Is higher choroid plexus ‘load’ an aetiologic factor in idiopathic intracranial hypertension? A clinico-imaging morphometric correlative study

Introduction

Idiopathic intracranial hypertension (IIH) is a condition of raised intracranial pressure (ICP) without clinical, laboratory or radiological evidence of intracranial pathology. It is a relatively frequent disease occurring in two in 100,000 of the general population (1). The syndrome is seen in all demographics but occurs most commonly in obese young women of childbearing age. In 1893 Heinrich Quincke first described a clinical syndrome resembling IIH, which he called “meningitis serosa”, in patients with increased ICP without a brain tumour (2,3). In 1937 Walter Dandy defined the diagnostic criteria of the syndrome we now refer to as IIH. The ‘Dandy Criteria’ described the presence of increased ICP, normal cerebrospinal fluid (CSF) findings and no sign of a brain tumour on ventriculography (1). JL Smith subsequently revised these criteria in 1985, incorporating CSF biochemical and imaging criteria, into the ‘Modified Dandy criteria’, which are currently used for diagnosing IIH (4).

The pathophysiology of IIH remains speculative and controversial. There are currently three major theories of the cause of IIH: increased CSF production, increased resistance to CSF absorption, and increased venous sinus pressure (5–9). Despite decades of research, the evidence for all of these proposed mechanisms remains inconclusive. It is likely that disordered CSF hydrodynamics plays a central role in the pathogenesis of IIH, and it is possible that increased CSF formation may at least play a part.

The choroid plexus (CP) produces CSF, and conditions which result in enlarged CP, such as papillomas and CP hyperplasia, are known to be associated with increased CSF production (10). We therefore hypothesised that if CSF hypersecretion is involved in the pathogenesis of IIH, this may be associated with an appreciable macroscopic enlargement of the CP. We carried out a retrospective review of patients with IIH and normal matched control patients in order to examine whether we could quantify a macroscopic difference in the size of the CP on neuroradiological imaging. OPEN IN VIEWER

Materials and methods

All imaging was performed during routine neuroradiological work-up of patients based on clinical indications; no prospective imaging of normal volunteers was performed in this study. As such, and according to guidelines of the revised Declaration of Helsinki, this study reported the retrospective analysis of data (anonymised images) obtained from ‘routine sources’ where consent of individual patients and ethical approval for research analysis was therefore not considered necessary (11,12).

First, we retrospectively reviewed the imaging studies of 50 patients who all met the modified Dandy criteria for the diagnosis of IIH. All patients had a documented CSF pressure of 25 cm H₂O or above, appropriate clinical symptoms and no other significant findings on neuroimaging. We reviewed the axial images of head computed tomography (CT) venograms of these patients, all of which had been performed within six weeks of a documented lumbar puncture and CSF pressure measurement. CT venograms with axial contiguous slices of 1 mm were performed on a Siemens SOMATOM Sensation® 16 CT scanner (Munich, Germany) after the intravenous injection of 150  ml non-ionic contrast medium (iopamidol, 300  mg iodine per ml). Two radiologists independently analysed all the anonymised images. An arbitrary simplified measure of the overall size or ‘load’ of each patient’s CP was calculated by adding together the measured maximum axial cross-sectional areas of CP at three separate levels and locations: the unilateral larger CP within either body of the lateral ventricles, the unilateral larger ChP glomus within either trigone of the lateral ventricles, and the CP at the fourth ventricle (Figure 1). The maximum cross-sectional area of the CP at these locations was measured in mm² using electronic calipers on a GE Healthcare Centricity™ PACS-IW review workstation (Little Chalfont, Buckinghamshire, UK).

We compared these results with those from 50 age- and sex-matched controls. For this, we retrospectively reviewed standard post-contrast head CT scans of patients who demonstrated normal brain imaging findings, or near-normal findings that were not relevant or likely to affect the size of their CP. CT head scans were performed using 4.5 mm thick axial contiguous slices on a Siemens SOMATOM Sensation® 16 CT scanner, and we obtained enhanced images after the intravenous injection of 50  ml of non-ionic contrast medium (iopamidol, 300  mg iodine per ml). We measured the same three CP maximum axial cross-sectional areas on these CT images. We also measured the Evans Index of ventricular size to exclude the presence of ventriculomegaly on both sets of patient and control images. We used a Student’s t test for independent samples to statistically compare IIH and control groups, with significance set at p  <  0.05. The effect of the variable of ICP value was also tested on the dependent variable (area of CP) using regression analysis.

Results

Patients with IIH had a mean age of 35 years; 98% of the patients were female and 2% were male. The 50 matched controls had a mean age of 38 years; 97% were female and 3% were male. None of the CT studies demonstrated ventriculomegaly and all had an Evans Index of less than <0.3. There was no significant difference in the total measured maximum axial cross-sectional areas of the CP between IIH patients, who had a mean CP area of 183 mm², and those of controls, with a mean CP area of 178 mm². In particular, the mean largest diameters of the glomus of the CP were 6.5 mm and 6 mm in the two groups, respectively. The range of CSF pressures was 25 to 50, with a mean value of 30 cm H₂O. There was no statistical correlation between the ‘load’ of CP and the degree of raised ICP (R²< 0.02). Overall, there was no appreciable macroscopic increase in size of the CP in patients with IIH when compared to normal individuals.

Discussion

To date, the exact aetiology of IIH remains controversial although the major current theories of the pathogenesis of IIH all involve disordered CSF hydrodynamics, either through the increased formation of CSF or through the reduced drainage of CSF secondary to increased resistance. One of the most favoured current theories is that increased resistance to CSF outflow occurs secondary to the presence of a primary venous sinus hypertension. However, conclusive evidence that venous hypertension is the primary causative factor in all cases is still lacking. Venous hypertension, demonstrated in IIH cases, is thought by many to be the result rather than the cause of IIH. Recently there have been attempts to examine the complex hormonal and molecular pathways that affect the biology of the CP that in turn affect CSF production; however, these studies remain inconclusive. We will briefly discuss the current evidence that CSF hypersecretion may at least play a part in the pathogenesis of IIH.

To date, much of the evidence for CSF hypersecretion as a causative factor in IIH remains anecdotal, controversial or limited. Indirect supporting evidence for the role of increased CSF formation includes the observation that there are often higher CSF volumes on lumbar drainage of IIH patients. There is also a recognised aetiological association with agents known to cause a disturbance of CSF hydrodynamics (13). Hypervitaminosis A is known to cause intracranial hypertension and an IIH-like syndrome. Moreover, inferred evidence comes from reported symptomatic treatment of the condition by medical therapies that act to reduce CSF secretion such as diuretics and carbonic anhydrase inhibitors (acetazolamide), although their efficacy remains disputed (13,14). Importantly, however, CSF is not a simple plasma ultrafiltrate. CSF secretion is a complex physiological and molecular process involving several ion transport proteins that move Na⁺, Cl^– and HCO3^– from the blood to the brain ventricles, thus creating an osmotic gradient which drives the secretion of H₂O (15). Based on our results we speculate that, whether or not CSF overproduction is in reality present in IIH, it is more likely if CSF hypersecretion were a contributory factor that this might require greater efficiency of existing molecular transport systems (perhaps by increased up-regulation of transport proteins) within a baseline number of CP epithelial cells (assuming that this does not result per se in individual cellular hypertrophy and consequent CP enlargement), rather than a hyperplasia of epithelial cells in the villi of the CP manifesting as an overall exaggerated load of CP throughout the ventricular system. The potential interplay of these factors and mechanisms could be fruitful areas of future investigations in IIH research.

There have been various studies that have attempted to objectively quantify the volumes of CSF production in IIH patients and normal controls, although the results of these have also proved inconclusive. Invasive infusion or perfusion techniques have been used to measure the production rate of CSF in patients. In part due to their invasive nature these studies have been limited in number, but the results principally demonstrate no significant differences in the production rate of CSF between patients and controls, or even a reduced rate. Attempts at measuring CSF production rates noninvasively, by recording CSF flow through the cerebral aqueduct using magnetic resonance imaging (MRI), have provided highly variable results that overall have also not supported the view that CSF hypersecretion is important in IIH. Accordingly, although most investigators have not found CSF hypersecretion in IIH, a small number of studies have shown increased CSF production rates (16,17). It is therefore plausible that in some patients at least, increased CSF production may contribute to the development of IIH.

One of the principal arguments made against excessive CSF production being an important cause of IIH is the lack of the presence of hydrocephalus in IIH patients. Most conditions known to be associated with the overproduction of CSF normally result in an increase in ventricular size caused by the increased pressure gradient between the CSF in the ventricles and the subarachnoid space. This relationship has been demonstrated in animal studies during intraventricular infusion studies. The well-known in vivo disease models of CSF hypersection, CP papilloma and CP hyperplasia, both result in hydrocephalus. The normal or decreased ventricular size found in IIH patients has been said to suggest that they do not have increased CSF production. However, some authors have attempted to explain the absence of ventricular dilatation in the presence of raised ICP in patients with IIH as being due to the simultaneous equalisation of the pressure of the brain interstitial fluid with the high ICP of the CSF. One mechanism by which this could happen is the presence of brain oedema. There has been some indirect evidence of the presence of brain oedema in patients with IIH, provided by early MRI studies, which showed increased water content and water diffusion in subcortical white matter. However, more recent studies using better MRI techniques have not been able to reproduce these earlier findings (18,19).

Iencean suggested an alternative hypothesis to help explain the paradox between raised ICP and small ventricles. He suggested that in IIH there might be a rapid circulation and absorption of CSF and brain interstitial fluid in the same ratio. IIH therefore might occur through the simultaneous hypersecretion of CSF and of cerebral interstitial fluid in the presence of rapid circulation and absorption of these fluids based on a fast cerebral blood flow (6,7). However, the evidence for this idea is also limited. Mathew et al. have reported some evidence of increased cerebral blood volume (CBV) in IIH, using intra-carotid tracer injection; however, the patients in this study were all anaesthetized (20). No significant differences in regional CBV in IIH compared with normal subjects were found in a subsequent study by Brooks et al. using positron emission tomography (20,21).

Conclusion

The exact aetiology and pathophysiology of IIH remain disputed. There is some indication that IIH may result from a disorder of CSF hydrodynamics, which might still in theory include some degree of CSF hypersecretion. Although the evidence that CSF hypersecretion per se plays a part in IIH is limited, disputed, or often anecdotal, our study adds credence to this by demonstrating that even if increased CSF production plays any part in the aetiology of IIH, this is not reflected in a demonstrable macroscopic hypertrophy of the CP or an increase in its ‘load’ on cross-sectional brain imaging.