Correspondence *Department of Neurology, University of Cincinnati, 231 Albert Sabin Way, MSB 4503, Cincinnati, OH 45267–0525, USA

Email alberto.espay@uc.edu

The case

An 80-year-old man presented to a movement disorders clinic for evaluation of step-wise progressive impairment of gait, balance, and memory. Thirty months earlier, he had developed a short-stepped gait and impaired balance. He suffered multiple falls during the ensuing year. Over the 9 months before presentation, the patient had become incontinent of urine. He had been hospitalized twice for sudden-onset freezing of gait. Although the patient had not noted changes in speech, swallowing, visual function, or handwriting, his daughter reported that his short-term memory had been impaired for the past few years. Levodopa, up to 600 mg/day, had failed to alleviate his symptoms. The patient resided at a rehabilitation facility where he received assistance with gait and transfers, but he was able to bathe, dress, or feed without assistance.

The patient's past medical history was positive for hypertension, hypercholesterolemia, hypokalemic episodes, and cerebral microangiopathic disease. He had accumulated a 60 pack-year smoking habit, but he had stopped smoking 40 years before presentation. His alcohol intake was restricted to social functions.

On examination, the patient's cognitive function was mildly impaired (Mini-Mental State Examination score 25 out of 30) and frontal release signs (snout and palmomental reflexes) were present. He exhibited a short-stride gait without stooping, shuffling, or festination. The patient had mild paratonia of the upper limbs but no rigidity in the legs or neck. Mild hypesthesia to temperature and light touch, in a stocking-glove distribution, was documented. His postural reflexes were impaired. Results of a thorough neuropsychological evaluation were compared with a baseline assessment ascertained at symptom onset 30 months previous (Table 1). On an overall index of mental functioning (Dementia Rating Scale-2), the patient's performance had dropped markedly from the average range (65^th percentile) to the extremely low range (<1^st percentile). Importantly, all executive functions (behavioral initiation, motor regulation, planning, organization, mental flexibility, problem solving, etc.) dropped from the high average range to below the 1^st percentile. With the exception of word finding and naming skills, a similar decline was noted in all other nonexecutive functions tested (visual-spatial construction, attention and memory).

Table 1 Selected tests of neuropsychological function.

Full tableFigures & Tables indexDownload Power Point slide (86K)

Brain MRI scans showed confluent punctate areas of increased T2-weighted signal intensity affecting subcortical and periventricular white matter in a pattern consistent with chronic small-vessel ischemic disease (Figure 1). Lateral and third ventricles were moderately dilated but not disproportionably to the degree of cortical atrophy. The combination of the patient's history and MRI findings prompted the diagnosis of vascular parkinsonism (VaP).

Figure 1 Brain MRI scans of a patient with suspected vascular parkinsonism.

Axial fluid-attenuated inversion recovery (A–C) and T2-weighted (D–F) MRI scans of the basal ganglia (A,D), and the mid-ventricular (B,E) and high-ventricular (C,F) regions. The increased periventricular and subcortical white matter signal is suggestive of small-vessel ischemic disease. The mild ventricular dilatation is roughly proportionate to the apparent degree of cortical atrophy.

Full figure and legend (36K)Figures & Tables indexDownload Power Point slide (97K)

Gait and cognitive assessments were performed before and immediately after a 3-day continuous external lumbar drainage (ELD). The patient's gait improved substantially in velocity and stride length, and modestly in cadence (Table 2; see also Supplementary Video 1 online). Although cognitive parameters remained in the abnormal range, most were relatively improved compared with the pre-ELD measurements (Table 1). On the basis of these data, the patient and his family agreed to ventriculoperitoneal shunt (VPS) placement.

Table 2 Gait assessments before and after ELD.

Full tableFigures & Tables indexDownload Power Point slide (78K)

Shortly after placement of a VPS, the patient unfortunately died as a result of peritonitis caused by bowel perforation. Post-mortem examination of the patient's brain (weight 1,400g; within normal range) revealed communicating hydrocephalus, with leptomeningeal fibrosis and superficial gliosis of the cerebral cortex, but no macroinfarcts or lacunes (Figure 2). Lewy bodies and tau pathology were not present. Additional neuropathology documented only minimal fibrinoid necrosis and arteriolosclerosis, associated with focal reduction in the number of oligodendrocytes in the basal ganglia and frontal white matter. The neuropathology was consistent with that of normal pressure hydrocephalus (NPH) and did not support the clinical diagnosis of VaP.

Figure 2 Photographs of two coronal sections of the patient's formalin-fixed cerebral hemispheres.

(A) At the level of the mamillary bodies. (B) Posterior to the splenium of the corpus callosum. The degree of dilatation of both lateral ventricles (which have rounded external angles of the frontal horns) appears disproportionate to the mild degree of cortical atrophy (unlike the interpretation of the brain MRI [Figure 1]), and seems unrelated to gray matter or white matter infarcts. The putaminal 'lacune' (arrow) histologically represents a dilated vein (varix).

Full figure and legend (27K)Figures & Tables indexDownload Power Point slide (86K)

Discussion of diagnosis

Guidelines have been proposed for the diagnosis of NPH and VaP, the two most common forms of the syndrome of lower-body parkinsonism.^{1, 2} The diagnosis of NPH is used when the triad of gait disturbance, urinary incontinence, and cognitive impairment is caused by communicating hydrocephalus with normal intracranial pressure (<20 mmHg); NPH often improves after placement of a VPS. VaP refers to the lower-body parkinsonism that results from multiple cerebral infarctions in the basal ganglia and/or white matter, with or without associated dementia and in the absence of neurodegeneration or medications that can cause parkinsonism. VaP responds poorly to dopaminergic medications and responds variably to VPS placement. Although NPH and VaP present as separate entities on clinical grounds, there is a paucity of post-mortem neuropathological data that validates them as distinct conditions. Most of the distinctions made in the literature have relied on expert clinical judgment rather than clinical–pathological correlations.

This case highlights the challenges to current diagnostic boundaries. This elderly man accumulated the triad of gait impairment, dementia, and urinary incontinence. Although characteristic of NPH, this triad is non-specific and can be evident in NPH-like conditions, such as those listed in Box 1. The patient met the clinical criteria for a diagnosis of VaP,¹ which was supported by the presence of multiple vascular risk factors and findings on brain MRI interpreted as periventricular white matter ischemic disease with mild hydrocephalus. He benefited dramatically from ELD—a predictor of response to VPS placement but also a potential feature of cases with clinically diagnosed VaP.³

The Evans' index has been helpful in the operationalization of hydrocephalus: ventriculomegaly is suspected when the ratio of the maximum anterior horn width to the transverse inner diameter of the skull is greater than 0.3. The present patient's Evans' index reached this cutoff, but the accompanying abnormal periventricular white matter signal—an indication of microangiopathic disease—was given greater weight in the pre-ELD assessment.

The classification of lower-body parkinsonism has been complicated by attempts to attribute features of gait, such as freezing or apraxia, to exclusively VaP or NPH. Freezing of gait is expressed as start-hesitation and blocks, especially during turns or when approaching narrow spaces. In a cohort of 347 patients with lower-body parkinsonism, the highest frequency of freezing was seen at comparable rates in those clinically diagnosed with VaP and NPH, suggesting that VaP and NPH could be similar 'frontal gait disorders'.⁴ Gait apraxia has been defined as the discrepancy between the severity of the gait impairment and the ability to perform normally other leg movements (e.g. cycling in the air while sitting), in the absence of primary motor deficits. Initially described as a feature of NPH, apraxia can also occur in VaP and other frontal gait impairments.⁵ In their seminal description of gait in CT-defined Binswanger's disease (BD; expressed as parkinsonism [VaP] and/or dementia [vascular dementia]), Thompson and Marsden suggested that the apraxia of BD could be indistinguishable from that of chronic hydrocephalus.⁶ Subsequent studies have reported most of the gait features of lower-body parkinsonism in both NPH and VaP (Box 2), although freezing of gait and difficulty in gait initiation may be more common in VaP and probably rare in NPH, and are often unresponsive to fluid diversion procedures. Freezing of gait is also a distinct feature of Parkinson's disease, a condition that can be confused, but sometimes coexist, with NPH and VaP.

Vascular risk factors coexist in a substantial number of patients with NPH. Hypertension, ischemic heart disease, diabetes, and low levels of HDL cholesterol are frequent comorbidities of NPH.⁷ In fact, excellent gait improvement following cerebrospinal fluid (CSF) shunting can occur in patients with radiologically defined VaP, or when the burden of periventricular white matter abnormalities suggestive of concomitant microangiopathy might seem extensive.^{3, 8} Even refined quantification methods with semi-automatic segmentation fail to predict a differential response to CSF shunting in patients with clinically diagnosed NPH or BD.⁹ To complicate matters further, extensive basal ganglia imaging abnormalities can be present in normal or asymptomatic individuals,¹⁰ and, counterintuitively, striatal infarcts are rarely followed by clinical parkinsonism.¹¹ Finally, post-mortem MRI–pathologic correlations of asymptomatic subjects have demonstrated that periventricular hyperintensities are associated with myelin pallor, dilatation of perivascular spaces, an increase in extracellular spaces, discontinuity of the ependymal lining, and subependymal gliosis,¹² suggesting that hyperintense periventricular abnormalities on MRI might not indicate microangiopathy, as is often interpreted. Thus, imaging-defined small-vessel ischemic disease could be part of the broad and complex definition of NPH, rather than a separate disorder.

Treatment and management

In view of the tentative diagnosis of VaP, we initially recommended gradual withdrawal of levodopa, blood pressure monitoring with orthostatic measurements, tighter control of dyslipidemia, and use of a walker to prevent falls. These treatment strategies, however, were not initiated following the ELD response that suggested shunt responsiveness, which—although non-specific as a diagnostic test of the patient's gait impairment—redirected the therapeutic efforts with the assumption of NPH. On this basis, the decision was made to proceed with VPS placement. Unfortunately, the patient died from peritonitis caused by bowel perforation, a rare but recognized complication of VPS placement.

When compared with a large-volume CSF tap test, prolonged CSF lumbar drainage for 3 days (i.e. ELD) provides higher sensitivity (50–100%) and a positive predictive value for a VPS response approaching 100%.^{13, 14} Although the predictive value of a positive ELD response is high, that of a negative ELD response is low because of the high rate of false negative results.¹⁵ Although the recently released NPH guidelines admit that insufficient data exist to support a management standard, it was concluded that, of the three recommended supplemental tests (ELD, CSF tap test, and outflow resistance determination), "ELD is most effective for identifying shunt responsiveness... The prognostic value of the ELD is most likely retained even with 'possible' and 'unlikely' NPH clinical designations."¹⁴ It should be recognized that ELD is an invasive procedure, carries a higher complication rate than either CSF tap or outflow resistance determination, and requires hospitalization and nursing care with competency in managing the external CSF drain and monitoring for potential complications, which include overdrainage, infection, catheter breakage, and nerve-root irritation. Nevertheless, the rate of VPS placement complications, which include cerebral hemorrhage, infection, and seizure, is much higher than that of ELD, reaching 28%, and death (as exemplified by this case) or persistent disability occurs in 7% of cases.¹⁶ Hence, it is important to identify patients who have the highest likelihood of benefiting from shunt placement surgery. At present, 3-day ELD is the best predictor of response to VPS placement.

Conclusions

VaP and NPH appear to represent overlapping phenotypic entities reflecting CSF dysfunction within a pathological continuum (Figure 3). Both disorders can have a substantial burden of periventricular and deep-white-matter T2-weighted hyperintensities, a degree of hydrocephalus, and responsiveness to CSF drainage. With a lack of controlled studies that assess response to fluid diversion procedures and provide neuroimaging–neuropathologic correlations, abnormalities suggestive of small-vessel disease that are shown on T2-weighted MRI scans cannot be accurately interpreted. As in the present patient, tightening control of already well-managed vascular risk factors might be a disappointing therapy option. Lowering the threshold for using ELD could help predict shunt-responsiveness in patients with lower-body parkinsonism, regardless of the clinical diagnosis and the extent of periventricular hyperintensities on MRI.

Figure 3 Diagram representing the clinical and pathologic spectrum in lower-body parkinsonism.

VaP and NPH are the standard clinical nomenclature at either end of a continuum on which both microangiopathy and CSF circulatory dysfunction are present. Shunt-responsiveness and/or continuous ELD-responsiveness can occur when the bulk of the microangiopathy is closer to the clinical VaP end of the spectrum and does not reliably distinguish NPH from VaP. Reproduced with permission from Mayfield Clinic. Abbreviations: CSF, cerebrospinal fluid; ELD, external lumbar drainage; NPH, normal pressure hydrocephalus; VaP, vascular parkinsonism.

Full figure and legend (10K)Figures & Tables indexDownload Power Point slide (67K)

Acknowledgments

The authors thank Martha Headworth, medical illustrator, and Mary Kemper, medical editor, of The Neuroscience Institute, University of Cincinnati, OH, for their expertise in organizing the figures and providing editorial review of this manuscript, respectively.

References

1. Zijlmans JC et al. (2004) Clinicopathological investigation of vascular parkinsonism, including clinical criteria for diagnosis. Mov Disord 19: 630–640 | Article | PubMed |
2. Relkin N et al. (2005) Diagnosing idiopathic normal-pressure hydrocephalus. Neurosurgery 57 (Suppl 3): S4–S16
3. Ondo WG et al. (2002) Vascular parkinsonism: clinical correlates predicting motor improvement after lumbar puncture. Mov Disord 17: 91–97 | Article | PubMed |
4. Giladi N et al. (1997) Freezing phenomenon in patients with parkinsonian syndromes. Mov Disord 12: 302–305 | Article | PubMed | ChemPort |
5. Snijders AH et al. (2007) Neurological gait disorders in elderly people: clinical approach and classification. Lancet Neurol 6: 63–74 | Article | PubMed |
6. Thompson PD and Marsden CD (1987) Gait disorder of subcortical arteriosclerotic encephalopathy: Binswanger's disease. Mov Disord 2: 1–8 | Article | PubMed | ChemPort |
7. Casmiro M et al. (1989) Risk factors for the syndrome of ventricular enlargement with gait apraxia (idiopathic normal pressure hydrocephalus): a case-control study. J Neurol Neurosurg Psychiatry 52: 847–852 | PubMed | ISI | ChemPort |
8. Krauss JK et al. (1996) Cerebrospinal fluid shunting in idiopathic normal-pressure hydrocephalus of the elderly: effect of periventricular and deep white matter lesions. Neurosurgery 39: 292–299 | Article | PubMed | ChemPort |
9. Tullberg M et al. (2002) White matter changes in normal pressure hydrocephalus and Binswanger disease: specificity, predictive value and correlations to axonal degeneration and demyelination. Acta Neurol. Scand 105: 417–426 | Article | PubMed | ChemPort |
10. Yamanouchi H and Nagura H (1997) Neurological signs and frontal white matter lesions in vascular parkinsonism: a clinicopathologic study. Stroke 28: 965–969 | PubMed | ChemPort |
11. Peralta C et al. (2004) Parkinsonism following striatal infarcts: incidence in a prospective stroke unit cohort. J Neural Transm 111: 1473–1483 | Article | PubMed | ChemPort |
12. Matsusue E et al. (2006) White matter changes in elderly people: MR-pathologic correlations. Magn Reson Med Sci 5: 99–104 | Article | PubMed |
13. Haan J and Thomeer RT (1988) Predictive value of temporary external lumbar drainage in normal pressure hydrocephalus. Neurosurgery 22: 388–391 | Article | PubMed | ChemPort |
14. Marmarou A et al. (2005) The value of supplemental prognostic tests for the preoperative assessment of idiopathic normal-pressure hydrocephalus. Neurosurgery 57: 17–28 | Article | PubMed |
15. Walchenbach R et al. (2002) The value of temporary external lumbar CSF drainage in predicting the outcome of shunting on normal pressure hydrocephalus. J Neurol Neurosurg Psychiatry 72: 503–506 | PubMed | ChemPort |
16. Vanneste J et al. (1992) Shunting normal-pressure hydrocephalus: do the benefits outweigh the risks? A multicenter study and literature review. Neurology 42: 54–59 | PubMed | ChemPort |
17. Stolze H et al. (2001) Comparative analysis of the gait disorder of normal pressure hydrocephalus and Parkinson's disease. J Neurol Neurosurg Psychiatry 70: 289–297 | Article | PubMed | ChemPort |
18. Zijlmans JC et al. (1996) Quantitative gait analysis in patients with vascular parkinsonism. Mov Disord 11: 501–508 | Article | PubMed | ChemPort |

Supplementary information

Supplementary Video (mpg 11 MB)

Video clip of patient demonstrating markedly improved straight gait and turning following 3-day continuous external lumbar drainage (ELD). Segment 1: straight gait before and after ELD. Segment 2: turning, also before and after ELD. Gait parameters are displayed.

Contact the journal about this article

Subject areas under which this article appears: Hydrocephalus | Movement disorders