Brain Biopsies in Normal Pressure Hydrocephalus and the Subsequent Diagnosis of Alzheimer's Disease
Abstract & Commentary
Synopsis: Many patients who are shunted for normal pressure hydrocephalus have Alzheimer's pathology, and many go on to develop Alzheimer's disease.
Sources: Leinonen V, et al. Amyloid and tau proteins in cortical brain biopsy and Alzheimer's disease. Ann Neurol 2010;68:446-453.
Hamilton R, et al. Lack of shunt response in suspected idiopathic NPH with AD pathology. Ann Neurol 2010;68:535-540.
McKhann G, Mayeux R. Brain drain: A bottom-up approach to normal pressure hydrocephalus. Ann Neurol 2010;68:415-417.
A frustrating aspect of diagnostic criteria for neurodegenerative dementias is the lack of biological testing. For the most common dementia, Alzheimer's disease (AD), diagnosis is purely clinical, based on time course, severity, and memory-dominant cognitive profile. For the much rarer illness, normal pressure hydrocephalus (NPH), diagnosis is based primarily on the clinical triad involving gait, cognition, and bladder control, together with hydrocephalus on imaging. Laboratory testing based on the biology of disease currently is not part of diagnostic criteria for either illness. However, surgical treatment of suspected NPH affords the opportunity for brain biopsy, which then can be correlated with subsequent outcomes. Two articles1,2 and an accompanying editorial3 in the Oct. 2010 issue of Annals of Neurology consider such biopsies in suspected NPH, taken at the time of intraventricular pressure monitoring or shunt insertion, and the subsequent clinical diagnosis of AD.
The work by Leinonen and colleagues is the largest study to date of brain biopsies and subsequent outcomes. Over 15 years, 433 patients with at least 1 symptom of NPH and hydrocephalus on imaging underwent intraventricular pressure monitoring at the Kuopio University Hospital in Finland, and were followed clinically for a median 4.4 years. Right frontal biopsies were obtained at the time of monitoring, and immunostained with antibodies against Aβ and hyper-phosphorylated tau (HPτ). Of the 433 patients, 10% were Aβ+/HPτ+, 33% were Aβ+/HPτ-, and 57% were Aβ-/HPτ- (a single Aβ-/HPτ+ case was excluded). Subsequent development of clinical AD occurred in 81% of the Aβ+/HPτ+ cases, 33% of the Aβ+/HPτ- cases, and 5% of the Aβ-/HPτ- cases. Substantial fractions of all groups also developed non-AD dementias or mild cognitive impairment. Thus, the presence of both Aβ and tau pathology strongly predicted development of AD (OR 68 vs. Aβ-/HPτ-). Aβ alone also significantly predicted subsequent AD but less strongly (OR 11 vs. Aβ-/HPτ-). Of note, shunting was still of benefit even in patients with AD pathology on biopsy. Patients in both Aβ+/HPτ+ and Aβ+/HPτ- groups were shunted based on ICP monitoring criteria not described, and 75% of patients in each group improved, including patients who subsequently developed clinical AD.
Hamilton and colleagues at the University of Pennsylvania studied a smaller group of 47 patients with at least 2 symptoms of NPH and hydrocephalus on imaging. All were shunted and underwent brain biopsy, and 30 were eventually followed. Patients with moderate to severe AD pathology on biopsy had significantly worse cognition than those without pathology, and only 2 of 8 patients with moderate to severe pathology were improved 4 months after shunting, compared to 18 of 22 patients with no or mild pathology.
That the presence of AD pathology correlates with subsequent diagnosis of AD clearly makes sense. At the same time, these studies also raise many questions. For one, not all cases with AD pathology subsequently developed clinical AD. Indeed, even at autopsy, AD pathology may be present without any cognitive impairment. It is possible that this phenomenon may be purely a matter of limited follow-upgiven enough time all such patients might develop clinical symptoms. However, the question would then still remain, what factors allow some subjects to maintain cognition longer? Thus, even the ultimate biologic testing, brain biopsy, is limited by our understanding of disease biology. Our understanding of NPH is similarly limited, particularly with respect to predicting success from shunting. The study by Hamilton suggests that success is likelier in the absence of other explanatory pathologies; however, using physiologic criteria based on ICP monitoring, Leinonen found benefit independent of subsequent AD diagnosis.
These studies suggest that biological testing can improve both diagnosis and therapy. The next step in improving our understanding of these diseases is clearly to make aggressive, systematic use of newer methodologies, including CSF and brain Aβ and tau measurements and amyloid imaging, and to incorporate them into studies with long-term clinical follow-up.
1. Leinonen V, et al. Amyloid and tau proteins in cortical brain biopsy and Alzheimer's disease. Ann Neurol 2010;68:446-453.
2. Hamilton R, et al. Lack of shunt response in suspected idiopathic NPH with AD pathology. Ann Neurol 2010;68:535-540.
3. McKhann G, Mayeux R. Brain drain: A bottom-up approach to normal pressure hydrocephalus. Ann Neurol 2010;68:415-417.