Aquaporin-4 antibodies and idiopathic intracranial hypertension: the jury is in and the channels are out

The aquaporins are transmembrane water channel proteins that regulate water homeostasis in the body; of the many aquaporin (AQP) subtypes, AQP-4 is the predominant channel expressed in brain, with expression localized to astrocytic foot processes surrounding capillary endothelium (1). We generally think of AQP-4 in the context of cerebral edema. AQP-4 appears to be important in the development of cellular (cytoxic) edema in response to ischemic and toxic injuries (1), as high levels of AQP-4 expression correlate with sites of early cellular edema. Moreover, mice that are deficient in AQP-4 are more resistant both to cellular edema that is induced experimentally by water loading and to ischemia (2,3). On the other hand, AQP-4 does not appear to influence the rate of extracellular (vasogenic) edema formation. Extracellular edema results from processes that interrupt the blood-brain barrier (BBB), such as tumors, intracranial hemorrhage and brain infections (1). AQP-4 may be involved in fluid clearance resulting from extracellular edema, and it has been hypothesized that ‘normal’ clearance of proteins through the ventricles and CSF pathways is reversed when the intracranial pressure (ICP) is elevated. According to this theory, when the system becomes overwhelmed in the setting of intracranial hypertension/hydrocephalus, extracellular fluid is absorbed via the brain parenchyma via capillary clearance (4). AQP-4-deficient mice, studied under conditions known to provoke BBB disruption and edema formation, developed greater increases in water content and ICP than wild-type mice (5).

Idiopathic intracranial hypertension (IIH) is characterized by increased ICP without ventriculomegaly, occurring predominantly in obese females in their child-bearing years. The absence of hydrocephalus makes IIH somewhat more usual than other disorders causing increased ICP, and this aspect of IIH has confounded clinicians and researchers since the disease was first described. A disorder of cerebrospinal fluid regulation is implicit but the pathogenesis of IIH remains elusive.

Ekizoglu and colleagues measured serum anti-AQP-4-antibody levels in subjects with IIH, neuromyelitis optica and controls with encephalitis and report their findings in the current issue of Cephalalgia (6). AQP-4 antibodies were not detected in any patients with IIH. Their study has potential limitations, as not all patients were studied at the onset of their disease and those with ongoing chronic headaches may have had normal intracranial pressure at the time they were studied. Nonetheless, the results were consistent with those of Dhungana et al, who did not detect anti-AQP-4-antibody levels in the serum and CSF of participants with IIH or in control participants being investigated for acute headache conditions (7).

AQP-4 antibodies are commonly (but not ubiquitously) found in patients with neuromyelitis optica (NMO), a demyelinating disorder ascribed to inflammatory and autoimmune processes and not associated with increased ICP. These antibodies are fairly specific for the diagnosis of NMO. The NMO immunoglobulin G binding pattern was also assessed in the current study (6); serum from four of the 29 patients with IIH showed cytoplasmic cerebellar Purkinje cell immunoreactivity and four showed diffuse neuronal nuclear staining that is typically associated with NMO. The significance of this pattern is uncertain, as it was also present in serum from patients with encephalitis and other inflammatory/demyelinating disorders. Is it possible that inflammation has a role in the development or progression of IIH? An inflammatory component has been postulated in the pathogenesis of IIH because of the high association of IIH and obesity, but a definitive correlation has not yet been proven (8).

Although the aquaporin family is an attractive candidate to explain the mechanism of IIH, the lack of antibody production in IIH patients is not entirely surprising. Perhaps AQP-4 is not the correct target to study; it is closely associated with choroid plexus and ventricular ependymal cells but it is not clear that either of these structures is dysfunctional in IIH. To date, there is no convincing evidence that IIH is accompanied by either cellular or extracellular edema. Early pathological surgical specimens from IIH patients that were examined in the 1950s and reported to show edema were later re-examined using modern techniques and found to have fixation artifact rather than true edema (9,10). More recent neuropathological specimens from the brains of patients with active IIH showed no evidence of edema (10). Observations ranging from Walter Dandy’s description in 1937 (11) to more recent studies using MRI and venous manometry suggest that cerebral venous hypertension is the final common pathway for IIH and the pseudotumor cerebri syndromes, rather than increased pressure in the brain parenchyma (edema) or intraventricular spaces (hydrocephalus). AQP-4 studies have not provided the answers but raise interesting questions regarding the precipitating factor(s) and underlying pathology required for developing IIH.