Brain Advance Access originally published online on March 19, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Brain, Vol. 127, No. 5, 965-972, 2004
© 2004 Guarantors of Brain
doi: 10.1093/brain/awh131
Pattern of white matter regional cerebral blood flow and autoregulation in normal pressure hydrocephalus
Academic Neurosurgery Unit, Addenbrooke’s Hospital and Wolfson Brain Imaging Centre, University of Cambridge, Cambridge, UK
Correspondence to: Shahan Momjian, Academic Neurosurgery Unit, Box 167, Addenbrooke’s Hospital, Cambridge CB2 2QQ, UKE-mail: sm471@medschl.cam.ac.uk
 |
Summary |
|---|
 Top Summary IntroductionMaterial and methodsResultsDiscussionReferences |
|---|
The mean cerebral blood flow (CBF) has generally been demonstrated to be lower in normal pressure hydrocephalus (NPH) than in normal controls. We investigated the distribution of the regional peri- and paraventricular white matter CBF (WM CBF) in NPH at baseline and during a controlled rise in intracranial pressure (ICP). Twelve patients with idiopathic NPH (mean age 69 years) underwent a CSF infusion study. CBF was measured by H215O PET at baseline and then during the steady-state plateau of raised ICP. The PET images were co-registered and resliced to 3D structural T1-weighted MRIs. Ten healthy normal volunteers served as control subjects for baseline CBF determination only. Profiles of the regional distribution of the baseline WM CBF and of the percentage change in WM CBF as a function of distance from the ventricles were plotted. The global mean baseline CBF in patients (28.4 ± 5.2 ml/100 ml/min) was lower than in the control subjects (33 ± 5.4 ml/100 ml/min) (P < 0.005). In patients, the profile of the regional WM CBF at baseline showed an increase with distance from the ventricles (P < 0.0001), with a maximal reduction adjacent to the ventricles and progressive normalization with distance, whereas in controls no relationship was apparent (P = 0.0748). In 10 patients, the rise in ICP during the infusion produced a fall in cerebral perfusion pressure (CPP) and a significant decrease of the global mean CBF from 27.6 ± 3.1 to 24.5 ± 2.9 ml/100 ml/min (P < 0.0001). The profile of the percentage changes in regional WM CBF in patients showed a U-shaped relationship with distance from the ventricles (P = 0.0007), with a maximal decrease skewed on the side of the lateral ventricles at around a mean distance of 9 mm. The WM CBF is reduced in NPH, with an abnormal gradient from the lateral ventricles towards the subcortical WM. An excessive decrease in CBF is brought about by reductions in CPP and appears to be maximal in the paraventricular watershed region. These results are discussed in the light of previous hypotheses concerning the aetiology of periventricular CBF reduction in NPH.
Key Words: normal pressure hydrocephalus; cerebral blood flow; autoregulation; white matter
Abbreviations: ABP = arterial blood pressure; CBF = cerebral blood flow; CPP = cerebral perfusion pressure; ICP = intracranial pressure; MCA = middle cerebral artery; NPH = normal pressure hydrocephalus; rCBF = regional CBF; SRAR = static rate of autoregulation; WM = white matter
Received September 4, 2003. Revised December 18, 2003. Accepted December 22, 2003.
|
Introduction |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
In normal pressure hydrocephalus (NPH), the physiological alterations related to the ventricular dilatation that lead to the Hakim and Adams clinical syndrome remain unclear, notably because the physiological abnormalities that have been noted so far are inconsistent and inconclusive (Mori and Mima, 1998
). This is particularly the case for the underlying mechanisms and the significance of the changes of cerebral blood flow (CBF) occurring in NPH, in which the global mean CBF has generally been demonstrated to be lower than in normal controls and where a postoperative increase of CBF has often been demonstrated after successful shunting (Owler and Pickard, 2001). A CBF increase has also often been revealed after CSF withdrawal by lumbar puncture in patients ultimately responding to shunting (Owler and Pickard, 2001; Mori et al., 2002; Hertel et al., 2003).
The reduction in baseline CBF is often more pronounced in the frontal region and appears to predominate in the white matter (WM) (Owler and Pickard, 2001). In some cases, although the mean global CBF was within the normal range, the regional CBF (rCBF) pattern was abnormal compared with normal controls, with a slightly reduced rCBF in the central WM (centrum semiovale) and with an enlargement of the area of the region of subcortical periventricular low CBF (Waldemar et al., 1993). Therefore, a possible primary dysfunction of the axons in the WM rather than a primary dysfunction of the grey matter might have a dominating role in the generation of the clinical syndrome.
However, both the relationship between the reduction in baseline CBF and the initial clinical impairment and the relationship between the magnitude of the postoperative CBF increase and the degree of clinical improvement are controversial. Therefore the correlation between the clinical syndrome and global/regional CBF is not established. The question whether the CBF reduction is the cause or the consequence of the neuronal dysfunction remains unresolved.
The present study was designed to assess the distribution of the regional periventricular WM CBF using H215O PET co-registered on MRI anatomy. The relationship between the WM CBF response to a controlled rise in intracranial pressure (ICP) [and hence a fall in cerebral perfusion pressure (CPP)] induced by an infusion of CSF was also studied.
|
Material and methods |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Patient population and controls
Twelve patients (eight women and four men) ranging in age from 46 to 76 years (mean age 69 years) were prospectively included in the present study. These patients were diagnosed as suffering from normal pressure hydrocephalus according to the following criteria. All patients presented with a Hakim and Adams clinical triad of gradually progressive dementia (Iddon et al., 1999
), gait disturbance and urinary incontinence. An NPH scale modified from Larsson and colleagues (Larsson et al., 1991) which assessed gait [0 = normal, 1 = insecure, 2 = insecure (cane), 3 = bimanual support, 4 = aided, 5 = wheelchair], living condition (0 = independent, 1 = at home with assistance, 2 = retirement home, 3 = nursing home, 4 = hospital) and urinary symptoms (0 = nil, 1 = present) was used to characterize and grade the clinical syndrome. This resulted in a score out of 10 points as well as a gait score. In addition, as a measure of overall function, a modified Stein–Langfitt score was determined (0 = no neurological deficit and able to work, 1 = minimal deficit and able to function independently at home, 2 = some supervision required at home, 3 = custodial care required despite considerable independent function, 4 = no capacity for independent function). In the patients, the mean Larsson gait score was 2.9 (median 2.5, range 1–5), the mean total Larsson score was 4.9 (median 4.5, range 2–7) and the mean modified Stein–Langfitt score was 2.3 (median 2, range 2–4). The course of symptoms was relatively short, with a mean duration of 11 months (range 6–18 months) and progressive, with no stepwise deterioration suggestive of cerebrovascular disease. Their brain MRI scan showed communicating hydrocephalus with a bicaudate ratio greater than 0.25. The ICP, measured in the ventricles through an Ommaya reservoir at the time of the infusion study (see below), was normal (mean 11.1, median 12.4, standard deviation 2.5, range 6.2–13 mmHg). In addition, on the T2-weighted MRI scans, nine of the patients had contiguous periventricular rims of hyperintensity. Three patients had limited foci of paraventricular subcortical white matter T2 hyperintensity suggestive of minor associated ischaemic changes (deep WM lesions). No cause for the hydrocephalus was defined in these patients, hence the diagnosis of idiopathic NPH. Five patients had a history of treated arterial hypertension, with a blood pressure consistently in the normal range. Their only antihypertensive medication was a thiazide diuretic.
According to the standard clinical practice established in our unit, patients presenting with NPH undergo a CSF infusion study so that we can determine the CSF circulatory and compensatory parameters (Czosnyka et al., 1996) and hence decide whether a shunt is appropriate. For this, a right-frontal ventricular catheter connected to an Ommaya reservoir is surgically placed and the infusion study is carried out a least 1 week after the procedure. In our patient group this test was performed on average 24 days (range 7–82 days) after the operation.
Ten healthy volunteers (eight men and two women) ranging in age from 30 to 60 years (mean age 45 years), without a history of neurological or cardiovascular disease, served as control subjects for CBF.
The present study was approved by the Local Research Ethics Committee of the Cambridge Health Authority. Informed written consent was obtained from the patients or their families, and from the control subjects, according to the Declaration of Helsinki.
CSF infusion study
For the infusion study, the patients were lying supine in the gantry of the PET scanner and two needles (gauge 25) were inserted into the Ommaya reservoir. One needle was connected to a pressure transducer via a stiff saline-filled tube and the other to an infusion pump. CSF pressure was monitored and recorded on a laptop personal computer. After a minimum of 30 min of baseline measurement, an infusion of normal saline at a constant rate of 1.0 or 1.5 ml/min was started and continued until a steady-state ICP plateau was reached, which usually occurred within approximately 10 min. This ICP plateau was then maintained for 20 min by continuing the infusion. CSF circulatory and compensatory parameters were then calculated using computer-supported methods based on physiological models of the CSF circulation (Czosnyka et al., 1996).
PET and MRI scans
PET studies were undertaken with a General Electric Advance scanner (GE Medical Systems, Milwaukee, USA) (Owler et al., 2003). A 10 min transmission scan using two rotating 68Ge/68Ga rod sources was first performed in all patients and used to correct subsequent emission scans for photon attenuation. Emission data were acquired during a 20 min steady-state intravenous infusion of 800 MBq of H215O, first at the baseline ICP and then once the steady-state plateau of ICP had been reached. The PET scan room was silent and the patients had their eyes closed during the scanning.
Images were reconstructed using the PROMIS 3D filtered back-projection algorithm (Kinahan and Rogers, 1989). Emission images were co-registered using Statistical Parametric Mapping (SPM99, Wellcome Department of Cognitive Neurology, London, UK) onto 3D structural T1-weighted MRIs obtained just before the PET scan on a Bruker MedSpec S300 scanner (Bruker, Ettlingen, Germany) operating at 3 T. The voxel size of this 3D MRI acquisition was 1 x 1 x 1 mm and the co-registered PET images were resliced to this resolution. Parametric maps of CBF were calculated by inputting simultaneous PET and arterial tracer activity measurements into standard models (Frackowiak et al., 1980).
The MRI scan also included the acquisition of T2-weighted images, which were used to judge whether the patients had contiguous periventricular rims and/or deep WM lesions.
The arterial catheter placed in each patient’s left radial artery for the collection of arterial blood samples for arterial tracer activity determination was also connected via a stiff saline-filled tube to a pressure transducer. The arterial blood pressure (ABP) was monitored and recorded continuously during the infusion study and the PET scan session. The arterial partial pressure of oxygen (PaO2), the arterial partial pressure of carbon dioxide (PaCO2), pH and haematocrit were measured at the baseline and during the infusion.
In the control subjects, image acquisition and reconstruction were accomplished in the same manner, except that only a baseline CBF determination was performed.
Analysis of global CBF and of its change
The global mean CBF (mCBF) at baseline and during the infusion was calculated from the PET parametric maps. For this, individual structural MRIs were segmented to extract a template that identified the brain tissue excluding the extracerebral tissue (scalp, skull, major blood vessels) and CSF. This brain template was then applied to the corresponding spatially co-registered PET parametric maps, thus restricting the analysis to the brain tissue. For each patient, CPP was calculated as the mean ABP minus the mean ICP, and hence the change in CPP induced by the increase in ICP was 
CPP = (mean ABPinfusion – mean ICPinfusion) – (mean ABPbaseline – mean ICPbaseline). The static rate of autoregulation (SRAR) was calculated as: SRAR = (CVR%/CPP%) x 100, where cerebrovascular resistance CVR = CPP/CBF (SRAR 100% indicates ideal autoregulation and 0% the absence of autoregulation).
Analysis of regional CBF and its change as a function of distance from the ventricles
The periventricular and paraventricular subcortical WM at the level of the lateral ventricles, above the caudate nucleus, was chosen because of the extensive range of distances from the ventricles to the cortex without intervening grey matter. In patients, the axial slices meeting these criteria ranged from the level of the corona radiata down to the level of the septum pellucidum, whereas in controls the slices spanned from the level of the corona radiata down to the level of the corpus callosum only (the slices at the level of the septum were excluded as they included portions of the caudate). The axial T1-weighted MRI slices meeting these criteria were retained and the WM was segmented on these slices to exclude the cortical grey matter. Only the WM on the external aspect of the lateral ventricles was retained. A binary map of this WM was generated (Fig. 1). The ventricles were also segmented on these slices and a binary map of their walls was created. For the regional analysis of the changes of CBF, as we did not have control data and as the aim was to study a possible influence of ventricular dilatation on the distribution of the CBF changes, we retained only the slices where the ventricular enlargement was maximal, i.e. at the level of the septum pellucidum.
|
A program written in-house (by S.M.) in MATLAB (MathWorks, Natick, MD, USA) was then used to calculate the shortest distance in millimetres between each WM voxel and the ventricular wall, along with the baseline CBF and the percentage change in CBF at that voxel. For each patient, the WM voxels were binned according to their distance from the ventricular wall. The mean baseline CBF and the mean percentage change in CBF at each millimetre increment in distance were calculated. As the largest thickness of WM was different across the patients, depending on the ventricular enlargement, the distance from the ventricular surface was normalized for each patient by dividing it by the largest thickness of WM in that particular patient. This yielded distances ranging from 0 (WM adjacent to the ventricles) to 1 (outermost subcortical WM), which were binned in increments of 0.1. A profile of the regional distribution of the baseline WM CBF and of the percentage change in WM CBF as a function of distance from the ventricles could then be plotted.
In control subjects, the analysis was performed as described above, except that only the profile of the regional distribution of the baseline WM CBF was available.
Statistical analysis
Data are presented as mean ± SD. The paired Student’s t-test was used for testing the changes in ICP, ABP, CPP and global mean CBF. The relationship between changes in CPP and changes in global mean CBF was tested and quantified by linear regression analysis. The relationships between the distance from the ventricles and the baseline WM CBF as well as the percentage change in WM CBF were analysed using the Kruskal–Wallis test. The threshold for significance was P < 0.05.
|
Results |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
The mean baseline ICP of the 12 patients was 11.1 ± 2.5 mmHg and the ICP increased significantly during the infusion in all patients to a mean of 28.4 ± 8 mmHg (P < 0.0001), the mean difference being a rise of 17.3 mmHg. A significant rise in ABP was also measured in all patients, from 89.9 ± 14.8 to 97.9 ± 15.7 mmHg (P < 0.001), the mean difference being a rise of 8 mmHg. CPP decreased in 10 patients, whereas in two the ABP rise outweighed the ICP rise, resulting in an increase in CPP. The mean CPP reduction in the 10 patients was 13.1 mmHg, from 79.6 ± 15 to 66.5 ± 14.7 mmHg (P < 0.0001).
The resistance to CSF outflow was 17.6 ± 8 mmHg/ml/min in the whole group of 12 patients and 19.4 ± 7.6 mmHg/ml/min in the subgroup of 10.
The global mean baseline CBF in the 12 patients was 28.4 ± 5.2 ml/100 ml/min, whereas it was 33 ± 5.4 ml/ 100 ml/min in the control subjects. This difference was significant (P < 0.005, unpaired Student’s t-test). In the 10 patients whose CPP fell, the global mean CBF significantly decreased during the infusion, from 27.6 ± 3.1 to 24.5 ± 2.9 ml/100 ml/min (P < 0.0001), whereas it increased in the two patients whose CPP rose. In the 12 patients, there was a linear correlation between the changes in CPP and the changes in CBF (r = 0.96, P < 0.0001) (Fig. 2). This relationship remained valid when we analysed only the 10 patients whose CPP and CBF fell (r = 0.95, P < 0.0001).
|
The static rate of autoregulation calculated from the global mean CBF indicated on average poor autoregulation in the patients (37 ± 12%).
The relationship between the distance from the ventricles and the mean regional WM CBF at that distance showed a logarithmic profile in all patients (P < 0.0001), whereas in controls no relationship was apparent (P = 0.0748) (Fig. 3). In the patients, WM CBF was maximally reduced adjacent to the ventricles and progressively normalized with distance, whereas in controls it remained constant and in the normal range through the entire thickness of WM.
|
In the subgroup of 10 patients whose CBF fell, the profile of the mean percentage change in regional WM CBF as a function of distance from the ventricles showed a polynomial relationship (P = 0.0007), with a regional CBF reduction increasing from the ventricles to a maximum at 0.3 of normalized distance, followed by a progressive decline towards the cortex (Fig. 4). As the mean thickness of explored WM in patients was 29 mm (range 21–39), this distance of 0.3 corresponded to a mean of 9 mm from the ventricular wall. The SRAR at the normalized distance of 0.3 was 31 ± 12% versus 66 ± 21% adjacent to the cortex (normalized distance of 1). This difference was significant at P = 0.0076 (paired sign test).
|
The PaO2, PaCO2, pH and haematocrit at the baseline were within the normal physiological ranges and did not change during the infusion.
|
Discussion |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
The controversy surrounding the regional cerebrovascular changes in NPH probably reflects both the variation between series in the characteristics of the patients studied and the lack of CBF technology of sufficient resolution (Owler and Pickard, 2001
). The present study used H2150 PET with co-registration onto high-field MRI to study both global and regional CBF in patients who fulfilled the clinical criteria for NPH and had ventriculomegaly and generally an elevated CSF outflow resistance. The main findings of the present study were that the distribution of WM CBF is different in NPH patients compared with normal controls, with a more pronounced CBF reduction adjacent to the lateral ventricles and a logarithmic normalization with distance from the ventricles. Furthermore, raising the CSF pressure reduced not only global CBF but also WM CBF, with a U-shaped distribution from the lateral ventricles to the subcortex. Our findings expand our previous study (Owler et al., 2003) and those of other authors who have described an enlargement of the area of the region of subcortical periventricular low CBF (Waldemar et al., 1993). This area shows a greater restoration of CBF after shunting than in the cortex which correlates with clinical improvement (Kimura et al., 1992; Tanaka et al., 1997). In more acute hydrocephalus, CBF is reduced in the areas of periventricular lucency, notably in the frontal WM, and increases after successful shunting (Matsuda et al., 1990; Kimura et al., 1992; Nakano et al., 1996; Pena et al., 1999). Furthermore, it has recently been shown in an animal model of chronic adult hydrocephalus that the only statistically significant regional CBF reduction persisting in the late stage of hydrocephalus was located in the periventricular WM (Klinge et al., 2003), which is consistent with our study. Various authors have advanced the following hypotheses, which are not mutually exclusive, to explain the reduction in periventricular CBF in NPH.
Tissue distortion
It has been hypothesized that distortion and stretching of the periventricular tissue, including vessels, lead to an increase in local cerebrovascular resistance and predispose to ischaemia (Greitz, 1969; Hakim et al., 1976; Meyer et al., 1984; Akai et al., 1987; Del Bigio and Bruni, 1988). Hakim further proposed that, in the early stages of development of NPH, the mechanical strain on the brain parenchyma (and hence the interstitial tissue pressure) would be more pronounced centrally in the periventricular region, with a decreasing gradient from there. Later, as the periventricular tissue yields, the brain parenchyma shrinks and CSF pressure returns to normal (i.e. when the state of NPH is reached), the gradient of stress distribution would be inverted, with more pronounced strain at the periphery of the brain (Hakim et al., 1976). Therefore, at this stage of the condition, CBF is expected to be predominantly reduced at the periphery of the cerebral mantle or, if some degree of added periventricular vessel stretching and distortion is also admitted, evenly reduced throughout the cerebral mantle and its WM.
Watershed ischaemia
Watershed ischaemia within the corona radiata might occur in the boundary zone between middle cerebral artery perforators and the medullary branches from the pial arteries, compounded by a disturbance of cerebrovascular autoregulation within the WM (Mathew et al., 1975; Meyer et al., 1984). Such a watershed phenomenon is consistent with the profile of the regional change of WM CBF brought about by a decrease in CPP, showing a polynomial U-shaped curve from the lateral ventricles to the subcortex. The maximal CBF decrease occurred at 0.3 of the normalized distance, which corresponded to a mean distance of 9 mm. Infarction in these watershed zones is generally presumed to be due to decreased cerebral perfusion (Bladin and Chambers, 1993). Our findings are consistent with vulnerability of the corona radiata to reductions in CPP within the conventionally defined autoregulatory range when accompanied by exhaustion or impairment of CBF pressure autoregulation. There is little normative data available on regional autoregulation, notably in the vascular border zones, following changes in ICP equivalent to the global studies of Johnston and colleagues (Johnston et al., 1972). The poor global autoregulation found in our group of patients is probably the average of mediocre autoregulation close to the ventricles, notably in the watershed territories, and relatively normal autoregulation close to the cortex. This is consistent with our previous study using transcranial Doppler ultrasonography with insonation of the distal middle cerebral artery (MCA), supplying the cortex and the superficial subcortical WM, where autoregulation was relatively preserved in patients with raised CSF outflow resistance (Czosnyka et al., 2002). There are remarkably few studies of CBF pressure autoregulation in NPH (Schmidt et al., 1990), where it has never been tested during a reduction of CPP brought about by an increase in ICP. Cerebral blood flow pressure autoregulation to reductions of CSF pressure has been considered by some authors to be impaired in NPH (Mathew et al., 1975; Mamo et al., 1987). Moreover, another form of cerebrovascular responsiveness, cerebrovascular reactivity to acetazolamide or CO2, has been demonstrated in some studies to be reduced in NPH compared with normal controls (Meyer et al., 1984; Lee et al., 1998; Chang et al., 2000).
Vasoactive metabolites
In NPH the CSF may suffuse from the ventricles into the parenchyma and the flow of interstitial fluid may be reversed, which can bring about interstitial/extracellular oedema, predominantly in the periventricular tissue (Tamaki et al., 1990). Such oedema does not cause ischaemia, but inappropriate vasoactive metabolites may stagnate in this oedema and, if the ISF/ECF (interstitial/extracellular fluid) contains vasoactive compounds, local cerebrovascular reactivity may be reduced (Marmarou et al., 1980; Whittle et al., 1992). These compounds could have deleterious effects by interfering with intrinsic mechanisms (such as nitric oxide) subserving cerebrovascular reactivity. Moreover, the accumulation of amyloid-ß protein by failure of drainage may damage various tissue components, including vessels (Silverberg et al., 2003). Such pathological processes could lead to failure of autoregulation in NPH. Our study showed heterogeneity of regional autoregulation, with deeper regions more and shallower regions less affected, which could be related to the predominance of the interstitial/extracellular oedema in the periventricular tissue.
Vascular disease
Changes in the material properties of the brain may predispose to ventricular enlargement and hence to stretching and distortion of the periventricular WM and its blood supply. In some idiopathic NPH cases, cerebral small vessel disease, notably due to hypertension, might be the initial pathological process leading to hydrocephalus, by multiple small deep cerebral infarctions in the periventricular WM and basal ganglia. It has been suggested that such multiple areas of infarction reduce the elasticity of the periventricular tissue, leading to enlargement of the ventricles provoked by intraventricular CSF pressure pulsations, which are also increased in case of systemic hypertension (Earnest et al., 1974; Koto et al., 1977). Arterial hypertension has been found to be more prevalent among idiopathic NPH patients than among age-matched demented Alzheimer’s disease patients (Graff-Radford and Godersky, 1987) and healthy age-matched controls (Casmiro et al., 1989; Krauss et al., 1996a). However, the direction of the association between idiopathic NPH and hypertension is still not clear, as Miller Fisher, who noted the association between NPH and lacunes, alternatively suggested that the compressive effect of NPH may predispose to lacunar white matter infarction (Fisher, 1982). Deep WM lesions on MRI suggestive of concomitant vascular encephalopathy do not exclude NPH patients from improving with a shunt, but the degree of improvement is negatively correlated with the extent of these lesions (Krauss et al., 1996b).
Interstitial fluid pressure increase
The suffusion of CSF into the periventricular tissue and the reversal of interstitial fluid flow might be supposed to compress local vessels, particularly underneath the ventricular wall and tapering off peripherally (Shulman et al., 1975; Tamaki et al., 1990; Kimura et al., 1992). The distribution of interstitial parenchymal pressure in NPH has recently been modelled using finite-element analysis based on Biot’s theory of deformation of a poroelastic medium (Pena et al., 2002). Pena and colleagues have proposed the hypothesis that the chronic ventricular dilatation of communicating hydrocephalus can be explained by a combination of reversal of interstitial fluid flow into the parenchyma and reduced tissue elasticity. In the presence of significant CSF suffusion into the brain parenchyma, the profile of the interstitial parenchymal pressure in the cerebral mantle would have a U curve from the ventricular surface to the cortical surface, with a minimum at the middle of the cerebral mantle. If the magnitude of CSF suffusion from the ventricles is larger than that from the subarachnoid CSF spaces, the distribution of the interstitial parenchymal pressure would have a skewed U shape, with a minimum located much closer to the brain convexity, possibly at the boundary between the WM and the cortical grey matter. The distribution of the interstitial pressure in the WM would appear as an exponential decrease from the ventricles through the WM. It is thus conceivable that the logarithmic profile of WM CBF found in the present study mirrors and results from such an exponential distribution of interstitial tissue pressure. However, the ICP essentially remains in the normal range in NPH, and hence it could be argued that any gradient of interstitial tissue pressure would be too low to compress the capillaries. We would suggest that failure of autoregulation and of drainage of vasoactive/toxic substances (see above) would provide the additional links in the chain of causation of border-zone ischaemia and tissue damage. It is not possible to say whether the reduced WM CBF observed in our study is sufficiently low to be categorized as ischaemic without measurements of cerebral oxygen extraction ratio by 15O2 PET. However, we suggest that the impairment of autoregulation and the reduction in WM CBF closer to the ischaemic threshold render the WM vulnerable to periods of modest systemic hypotension or rises in CSF pressure; for example, during REM sleep. Pantoni et al. (1996) have suggested that WM is more vulnerable to ischaemia than previously thought. Three of our 12 NPH patients had deep WM ischaemic changes on the T2 MRIs. Repeated falls of CBF below the ischaemic threshold could also produce delayed ischaemic neuronal damage in the ischaemia-vulnerable hippocampus (Klinge et al., 2003).
In our study the control subjects were younger than the patients. However, Meltzer et al. (2000) have shown that, if account is taken in H215O PET CBF studies of the partial volume effect created by age-related cerebral atrophy, CBF does not decline with age in healthy individuals. Moreover, an intra-group analysis showed no relationship between CBF and age in our control population (r2 = 0.217, P = 0.127). Our younger control group can therefore be considered as legitimate.
Conclusions
The WM CBF is reduced in NPH, with an abnormal gradient from the lateral ventricles to the outermost subcortical WM. Furthermore, CBF pressure autoregulation to reductions in CPP produced by CSF infusion is impaired, with particular vulnerability of the paraventricular watershed region of the corona radiata between the deep and superficial territories of the MCA. These findings are consistent with a contribution from various processes, including tissue distortion, reversal of CSF and interstitial fluid flow and hence failure of drainage of vasoactive and toxic substances from the parenchyma, watershed ischaemia compounded by failure of WM CBF autoregulation and, finally, changes in the material properties of the brain provoked by cerebral small vessel disease. Such a combination of factors could explain the increased incidence of deep WM lesions in NPH and why a primary CSF disorder might promote cerebral watershed ischaemia. When assessing elderly patients, who often harbour a number of possible pathologies, it is important to look for a remediable hydrocephalic component that might respond to a shunt.
|
Acknowledgements |
|---|
S.M. is supported by the Swiss National Science Foundation. B.K.O. was supported by a Wood grant from the Sydney University Medical Foundation and the Madeline Foundation for Neurosurgical Research. A.P. is supported by a Wellcome Trust training fellowship in mathematical biology. M.C. is on unpaid leave from Warsaw University of Technology, Warsaw, Poland. This work was supported by the MRC Programme Grant number G0001237/G9439390 and by the MRC WBIC Cooperative Group Grant.
|
References |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Akai K, Uchigasaki S, Tanaka U, Komatsu A. Normal pressure hydrocephalus. Neuropathological study. Acta Pathol Jpn 1987; 37: 97–110.[Medline]
Bladin CF, Chambers BR. Clinical features, pathogenesis, and computed tomographic characteristics of internal watershed infarction. Stroke 1993; 24: 1925–32.
Casmiro M, D’Alessandro R, Cacciatore FM, Daidone R, Calbucci F, Lugaresi E. Risk factors for the syndrome of ventricular enlargement with gait apraxia (idiopathic normal pressure hydrocephalus): a case–control study. J Neurol Neurosurg Psychiatry 1989; 52: 847–52.[Abstract]
Chang CC, Kuwana N, Ito S, Ikegami T. Impairment of cerebrovascular reactivity to acetazolamide in patients with normal pressure hydrocephalus. Nucl Med Commun 2000; 21: 139–41.[CrossRef][ISI][Medline]
Czosnyka M, Whitehouse H, Smielewski P, Simac S, Pickard JD. Testing of cerebrospinal compensatory reserve in shunted and non-shunted patients: a guide to interpretation based on an observational study. J Neurol Neurosurg Psychiatry 1996; 60: 549–58.[Abstract]
Czosnyka ZH, Czosnyka M, Whitfield PC, Donovan T, Pickard JD. Cerebral autoregulation among patients with symptoms of hydrocephalus. Neurosurgery 2002; 50: 526–32.[ISI][Medline]
Del Bigio MR, Bruni JE. Changes in periventricular vasculature of rabbit brain following induction of hydrocephalus and after shunting. J Neurosurg 1988; 69: 115–20.[ISI][Medline]
Earnest MP, Fahn S, Karp JH, Rowland LP. Normal pressure hydrocephalus and hypertensive cerebrovascular disease. Arch Neurol 1974; 31: 262–6.[ISI][Medline]
Fisher CM. Lacunar strokes and infarcts: a review. Neurology 1982; 32: 871–6.
Frackowiak RS, Lenzi GL, Jones T, Heather JD. Quantitative measurement of regional cerebral blood flow and oxygen metabolism in man using 15O and positron emission tomography: theory, procedure, and normal values. J Comput Assist Tomogr 1980; 4: 727–36.[ISI][Medline]
Graff-Radford NR, Godersky JC. Idiopathic normal pressure hydrocephalus and systemic hypertension. Neurology 1987; 37: 868–71.
Greitz T. Effect of brain distension on cerebral circulation. Lancet 1969; 1: 863–5.[ISI][Medline]
Hakim S, Venegas JG, Burton JD. The physics of the cranial cavity, hydrocephalus and normal pressure hydrocephalus: mechanical interpretation and mathematical model. Surg Neurol 1976; 5: 187–210.[Medline]
Hertel F, Walter C, Schmitt M, Morsdorf M, Jammers W, Busch HP, et al. Is a combination of Tc-SPECT or perfusion weighted magnetic resonance imaging with spinal tap test helpful in the diagnosis of normal pressure hydrocephalus? J Neurol Neurosurg Psychiatry 2003; 74: 479–84.
Iddon JL, Pickard JD, Cross JJ, Griffiths PD, Czosnyka M, Sahakian BJ. Specific patterns of cognitive impairment in patients with idiopathic normal pressure hydrocephalus and Alzheimer’s disease: a pilot study. J Neurol Neurosurg Psychiatry 1999; 67: 723–32.
Johnston IH, Rowan JO, Harper AM, Jennett WB. Raised intracranial pressure and cerebral blood flow. I. Cisterna magna infusion in primates. J Neurol Neurosurg Psychiatry 1972; 35: 285–96.[ISI][Medline]
Kimura M, Tanaka A, Yoshinaga S. Significance of periventricular hemodynamics in normal pressure hydrocephalus. Neurosurgery 1992; 30: 701–4.[ISI][Medline]
Kinahan PE, Rogers JG. Analytic 3D image reconstruction using all detected events. IEEE Trans Nucl Sci 1989; 36: 964–8.[CrossRef][ISI]
Klinge PM, Samii A, Muhlendyck A, Visnyei K, Meyer GJ, Walter GF, et al. Cerebral hypoperfusion and delayed hippocampal response after induction of adult kaolin hydrocephalus. Stroke 2003; 34: 193–9.
Koto A, Rosenberg G, Zingesser LH, Horoupian D, Katzman R. Syndrome of normal pressure hydrocephalus: possible relation to hypertensive and arteriosclerotic vasculopathy. J Neurol Neurosurg Psychiatry 1977; 40: 73–9.[Abstract]
Krauss JK, Regel JP, Vach W, Droste DW, Borremans JJ, Mergner T. Vascular risk factors and arteriosclerotic disease in idiopathic normal-pressure hydrocephalus of the elderly. Stroke 1996a; 27: 24–9.
Krauss JK, Droste DW, Vach W, Regel JP, Orszagh M, Borremans JJ, et al. Cerebrospinal fluid shunting in idiopathic normal-pressure hydrocephalus of the elderly: effect of periventricular and deep white matter lesions. Neurosurgery 1996b; 39: 292–300.[CrossRef][ISI][Medline]
Larsson A, Wikkelso C, Bilting M, Stephensen H. Clinical parameters in 74 consecutive patients shunt operated for normal pressure hydrocephalus. Acta Neurol Scand 1991; 84: 475–82.[ISI][Medline]
Lee EJ, Hung YC, Chang CH, Pai MC, Chen HH. Cerebral blood flow velocity and vasomotor reactivity before and after shunting surgery in patients with normal pressure hydrocephalus. Acta Neurochir (Wien) 1998; 140: 599–604.[CrossRef][Medline]
Mamo HL, Meric PC, Ponsin JC, Rey AC, Luft AG, Seylaz JA. Cerebral blood flow in normal pressure hydrocephalus. Stroke 1987; 18: 1074–80.
Marmarou A, Takagi H, Shulman K. Biomechanics of brain edema and effects on local cerebral blood flow. Adv Neurol 1980; 28: 345–58.[Medline]
Mathew NT, Meyer JS, Hartmann A, Ott EO. Abnormal cerebrospinal fluid–blood flow dynamics. Implications in diagnosis, treatment, and prognosis in normal pressure hydrocephalus. Arch Neurol 1975; 32: 657–64.[Abstract]
Matsuda M, Nakasu S, Nakazawa T, Handa J. Cerebral hemodynamics in patients with normal pressure hydrocephalus: correlation between cerebral circulation time and dementia. Surg Neurol 1990; 34: 396–401.[CrossRef][ISI][Medline]
Meltzer CC, Cantwell MN, Greer PJ, Ben-Eliezer D, Smith G, Frank G, et al. Does cerebral blood flow decline in healthy aging? A PET study with partial-volume correction. J Nucl Med 2000; 41: 1842–8.
Meyer JS, Tachibana H, Hardenberg JP, Dowell RE Jr, Kitagawa Y, Mortel KF. Normal pressure hydrocephalus. Influences on cerebral hemodynamic and cerebrospinal fluid pressure–chemical autoregulation. Surg Neurol 1984; 21: 195–203.[CrossRef][ISI][Medline]
Mori K, Maeda M, Asegawa S, Iwata J. Quantitative local cerebral blood flow change after cerebrospinal fluid removal in patients with normal pressure hydrocephalus measured by a double injection method with N-isopropyl-p-[123I] iodoamphetamine. Acta Neurochir (Wien) 2002; 144: 255–63.[CrossRef][Medline]
Mori K, Mima T. To what extent has the pathophysiology of normal-pressure hydrocephalus been clarified. Crit Rev Neurosurg 1998; 8: 232–43.
Nakano H, Bandoh K, Miyaoka M, Sato K. Evaluation of hydrocephalic periventricular radiolucency by dynamic computed tomography and xenon-computed tomography. Neurosurgery 1996; 39: 758–62.[CrossRef][ISI][Medline]
Owler BK, Pickard JD. Normal pressure hydrocephalus and cerebral blood flow: a review. Acta Neurol Scand 2001; 104: 325–42.[CrossRef][ISI][Medline]
Owler BK, Momjian S, Czosnyka Z, Czosnyka M, Pena A, Harris NG, et al. Normal pressure hydrocephalus and cerebral blood flow. A PET study of baseline values. J Cereb Blood Flow Metab 2004; 24: 17–23.[CrossRef][ISI][Medline]
Pantoni L, Garcia JH, Gutierrez JA. Cerebral white matter is highly vulnerable to ischemia. Stroke 1996; 27: 1641–7.
Pena A, Bolton MD, Whitehouse H, Pickard JD. Effects of brain ventricular shape on periventricular biomechanics: a finite-element analysis. Neurosurgery 1999; 45: 107–16.[CrossRef][ISI][Medline]
Pena A, Harris NG, Bolton MD, Czosnyka M, Pickard JD. Communicating hydrocephalus: the biomechanics of progressive ventricular enlargement revisited. Acta Neurochir Suppl (Wien) 2002; 81: 59–63.[Medline]
Schmidt JF, Andersen AR, Paulson OB, Gjerris F. Angiotensin converting enzyme inhibition, CBF autoregulation, and ICP in patients with normal-pressure hydrocephalus. Acta Neurochir (Wien) 1990; 106: 9–12.[CrossRef][Medline]
Shulman K, Marmarou A, Weitz S. Gradients of brain interstitial fluid pressure in experimental brain infusion and compression. In: Lundberg N, Ponten U, Brock M, editors. Intracranial pressure II. Berlin: Springer-Verlag; 1975. p. 221–3.
Silverberg GD, Mayo M, Saul T, Rubenstein E, McGuire D. Alzheimer’s disease, normal-pressure hydrocephalus, and senescent changes in CSF circulatory physiology: a hypothesis. Lancet Neurol 2003; 2: 506–11.[CrossRef][ISI][Medline]
Tamaki N, Nagashima T, Ehara K, Shirakuni T, Matsumoto S. Hydrocephalic oedema in normal-pressure hydrocephalus. Acta Neurochir Suppl (Wien) 1990; 51: 348–50.[Medline]
Tanaka A, Kimura M, Nakayama Y, Yoshinaga S, Tomonaga M. Cerebral blood flow and autoregulation in normal pressure hydrocephalus. Neurosurgery 1997; 40: 1161–5.[CrossRef][ISI][Medline]
Waldemar G, Schmidt JF, Delecluse F, Andersen AR, Gjerris F, Paulson OB. High resolution SPECT with [99mTc]-d,l-HMPAO in normal pressure hydrocephalus before and after shunt operation. J Neurol Neurosurg Psychiatry 1993; 56: 655–64.[Abstract]
Whittle IR, Piper IR, Miller JD. The role of bradykinin in the etiology of vasogenic brain edema and perilesional brain dysfunction. Acta Neurochir (Wien) 1992; 115: 53–9.[CrossRef][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G.A. Bateman The Pathophysiology of Idiopathic Normal Pressure Hydrocephalus: Cerebral Ischemia or Altered Venous Hemodynamics? AJNR Am. J. Neuroradiol., January 1, 2008; 29(1): 198 - 203. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Malm and A. Eklund Idiopathic normal pressure hydrocephalus Practical Neurology, February 1, 2006; 6(1): 14 - 27. [Full Text] [PDF]
|
|
|
|
 |
|
 G. D. Silverberg Normal pressure hydrocephalus (NPH): ischaemia, CSF stagnation or both Brain, May 1, 2004; 127(5): 947 - 948. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||