.
The discovery of previously elevated CSF pressure in 3% of patients with SIH, as well as the appearance of intracranial hypertension following surgical closure of the CSF leak in some reports, suggest that increased pressure could predispose a patient for a CSF leak at a weak point in the dura. This would be especially true for patients who develop delayed RIH after EBP . Our patient, however, had clinical and imaging presentations of decreased intracranial pressure from the onset of the disease.