Manometry combined with cervical puncture in idiopathic intracranial hypertension

We thank Quattrone et al. for their comments about our article on manometry combined with cervical puncture in idiopathic intracranial hypertension. [1] Our patients all had idiopathic intracranial hypertension (IIH) which is a different condition to isolated intracranial hypertension without papilledema. There remains however a valid question as to the cause of the functional obstruction to cerebral venous outflow at the level of the transverse sinuses (TS). In our first paper we considered mural thrombus, in some cases forming on arachnoidal granulations, the likely cause. [2] However in IIH where cerebral venography typically shows smooth bilateral tapered narrowing rather than focal sessile lesions, we now think that the changes are all due to stretching of the walls of the TS, given the immediate relief of elevated venous sinus pressures by lowering intracranial pressure with cervical puncture.

The case report of successful stenting of one TS in a patient with IIH, [5] confirmed our findings of a pressure drop across the TS with raised pressures at the level of the torcula. The stent in the TS abolished the venous hypertension and after 3 weeks the opening pressure at lumbar puncture was normal. The stent opened the lumen of the TS, dropped the pressure in the superior sagittal sinus and allowed passive absorption of cerebrospinal fluid (CSF), thereby lowering the intracranial pressure. Lowering the intracranial pressure by C1-2 puncture has the same effect, which leads us to consider that the TS stenosis is caused by extrinsic pressure on the walls of the TS rather than by intraluminal processes such as mural thrombus. This argument begs the question as to what flattens the walls of the TS. We believe there is some process in IIH involving the arachnoidal granulations which impairs CSF absorption and initiates the rise in intracranial pressure.

Magnetic resonance venography is recognized to be sensitive to altered flow, but less accurate in assessing the anatomy of the venous sinuses. We consider conventional venography to be superior to MRV in displaying anatomical detail and at this stage the smooth bilateral narrowing of the TS in IIH is unlikely to be due to mural thrombus.

Dr Lee has drawn attention to the uncertain place of MRV in IIH. In typical cases of IIH we found the MRV lacked adequate definition in the TS and flow voids could easily be misinterpreted as sinus thrombosis. MRV and conventional cerebral venograms were performed in patients with IIH and most patients showed apparent narrowing of the TS on MRV however T2 and T1 weighted MRI excluded thromboses and arachnoidal granulations. We did not perform pre- and post- cervical puncture MRV but would be surprised if this technique would be helpful because conventional venography did not shown striking changes despite lowering of pressures in the superior sagittal sinus and proximal TS after cervical puncture.

Drs. Higgins, Nicholas and Professor Pickard question our conclusion that the venous outflow obstruction in IIH is due to partial collapse of the walls of the transverse sinuses from raised intracranial pressure as a secondary phenomonen. By lowering intracranial pressure we found the pressure gradient in the transverse sinuses largely disappeared. It is reasonable to assume that the cross-sectional area of the TS increased when the extrinsic compression was reduced. This would have the same effect as enlarging the internal dimensions by placement of a stent.

Conventional venograms in most instances of IIH show smooth tapered narrowing of the TS bilaterally. Although we originally felt this could be due to mural thrombus, it seems unlikely that such a symmetrical appearance could result from acute, organized or recanalised clot. If the process were some form of sclerosis of the TS one would not expect lowering the intracranial pressure to have any significant effect on the venous hypertension.

Our hypothesis requires a subclinical elevation of intracranial pressure, possibly due to some change in permeability of arachnoidal villi to CSF. This could be produced by an as yet unidentified hormone in overweight females or by drugs such as minocycline. Over a period of a few months in susceptible individuals the raised intracranial pressure would start to flatten the walls of the TS and push up the venous pressure in the SSS and proximal TS, further impairing CSF absorption and sharply elevating intracranial pressure. Stenting one TS would allow venous pressure to fall but would the intracranial pressure fall to normal as happened in the case reported by Higgins et al? [3] The early value of stenting the TS in IIH has been confirmed in a further four cases [4] and the procedure offers a new treatment option. The follow-up results are awaited, however these cases suggest that cerebral venography and manometry should be done routinely in IIH.

References:

1. King JO, Mitchell PR, Thomson KR, Tress BM. Manometry combined with cervical puncture in idiopathic intracranial hypertension. Neurology 2002;58:26-30.

2. King JO, Mitchell PR, Thomson KR, Tress BM. Cerebral venography and manometry in idiopathic intracranial hypertension. Neurology 1995;45:2224-2228

3. Higgins JNP, Owler BK, Cousins C, Pickhard JD. Venous sinus stenting for refractory benign intracranial hypertension. Lancet 2002;359:228-230.

4. Owler BK, Parker G, Dunn V, Halmagyi GM, McDowell D. Besser M. Pseudotumor cerebri: Treatment with venous sinus stenting. Australian and New Zealand Journal of Surgery 2002;72(Suppl):A62.

5. Higgins JN, Owler BK, Cousins C, et al. Venous sinus stenting for refractory benign intracranial hypertension. Lancet 2002;359:228-230.

We read with interest the article by King et al. [1] We agree that there is a relationship between intracranial pressure and sinus venous pressure. We do, however, have concerns over the authors' conclusion that increased cerebral venous pressure found in most patients with isolated intracranial hypertension (IIH) is due to a functional obstruction (collapse of the walls) of the transverse sinuses (TS) by raised intracranial pressure and not due to a primary obstructive process in the TS.

We recently demonstrated on MR venography (MRV) that a number of subjects with IIH with or without papilledema had flowing abnormalities of both transverse sinuses highly suggestive of cerebral venous thrombosis. [2,3] It is noteworthy that the flowing abnormalities seen on MRV in IIH occurred mainly in the distal portion of the TS. [2, 3] This observation was confirmed by King et al. [1] who showed a pressure gradient between the proximal and distal part of the TS in patients with IIH. These findings suggest there must be some anatomical reason that makes the distal part of the transverse sinus the preferential site for developing an obstructive process. Since arachnoidal granulations typically occur in the distal portion of the TS, [4] it is reasonable to hypothesize that in some individuals large arachnoid granulations could produce relative luminal compromise and lead to a disturbed flow with a pressure gradient or an increased risk of venous thrombosis. Taken together, these data indicate that obstruction of the distal portion of one or both TS, which occurs in many patients with IIH, [2, 3, 5] is probably due to an intraluminal process (prominent arachnoidal granulations, thrombus forming on arachnoidal granulations, or venous thrombosis) rather than to an extrinsic cause (i.e. raised intracranial pressure), which should collapse the walls of the entire TS and not just the walls of the distal portion. Consistent with this hypothesis, a recent paper [5] described a patient with IIH who showed, on venography and manometry, a partial obstruction of the distal portion of both TS with raised pressure proximal to the obstruction. Dilatation of one of the transverse sinuses with a stent reduced both the pressure gradient and CSF opening pressure with striking symptomatic improvement, suggesting a causal relationship between venous outflow obstruction and IIH. Finally, we agree with King et al. [1] that raised intracranial pressure could make the obstruction worse by collapsing the walls of the sinus, thus further exacerbating both venous hypertension and CSF pressure.

References:

1. King JO, Mitchell PJ, Thomson KR, et al. Manometry combined with cervical puncture in idiopathic intracranial hypertension. Neurology 2002;58:26-30.

2. Quattrone A, Gambardella A, Carbone AM, et al. A hypofibrinolytic state in overweight patients with cerebral venous thrombosis and isolated intracranila hypertension. J. Neurol 1999;246:1086-1089.

3. Quattrone A, Bono F, Oliveri RL, et al. Cerebral venous thrombosis and isolated intracranial hypertension without papilledema in CDH. Neurology 2001;57:31-36.

4. Leach JL, Jones BV, Tomsick TA, et al. Normal appearance of arachnoid granulations on contrast-enhanced CT and MR of the brain : differentiation from dural sinus disease. AJNR Am J Neuroradiol 1996;17:1523-1532.

5. Higgins JN, Owler BK, Cousins C, et al. Venous sinus stenting for refractory benign intracranial hypertension. Lancet 2002; 359:228-230.

We have read with great interest the paper by King et al. [1] but would disagree with their conclusions and with the enthusiasm with which they were greeted in the accompanying editorial. [2] We also have found pressure gradients across stenoses in the lateral sinuses in patients with apparently idiopathic intracranial hypertension. Furthermore we have dilated one of these stenoses with a stent thereby reducing the pressure gradient which resulted in almost complete resolution of symptoms. [3] Hence we would strongly support the original King et al. hypothesis that venous outflow obstruction is the primary cause of idiopathic intracranial hypotension at least in some cases.

There is no doubt that the transverse sinuses may collapse in response to raised intracranial pressure and that in this situation pressure gradients will be detected along them. Moreover, these gradients will resolve if intracranial pressure is reduced. [4] Equally, there is narrowing or occlusion of the intracranial venous sinuses can cause no doubt that raised intracranial pressure. With reference to this paper, it is important to reiterate that raised intracranial pressure in patients with unequivocal cerebral venous thrombosis is relieved by CSF diversion. [1] Where there is stenosis or thrombosis of the sagittal or transverse sinuses, a secondary rise in cerebral venous pressure must be accompanied by an even greater rise in CSF pressure if CSF absorption is to continue. Where cerebral venous pressure or intracranial pressure is raised, there will be an autoregulatory vasodilatation to maintain cerebral blood flow constant in the otherwise healthy brain. If CSF is diverted and intracranial pressure reduced, such cerebral vasodilatation will reverse very likely with a fall in cerebral venous pressure (see 5 for a discussion of pressure distribution along the cerebral venous outflow). A fall in cerebral venous pressure in response to a withdrawal of cerebrospinal fluid, therefore, does not exclude venous outflow obstruction as the cause of raised intracranial pressure.

Another question, not explored, is any secondary effect of raised intracranial pressure on the venous sinuses when intracranial hypertension is due to venous sinus obstructions, especially at points where the sinuses are known to be compressible. In the cranial cavity the raised intracranial pressure itself will act as a force on the sinus wall resisting any expansion that might mitigate the obstructing lesion. Reducing intracranial pressure by removing CSF would alter this transmural gradient and might allow the sinus to expand. If this expansion were sufficient, then the intrasinus pressure gradients across the stenotic lesions would fall acutely in the manner recorded by King et al. One can speculate further that following this acute response equilibrium will be restored at a rate depending on the degree of "primary" sinus stenosis and the compliance of the sinus wall under strain from rising intracranial pressure - paralleling the clinical effects of CSF withdrawal.

Whatever the mechanism operating here, we suggest that if pressure gradients along the transverse sinuses were always secondary to idiopathic intracranial hypertension then dilating one of these stenotic areas would not have caused the intracranial pressure to fall, nor effected the clinical improvement we observed in our reported case. We would applaud King et al. pioneering observations in 1995, but are anxious that a misinterpretation of his most recent results, unquestioned in your editorial, will stall a line of research that may yet revolutionize our understanding of this condition.

References:

1. King JO, Mitchell PJ, Thomson KR, Tress BM. Manometry combined with cervical puncture in idiopathic intracranial hypertension. Neurology 2002; 58:26-30.

2. Corbett JJ, Digre K. Idiopathic intracranial hypertension; an answer to, "the chicken or the egg?" Neurology 2002; 58: 5-6.

3. Higgins JNP, Owler BK, Cousins C, Pickard JD. Venous sinus stenting for refractory benign intracranial hypertension. Lancet 2002; 359: 228-230.

4. Osterholm J. Reaction of the cerebral venous system to acute intracranial hypertension. Journal of Neurosurgery 1970. 32;654-659.

5. Piechnik SK, Czosnyka M, Richards HK, Whitfield PC, Pickard JD. Cerebral venous blood outflow: a theoretical model based on laboratory simulation. Neurosurgery 2001; 49: 1214-1223.

King et al. [1] reported the results of cerebral venous sinus manometry and cervical puncture in idiopathic intracranial hypertension (IIH). They reported cerebral venous sinus hypertension in the superior sagittal and proximal transverse sinuses that was reversed by reducing intracranial pressure. They concluded that the elevated intracranial pressure and not the other way around caused compression of the dural walls of the transverse sinus. Lee and Brazis [2] previously performed a prospective study to evaluate for the presence or absence of dural sinus thrombosis using magnetic resonance (MR) imaging and MR venography of the brain in 22 consecutive young, overweight women patients with typical IIH. None of the 22 MR imaging and MR venography studies showed venous sinus thrombosis and they concluded that MR venography might not add significantly to the evaluation of typical IIH. I still order cranial MRI/MRV however in atypical IIH cases (e.g., male, thin or elderly patients).

I have been impressed however by the number of MRV studies in IIH that have shown findings that we have in the past interpreted as being suggestive of venous sinus stenosis or flow related turbulence at the level of the distal transverse sinus. Some of these patients underwent standard catheter venography and a few were even considered for possible stenting. Thus, the MRV studies in these patients actually confounded the evaluation of their IIH. My questions for the authors are as follows:

1. Did any of their patients undergo MRV in addition to cranial MRI and if so did these MRVs show anything that might have been misinterpreted as venous sinus thrombosis or obstruction at the distal transverse sinus?

2. Do the authors believe that performing an MRV in typical IIH might be misleading in the management of typical IIH in cases with flow related abnormalities (but not true obstruction) at the distal transverse sinus?

3. Would pre- and post-lumbar puncture MRV be able to demonstrate the reversibility of flow related signal abnormalities at this level.

This work is fascinating and I commend the authors for their efforts in this area.

References:

1. King JO, Mitchell PJ, Thompson KR, Tress BM. Manometry combined with cervical puncture in idiopathic intracranial hypertension. Neurology 2002;58:26-30.

2. Lee AG, Brazis PW. Magnetic resonance venography in idiopathic pseudotumor cerebri. J Neuro-ophthalmology 2000;20: 12-13.