Abstract
Dear Sir We read with great interest the paper “Abnormal pressure waves in headache sufferers with bilateral transverse sinus stenosis” by Bono et al. (1). The study replicates the findings of a previous study by Torbey et al. entitled “Utility of CSF pressure monitoring to identify idiopathic intracranial hypertension without papilledema in patients with chronic daily headache” (2), providing an independent confirmation that the diagnosis of idiopathic intracranial hypertension without papilledema (IIHWOP) may be overlooked even in the case of direct opening pressure (OP) measurement by lumbar puncture (LP). In fact, in most such cases the intracranial hypertensive status presents large intra-day fluctuations so that prolonged intracranial pressure (ICP) monitoring is crucial in recognizing the condition. However, the study design of Bono et al. presents a number of important differences from that of Torbey, four of which are of major relevance:
The study is not limited to chronic/transformed migraine (CM) but is extended to cases presenting with chronic tension type headache (CTTH) in comparable proportions and to a minority of “other forms”. Cases are further selected on the basis of a poor response to treatments. Rather than a specific painful condition, the main variables under study are therefore the high frequency of attacks and the resistance to treatments of different primary headache pain. The period of ICP monitoring is reduced from 24–48 hours to 1 hour. This represents a much more reasonable time span and might promote the use of this valuable procedure in clinical practice. The experimental design includes a magnetic resonance venography (MRV) performed in all patients before the diagnostic LP. This allows evaluation of the possible correlation between ICP values and cerebral venous flow disturbances. The study recruits a large series of consecutive cases of chronic headache with minimal response to treatments; conversely in Torbey’s paper cases are selected. This means that the findings of Bono et al. may be generalized to clinical populations of unresponsive or almost unresponsive chronic headache sufferers. It is noteworthy that this kind of patient represents a large percentage of those referring to specialized headache clinics and greatly contributes to the overall burden of primary headaches, both from a clinical and a socioeconomic point of view.
Briefly, the authors’ findings may be summarized as follows:
Forty-eight out of 98 (48.9%) of the chronic headache sufferers studied show bilateral transverse sinus stenosis (BTSS). This percentage is significantly higher than expected: BTSS has been found by the same research group in 50 out of 217 (23.0%) of neurological patients without persistent headache or any other signs or symptoms of raised ICP (3). In another paper “sinus flow gaps or aplasias” are described in approximately 24% of a series of 105 individuals with normal MR of the brain (4). Using all the data reported in both Bono’s papers (1,3) (48 BTSS positive on 50 BTSS negative among chronic headache group versus 50 BTSS positive on 167 BTSS negative among control group) we can compute a p-value of 7 × 10−6 using Fisher’s exact test and an odds ratio of 3.2064 (95% confidence interval 1.9319 to 5.3217); both tests have a power of 99.49%. IIHWOP, defined as OP > 200 mmH2O without any evidence of intracranial pathology, has been found in 18 out of 98 (18.3%) of chronic headache sufferers by a single-spot opening pressure (OP) measurement and in an adjunctive 26 out of 98 (26.5%) after 1-hour ICP monitoring. This leads to a cumulative 44.8% (44 out of 98) prevalence of IIHWOP in this consecutive chronic headache sufferer series. This finding is extraordinarily elevated compared to the very low prevalence attributed to IIHWOP, considered at present an infrequent variant of IIH with papilledema, a disease with estimated incidence of 1–19/100,000 (5,6). The data suggest that 1-hour ICP monitoring is a more sensitive method than OP measurement. Considering the 48 headache suffers with BTSS (1), 18 (37.5%) had elevated CSF opening pressure but 1-hour ICP monitoring revealed waves of elevated CSF pressure also in 26 out of 30 patients who had normal OP. This means that 86.6% of the subjects without elevated CSF pressure at single-spot OP can be reclassified as IIHWOP after 1 hour of ICP monitoring. A strong bidirectional correlation, which leads to an almost complete overlap, links BTSS and IIHWOP. BTTS has been found in all (100%) chronic headache patients with intracranial hypertension vs. only 7.4% of those with normal ICP. On the other hand, IIHWOP has been found in 44 out of 48 (91.6%) BTSS carriers. Conversely, none out of 50 patients BTSS-free showed an OP > 200 mmH2O. In a previous work by the same group (7), none of 70 recurrent headache patients without BTSS at MRV showed an OP > 200 mmH2O. A 2–4 week improvement of headache intensity and frequency follows the diagnostic LP “in the majority of the patients with elevated intracranial pressure”. A similar 2–4 weeks transient improvement after LP has been previously described in a series of CTTH patients with BTSS-associated IIHWOP (8).
The authors propose “the accuracy of short-term CSF pressure monitoring through a lumbar needle in estimating CSF pressure” as the main finding of the study. We would suggest a more extensive interpretation of their results. In fact, their data further demonstrate that almost half of consecutive clinical series of unresponsive chronic headache sufferers harbour a BTSS-associated IIHWOP, strongly suggesting, at the same time, that raised ICP is responsible for the high frequency and unresponsiveness of pain in these subjects. Although compelling, the practical utility of these findings should remain uncertain until they can be independently reproduced and verified.
Based on their observations, the authors classify these patients as having a “secondary headache and mild IIHWOP”. We would like to propose an alternative interpretation of the data. Most cases of this series received either a CM or a CTTH diagnosis according to the IHS 2004 criteria (9). This implies that a previous history of episodic headaches was identifiable in all CTTH and probably in most of CM patients of this series. Moreover the authors observed the improvement of frequency and intensity of attacks after LP, not their disappearance, as is expected after removal of the causative agent in a secondary headache (9). Rather than a confirmation of the secondary nature of headache in these patients, this could reflect the return to an episodic pattern of attacks in primary headache predisposed subjects. The following evidence further sustains this view.
In a previous paper Bono et al. (3) described BTSS in 50 out of 217 (23.0%) patients undergoing LP for various neurological conditions with the exclusion of signs and symptoms of intracranial hypertension, including “persistent daily headache”. In 24 of such cases (11% of the sample) the reported OP was >200 mmH2O, but this data could be greatly underestimated because LP was not followed by prolonged ICP monitoring. Conversely, no BTSS-free patients showed an elevated OP. Interestingly, all 50 BTSS carriers complained of episodic headaches and some reported a transient reduction of headache after performing the LP. These data indicate that IIHWOP could be much more prevalent than thought in the general population and that it can run almost asymptomatically in most affected individuals. IIH without headache has been described in a woman without personal or familial history of migraine, and headache has been reported as remitting in the course of pregnancy, a strong migraine protective factor (10), suggesting that a chronic headache presentation of IIH may require a migraneous background. The absence of both BTSS and raised ICP in more than half of unresponsive chronic headache sufferers recruited in the study (1) and the high prevalence of BTSS-associated IIHWOP found in neurological patients without chronic headache (3) indicate that IIHWOP is neither a necessary nor a sufficient factor for the development of frequent headaches. Consequently, taken together, all the above reported data strongly suggest that IIHWOP is a powerful and very common, although largely unrecognized, risk factor for primary headache pain progression and refractoriness. Because of the modifiable nature of IIHWOP this observation may have a crucial impact in the clinical management of many chronic headache patients. We have recently presented a number of indirect evidences supporting this view (11).
Conclusions
Although the authors do not discuss their results in this regard, their recent findings (1,3) raise the following hypotheses with a potential high clinical impact that need to be urgently tested in adequately designed experimental studies. Still, it must be stated that until these data are independently confirmed and verified any change in the current clinical practice standards is unjustified. Future studies should also define the clinical outcome of IIHWOP treatment on chronic headache as well as which patients should be studied and how.
Population prevalence of IIHWOP is much higher than believed. Cases with papilledema might represent only the tip of the iceberg (11). A comorbid IIHWOP represents a very common but largely underestimated modifiable risk factor for chronification and refractoriness of pain in primary episodic headache subjects. It could represent the pathogenetic key leading to progression of pain in about half of primary chronic headache subjects with minimal response to treatments referred to specialized headache clinics (11). BTSS at MRV are sensitive and specific radiological predictors of raised ICP, even in individuals without signs or symptoms of intracranial hypertension. Cerebral venous outflow disturbances play a crucial pathogenetic role in IIH with and without papilledema. We have recently presented a comprehensive pathogenetic model of IIH focused on the role of sinus venous stenosis (12,13) supported by independent, clinical data-based mathematic modelling studies (14).
Finally, in agreement with the authors’ own final findings (1), the persistence of BTSS in medically treated IIH patients despite normalization of OP at LP does not imply that BTSS must play a marginal role in IIH pathogenesis, as they conclude in a previous work on a small series of IIH cases (15). In fact, the possible persistence of a fluctuating intracranial hypertension might have been overlooked by a single-spot OP measurement (16). Therefore, a crucial pathogenetic role of BTSS in IIH cannot be ruled out on the bases of the presented data.