brain tissue aβ42 levels are linked to shunt response in ...the number of steps and seconds needed...

CLINICAL ARTICLEJ Neurosurg 130:121–129, 2019

IdIopathIc normal pressure hydrocephalus (iNPH), a neurological disorder seen in the elderly popula-tion, is characterized by impairment in gait and bal-

ance, urinary incontinence, and cognitive decline.20 The pathophysiological background of iNPH without external causes has not been fully elucidated.2 Regardless of the underlying etiology, the clinical symptoms of iNPH can be alleviated with the insertion of a ventriculoperitoneal (VP) or a lumboperitoneal (LP) shunt for CSF diversion.15,30 Candidates for shunt insertion are selected on the basis of a detailed clinical examination and brain imaging, often followed by CSF dynamics tests,29 with one aim being to exclude patients with differential diagnoses such as Alz-heimer’s disease (AD).

Alzheimer’s disease is characterized by the accumula-tion of insoluble amyloid beta (Ab) aggregates and neu-rofibrillary tangles consisting of hyperphosphorylated tau protein (HPtau). The Ab aggregates are the result of oligo-merization of monomeric Ab species, mainly Ab40 and Ab42, into oligomers and protofibrils, which aggregate further into insoluble fibrils.7 To differentiate iNPH and AD in the workup preceding shunt placement, CSF levels of Ab42, total tau, and HPtau may be used as biomarkers. Decreased levels of Ab42 and elevated total tau and HPtau in CSF may indicate AD and a lower probability of symp-tom relief after VP shunt insertion in iNPH.37 Nonetheless, AD-related neuropathological changes such as Ab depos-its, neurofibrillary tangles, and reduced CSF Ab42 may

ABBREVIATIONS Aβ = amyloid beta; AD = Alzheimer’s disease; ELISA = enzyme-linked immunosorbent assay; HPtau = hyperphosphorylated tau protein; iNPH = idio-pathic normal pressure hydrocephalus; LOD = limit of detection; LP = lumboperitoneal; MMSE = Mini-Mental State Examination; VP = ventriculoperitoneal.SUBMITTED April 25, 2017. ACCEPTED July 24, 2017.INCLUDE WHEN CITING Published online January 19, 2018; DOI: 10.3171/2017.7.JNS171005.

Brain tissue Aβ42 levels are linked to shunt response in idiopathic normal pressure hydrocephalusSami Abu Hamdeh, MD,1 Johan Virhammar, MD, PhD,2 Dag Sehlin, PhD,3 Irina Alafuzoff, MD, PhD,4 Kristina Giuliana Cesarini, MD, PhD,1 and Niklas Marklund, MD, PhD1

1Department of Neuroscience, Section of Neurosurgery, Uppsala University; 2Department of Neuroscience, Neurology, Uppsala University Hospital, Uppsala University; 3Department of Public Health and Caring Sciences/Geriatrics, Uppsala University; and 4Department of Pathology, Uppsala University Hospital, Uppsala University, Uppsala, Sweden

OBJECTIVE The authors conducted a study to test if the cortical brain tissue levels of soluble amyloid beta (Aβ) reflect the propensity of cortical Aβ aggregate formation and may be an additional factor predicting surgical outcome following idiopathic normal pressure hydrocephalus (iNPH) treatment.METHODS Highly selective ELISAs (enzyme-linked immunosorbent assays) were used to quantify soluble Aβ40, Aβ42, and neurotoxic Aβ oligomers/protofibrils, associated with Aβ aggregation, in cortical biopsy samples obtained in patients with iNPH (n = 20), sampled during ventriculoperitoneal (VP) shunt surgery. Patients underwent pre- and postoperative (3-month) clinical assessment with a modified iNPH scale. The preoperative CSF biomarkers and the levels of soluble and insoluble Aβ species in cortical biopsy samples were analyzed for their association with a favorable outcome follow-ing the VP shunt procedure, defined as a ≥ 5-point increase in the iNPH scale.RESULTS The brain tissue levels of Aβ42 were negatively correlated with CSF Aβ42 (Spearman’s r = -0.53, p < 0.05). The Aβ40, Aβ42, and Aβ oligomer/protofibril levels in cortical biopsy samples were higher in patients with insoluble corti-cal Aβ aggregates (p < 0.05). The preoperative CSF Aβ42 levels were similar in patients responding (n = 11) and not responding (n = 9) to VP shunt treatment at 3 months postsurgery. In contrast, the presence of cortical Aβ aggregates and high brain tissue Aβ42 levels were associated with a poor outcome following VP shunt treatment (p < 0.05).CONCLUSIONS Brain tissue measurements of soluble Aβ species are feasible. Since high Aβ42 levels in cortical biopsy samples obtained in patients with iNPH indicated a poor surgical outcome, tissue levels of Aβ species may be associated with the clinical response to shunt treatment.https://thejns.org/doi/abs/10.3171/2017.7.JNS171005KEY WORDS amyloid-β; iNPH; Alzheimer’s disease; Aβ oligomers; hydrocephalus

J Neurosurg Volume 130 • January 2019 121©AANS 2019, except where prohibited by US copyright law

Unauthenticated | Downloaded 09/23/20 09:51 PM UTC

S. Abu Hamdeh et al.

J Neurosurg Volume 130 • January 2019122

be observed in iNPH patients.3,13,28 In addition, insoluble Ab aggregates are frequently found in iNPH brain tissue biopsy samples, despite normal CSF Ab42 levels.3,28

In iNPH, low CSF turnover may lead to impaired Ab clearance, decreased CSF Ab levels,34,39 and Ab aggre-gation. Insoluble Ab aggregates in cortical brain biopsy samples were negatively correlated to CSF Ab levels in iNPH.10 Moreover, similar to the Ab42 CSF levels24 the presence or absence of Ab aggregates in brain biopsy sam-ples are associated with the response to shunt surgery.28 Soluble Ab levels were measured in ventricular CSF and brain interstitial fluid of iNPH patients,10,13 although the corresponding brain tissue levels of Ab are unknown.

In this study, tissue levels of Ab40, Ab42, and Ab oligomers/protofibrils in cortical brain biopsy samples from iNPH patients were analyzed and correlated to lum-bar CSF levels of Ab42, total tau, and HPtau. We report higher levels of brain tissue Ab40 and Ab42 and Ab oligo-mers/protofibrils in iNPH patients with histopathological evidence of insoluble Ab aggregates, and we demonstrate an association between low tissue levels of soluble Ab42 and the likelihood of a favorable response to shunt surgery.

MethodsEthical Statement

The Regional Research Ethics Committee at Uppsala University granted permission for all studies included herein. Written informed consent was obtained from each patient, and all research was conducted in accordance with the ethical standards given in the Helsinki Declara-tion of 1975, as revised in 2008.

Patient PopulationPatients diagnosed with iNPH according to the inter-

national consensus guidelines29 (n = 20) and planned for VP shunt surgery were included in the study. All patients had ventricular enlargement (Evan’s index > 0.3)5 and progressively evolving symptoms, including a typical gait disturbance with or without cognitive decline or urinary incontinence. Selection of shunt candidates was done by a multidisciplinary iNPH team consisting of neurosurgeons, neurologists, and specially trained physiotherapists and oc-cupational therapists. The preoperative evaluation included brain MRI, high-volume lumbar puncture—i.e., tap test—and infusion tests. Patients with secondary causes of iNPH or clinical and radiological signs of AD or other neurode-generative diseases were not included. The multidisciplin-ary iNPH team evaluated the patients preoperatively and at 3 months postoperatively. A postoperative CT scan (mean 41 ± 22 days postoperatively) was obtained in all patients.

Clinical EvaluationAll patients had symptoms indicative of iNPH, includ-

ing a typical gait disturbance with or without cognitive de-cline or urinary incontinence, which were gradually dete-riorating. Patients were clinically evaluated preoperatively and at follow-up 3 months postoperatively with a modified iNPH scale.8 The iNPH scale is the weighted mean score of assessment in 4 domains: gait, neuropsychology, static balance, and continence. Gait was evaluated by measuring

the number of steps and seconds needed to walk 10 m at a self-chosen pace and an ordinal rating of gait. Neuropsy-chology domain scores were the mean of converted scores from 4 available assessments/tasks, the Mini-Mental State Examination (MMSE), Grooved pegboard (Lafayette In-strument Co.), and Swedish Stroop test, which consists of 2 tasks.9 MMSE scores were converted to a minimum–maximum range of 0–100 (100/30 × MMSE score). The Swedish Stroop test used at our department is shortened compared with the test described for the iNPH scale and consists of 24 rectangles in contrast to the 100 rectangles used in the previous study.8 This was adjusted for using the formula 100/24, which was multiplied by the time used by the patient to complete the task. Static balance and con-tinence were assessed with ordinal scales. The iNPH scale consists of scores from 0 to 100, where 100 represents nor-mal performance by healthy individuals in a typical iNPH age range of 70–74 years.8 An increase of at least 5 points on the iNPH scale at follow-up was considered a signifi-cant shunt-related response.8,16,36

CSF AnalysisLumbar CSF samples were obtained as part of the

preoperative evaluation. Levels of Ab42, total tau, and HPtau in CSF were assessed by routine clinical analysis using commercial enzyme-linked immunosorbent assay (ELISA) kits following the manufacturer’s instructions.3

Tissue HandlingBrain biopsy samples were taken in the right frontal

area anterior to the coronal suture at the time of shunt sur-gery using the same corticotomy, performed with a sharp syringe to avoid thermal injury. Biopsy needles with a 14-gauge (2.11-mm) diameter and an 8-mm side cutting window (Elekta Instrument AB) were used, and the biopsy samples included cortical and subcortical tissue. A first bi-opsy sample was put in prelabeled 1.5-ml Eppendorf tubes and stored in a -80°C freezer until analyzed. A second bi-opsy sample was immediately placed in 4% formaldehyde (HistoLab Products AB, catalogue no. 02176) at surgery.

The samples were fixed in 4% formaldehyde for 24 hours, paraffin embedded using Histovax (HistoLab Products AB), and processed by hardware Tissue-Tek VIP (Sakura). Immunohistochemistry was performed on 6-mm microtome sections according to previously published protocols3 using H & E staining and Ab (6F/3D, M0872; dilution 1:100, pretreatment with 80% formic acid for 1 hour; Dako) and HPtau immunohistochemistry (AT8, MN1020; dilution 1:500; Thermo Scientific). The Dako EnVision FLEX detection system was used for visualiza-tion of staining results and sections assessed using light microscopy.

Protein ExtractionThe brain biopsy samples were homogenized for 60

seconds (POLYTRON PT 1200, Kinematica) with 1 ml of lysis buffer (10 mM Tris-HCl pH 7.4, 0.15 M NaCl, 1mM EDTA, and phosphate-buffered saline containing 1% b-octyl glucopyranoside). Protease inhibitor cocktail (10 ml) was added, and after homogenization, the samples were


J Neurosurg Volume 130 • January 2019 123


incubated for 90 minutes at 4°C. The tissue lysates were centrifuged for 30 minutes (10,000g at 4°C). The total pro-tein concentration in the supernatant was determined us-ing the DC Protein Assay Kit (BioRad Laboratories). The DC assay was carried out using a 96-well microtiter plate reader (model 680, BioRad Laboratories).

Aβ40 and Aβ42 ELISA AnalysesThe brain extracts were denatured by 5 minutes of boil-

ing in 1% sodium dodecyl sulfate to obtain a preparation of monomeric Abx-40 and Abx-42, since soluble Ab oligo-mers and protofibrils are poorly detected with ELISA.35 Samples were diluted in ELISA sample buffer and ana-lyzed in duplicates with WAKO Ab40 and WAKO Ab42 High Sensitive ELISA kits (Wako Chemicals USA, Inc.). All samples, as well as the Ab standard, had the same final concentration of sodium dodecyl sulfate (0.1%) and lysis buffer (10%). This procedure does not interfere with Ab detection in these ELISA kits.31 Analyses were conducted by a researcher (D.S.) blinded to the immunohistochemi-cal Ab results and the clinical information.

Aβ Oligomer/Protofibril ELISAA homogenous sandwich ELISA, using the 82E1 an-

tibody (specific to the N-terminal neoepitope of Ab) as both the capture and detection antibody, was performed to

analyze the levels of Ab oligomers and protofibrils.38 This ELISA detects soluble Ab oligomers in the range from di-mers up to large protofibrils, but it will not detect any mo-nomeric Ab. Ninety-six-well half-area plates were coated overnight with 12.5 ng/well of 82E1 (IBL International/Tecan Trading AG) according to previous protocols.33 Bio-tinylated 82E1 (0.25 mg/ml) and streptavidin-conjugated horseradish peroxidase (Mabtech AB) were used for de-tection, developed with K blue aqueous TMB substrate (Neogen Corp.), and read at 450 nm.

Statistical AnalysisStatistica 12.0 (Stat Soft, Inc.) was used for descriptive

and analytical statistics. With the exception of age, ex-pressed as mean ± SD, the data were not normally distrib-uted and are expressed as median and range. The Fisher exact test was used for comparison of proportions and the Mann-Whitney U-test to determine significant differences between groups. Spearman rank-order correlations were used to analyze the relationships between variables. A p value < 0.05 was considered statistically significant.

ResultsDetailed demographic and clinical characteristics are

shown in Table 1. The mean age of the 13 male and 7 female patients with iNPH was 73.5 ± 5.4 years. Nineteen

TABLE 1. Patient characteristics in iNPH patients with and without insoluble Aβ aggregates

Patient No.Age (yrs) Sex

Evan’s Index

CSF Levels (ng/L)* iNPH Scale ScoreNeuropathologyAβ42 Tau HPtau Preop Postop (3 mos)

iNPH (Aβ+) 1 76 M 0.32 548 490 61 46 48 Aβ aggregates 2 70 M 0.37 359 279 42 67 80 Aβ aggregates, HPtau 3 86 M 0.40 437 291 38 40 45 Aβ aggregates 4 72 M 0.31 676 211 34 54 57 Aβ aggregates 5 65 F 0.33 364 267 33 32 59 Aβ aggregates, HPtau 6 70 F 0.36 649 156 23 75 70 Aβ aggregates 7 77 M 0.44 576 272 35 42 23 Aβ aggregates 8 74 F 0.30 251 <75 17 48 50 Aβ aggregates 9 70 M 0.38 462 196 28 42 22 Aβ aggregates 10 80 F 0.31 756 112 19 58 51 Aβ aggregates 11 80 F 0.43 416 236 24 21 22 Aβ aggregates 12 77 M 0.34 448 206 36 49 57 Aβ aggregatesiNPH (Aβ-) 1 67 M 0.38 673 197 29 83 94 Gliosis 2 72 M 0.39 639 204 28 65 75 No pathology 3 71 M 0.36 365 130 18 42 57 HPtau 4 71 M 0.32 498 99 24 59 68 No pathology 5 78 M 0.36 759 223 24 46 55 Gliosis, HPtau 6 64 F 0.38 887 370 48 58 48 Gliosis 7 75 F 0.30 NA NA NA 54 67 Gliosis 8 76 M 0.38 541 115 19 68 78 Gliosis

NA = not established.* Reference intervals: CSF Aβ42, > 550 ng/L; CSF-tau, < 400 ng/L; and CSF HPtau, < 80 ng/ L.




patients fulfilled the criteria for probable iNPH and 1 pa-tient for possible iNPH.29 Median Evan’s index was 0.36 (range 0.30–0.44). Two patients required a shunt revision due to a misplaced proximal catheter outside the ventricle (n = 1) or distal catheter dislocation (n = 1). Two patients with signs of overdrainage on postoperative CT scans were successfully treated by increasing the shunt valve pressure setting. There were no adverse events related to the brain biopsy sampling procedure.

On immunohistochemical analysis, insoluble Ab ag-gregates were found in 12 patients (Ab+), and were absent in 8 patients (Ab-), while tau pathology was found in 4 of

20 patients (Table 1; Fig. 1). Age and Evan’s index were similar in the Ab+ and Ab- groups (p = 0.34 and p = 0.85; Table 2).

Clinical Evaluation Following High-Volume CSF Tap Test: No Association Between Clinical Response and Aβ Status

A clinical improvement, measured by the iNPH scale, following high-volume CSF tap was observed in 14 of 20 patients. Five patients did not improve following the CSF tap test, and, in 1 case, a lumbar puncture could not be performed for patient-related reasons. There was no as-sociation between the clinical response to the CSF tap test

TABLE 2. Statistical analysis of 20 iNPH patients with and without the presence of insoluble Aβ aggregates

FactorGroup* p

ValueAβ+ (n = 12) Aβ- (n = 8)

Mean age in yrs (range) 75 (65–86) 72 (64–78) 0.34M/F ratio 7:5 6:2 0.44Evan’s index 0.35 (0.30–0.44) 0.37 (0.30–0.39) 0.85Preop Gait domain 28 (5–62) 55 (18–89) 0.06 Neuropsychology domain 49 (23–89) 71 (53–100) 0.06 Balance domain 67 (17–83) 67 (33–67) 0.68 Continence domain 60 (0–100) 50 (0–80) 0.52 iNPH scale score 47 (21–75) 58 (42–83) 0.123 mos postop Gait domain 37 (10–82) 60 (16–100) 0.07 Neuropsychology domain 50 (13–88) 82 (70–100) 0.002 Balance domain 67 (17–83) 67 (50–83) 0.27 Continence domain 60 (0–100) 70 (0–100) 0.73 iNPH scale score 51 (22–80) 67 (48–94) 0.06CSF levels in ng/L Aβ42 455 (251–756) 639 (365–887) 0.14 CSF-tau 223 (75–490) 197 (99–370) 0.34 HPtau 34 (17–61) 24 (18–48) 0.34

Values are presented as the median (range) unless otherwise stated.* Reference intervals: CSF Aβ42, > 550 ng/L; CSF-tau, < 400 ng/L; and CSF HPtau, < 80 ng/ L.

FIG. 1. Photomicrographs of insoluble Aβ aggregates stained with antibodies to Aβ in cortical biopsy samples obtained in patients with iNPH. A: Absence of insoluble Aβ aggregates. B: Sparse accumulation of insoluble Aβ aggregates. C: Extensive accumula-tion of insoluble Aβ aggregates.




prior to surgery and CSF-Ab42 (p = 0.09), CSF-tau (p = 0.21), or CSF-Hptau (p = 0.17) or brain tissue Ab40 (p = 0.96), Ab42 (p = 0.56), or Ab oligomers/protofibril (p = 0.57). In addition, the number of patients with a clinical response to the CSF tap test did not significantly differ between the Ab+ (n = 8 improved) and Ab- groups (n = 6 improved) (p = 0.91).

Preoperative Clinical Evaluation and Lumbar CSF Samples: No Significant Difference in iNPH Symptoms or Lumbar CSF Biomarkers Between Patients With and Without Insoluble Aβ Aggregates

Individual preoperative and 3-month follow-up iNPH scale scores are provided in Table 1. There were near-significant differences in the preoperative median values between the Ab+ and Ab- groups in the gait (p = 0.06) and the neuropsychology (p = 0.06) domains, while pre-operative median values for the balance (p = 0.68) and continence (p = 0.52) domains and total iNPH scale score (p = 0.12) were similar between the Ab+ and Ab- groups (Table 2).

Median CSF Ab42 levels were 541 ng/L (range 251–887 ng/L) and were similar between the Ab+ and Ab- groups (p = 0.14; Table 2). Median CSF-tau and HPtau levels were 206 ng/L (range 75–490 ng/L) and 28 ng/L (range 17–61

ng/L), respectively, with no differences between the Ab+ and Ab- groups (p = 0.34 for both comparisons; Table 2).

Brain Tissue Levels of Soluble Aβ40 and Aβ42: Significantly Higher Levels of Aβ40 and Aβ42 in iNPH Patients With Insoluble Aβ Aggregates

The limit of detection (LOD) for the Ab40 and Ab42 ELISA assays were 26 ng/l and 46 ng/l, respectively. No patient had brain tissue levels below the LOD for either Ab40 or Ab42.

Median brain tissue Ab40 levels were 340 ng/L (range 104–21,698 ng/L), with significantly higher levels in the Ab+ than the Ab- group (median 416 ng/L [range 104–21,698 ng/L] and median 187 ng/L [range 113–368 ng/L], respectively; p = 0.012; Fig. 2). Median brain tissue Ab42 levels were 3234 ng/L (range 59–115,614 ng/L) . The Ab+ group had significantly higher levels of brain tissue Ab42 than the Ab- group (median 16,360 ng/L [range 180–115,614 ng/L] and median 153 ng/L [range 59–555 ng/L], respectively; p = 0.0001; Fig. 2). Brain tissue levels of Ab40 and Ab42 did not correlate significantly to patient age (p = 0.61), preoperative iNPH scale score (p = 0.25), or the total tau (p = 0.69) and HPtau (p = 0.64) CSF levels. Brain tissue levels of Ab42 had a moderately negative cor-relation to the CSF levels of Ab42 (Spearman’s r = -0.53,

FIG. 2. Dot plots of increased levels of brain tissue Aβ40, Aβ42, and Aβ oligomers/protofibrils in cortical biopsy samples of iNPH patients with insoluble Aβ aggregates. A: Individual and median levels of brain tissue Aβ40 in patients with (Aβ+) and without (Aβ-) insoluble Aβ aggregates. Levels of brain tissue Aβ40 were significantly higher in Aβ+ than Aβ- patients (p = 0.012). B: Lev-els of brain tissue Aβ42 were significantly higher in Aβ+ than Aβ- patients (p = 0.0001). C: Levels of brain tissue Aβ oligomers/protofibrils were significantly higher in Aβ+ than Aβ- patients (p = 0.012). D: Brain tissue levels of Aβ42 had a moderately nega-tive correlation with the CSF Aβ42 levels (Spearman’s r = -0.53, p = 0.02). E: Brain tissue levels of Aβ oligomers/protofibrils had a strong positive correlation with brain tissue Aβ40 levels (Spearman’s r = 0.87, p < 0.0001). F: Brain tissue levels of Aβ oligomers/protofibrils had a strong positive correlation with brain tissue Aβ42 levels (Spearman’s r = 0.86, p < 0.0001). OL/PF = oligomers/protofibrils.




p = 0.02; Fig. 2). Brain tissue Ab40 and Ab42 levels were positively correlated (Spearman’s r = 0.80, p < 0.0001).

Brain Tissue Levels of Aβ Oligomers/Protofibrils: Aβ Oligomers/Protofibrils Correlated to Soluble Aβ40 and Aβ42 Levels

The LOD for the Ab oligomer/protofibril ELISA assay was 135 ng/L. No oligomers/protofibrils were detected in 6 patients. Five patients with levels below the LOD were assigned the value 68 ng/L to separate them from those in whom no oligomers/protofibrils were found. In 1 patient, the amount of brain extract was insufficient for this ELISA assay.

Median brain tissue oligomer/protofibril levels were 68 ng/L (range 0–3348 ng/L), significantly higher in the Ab+ than the Ab- group (median 401 ng/L [range 0–2248 ng/L] and median 34 ng/L [range 0–68 ng/L], respectively; p = 0.012; Fig. 2). Brain tissue levels of oligomers/protofi-brils did not correlate to patient age (p = 0.58), preopera-tive iNPH scale score (p = 0.78), or the CSF levels of Ab42 (p = 0.18), total tau (p = 0.94), or HPtau (p = 0.93). Brain tissue levels of oligomers/protofibrils had a positive cor-relation to Ab40 (Spearman’s r = 0.87, p < 0.0001; Fig. 2) and Ab42 levels (Spearman’s r = 0.86, p < 0.0001; Fig. 2).

Postoperative Clinical Evaluation: Low Brain Tissue Aβ42 Levels Associated With a Favorable Shunt Response

Three months postoperatively, a favorable shunt re-

sponse was seen in 11 of 20 patients. There were no as-sociations between shunt response and age (p = 0.82), sex (p = 0.08), Evan’s index (p = 0.88), or preoperative iNPH scale score (p = 0.55) or CSF levels of Ab42 (p = 0.32), total tau (p = 0.84), or HPtau (p = 1.0).

Preoperative and postoperative iNPH scale scores for Ab+ and Ab- patients are shown in Fig. 3. Both groups had similar postoperative median values for the gait (p = 0.07), balance (p = 0.27), and continence (p = 0.73) do-mains, as well as the total iNPH scale score (p = 0.06). The Ab+ group had lower median values for the neuropsychol-ogy domain than the Ab- group (p = 0.002).

A favorable shunt response was observed in 4 of 12 patients in the Ab+ group and 7 of 8 patients in the Ab- group. The absence of insoluble Ab aggregates on immu-nohistochemistry was associated with a favorable shunt response (p = 0.02). The brain tissue levels of Ab40 were similar in patients with a favorable shunt response and nonresponders (median Ab40 195 ng/L [range 104–3369 ng/L] and median 411 ng/L [range 143–21,698 ng/L], re-spectively; p = 0.08; Fig. 3), while brain tissue Ab42 lev-els were lower in shunt responders than in nonresponders (median Ab42 180 ng/L [range 59–81,703 ng/L] and me-dian 10,233 ng/L [range 180–115,614 ng/L], respectively; p = 0.038; Fig. 3). The association between oligomers/proto-fibrils and a favorable shunt response (median oligomers/protofibrils in responders 34 ng/L [range 0–2939 ng/L] compared with nonresponders 302 ng/L [range 0–3348 ng/L]) did not reach statistical significance (p = 0.053).

FIG. 3. Insoluble Aβ aggregates and high brain tissue levels of soluble Aβ42 are associated with a poor response to shunt treat-ment. A: iNPH scale scores prior to (preoperative) and at the 3-month follow-up examination (postoperative) in patients with (Aβ+) and without (Aβ-) insoluble Aβ aggregates. An increase of at least 5 points at follow-up is considered a significant response to shunt treatment. A significant shunt response at follow-up was observed in 7 of 8 Aβ- patients and in only 4 of 12 Aβ+ patients (p = 0.02). B: Gait domain scores prior to (preoperative) and at 3-month follow-up examination (postoperative) in Aβ+ and Aβ- patients. C: Neuropsychology domain scores prior to (preoperative) and at 3-month follow-up examination (postoperative) in Aβ+ and Aβ- patients. The Aβ+ group had significantly lower scores than the Aβ- group at follow-up (p = 0.002). D: Dot plot of individual and median levels of brain tissue Aβ40 in iNPH patients with (responders) and without (nonresponders) a significant shunt response. The Aβ40 brain tissue levels were similar in responders and nonresponders (p = 0.08). E: Nonresponders had significantly higher Aβ42 levels in brain tissue than responders (p = 0.038). F: There was a trend toward higher Aβ oligomer/proto-fibril brain tissue levels in responders than in nonresponders (p = 0.053).




DiscussionIn the present study, we presented the first measure-

ments of brain tissue levels of soluble Ab40, Ab42, and Ab oligomers/protofibrils from cortical biopsy samples ob-tained in patients with iNPH. Our main findings were that the brain tissue levels of Ab40, Ab42, and Ab oligomers/protofibrils were higher in cortical brain tissue obtained in iNPH patients with histopathological evidence of insoluble Ab aggregates (Ab+ patients). Additionally, high levels of brain tissue Ab42 were associated with a poor response to VP shunt treatment.

Our study revealed considerable variability in the levels of brain tissue Ab42 among iNPH patients, in contrast to the CSF Ab42 levels. Nonetheless, the Ab42 tissue levels in cortical biopsy samples were negatively correlated to the CSF Ab42 levels, although, to date, the Ab brain levels in healthy individuals are unknown. Soluble Ab monomers were previously measured in ventricular CSF and brain interstitial fluid of patients with iNPH and traumatic brain injury (TBI).1,10,13 Although AD-related neuropathological changes are common features in iNPH3,13,28 and comorbid-ity is frequent,6 we cannot exclude the possibility that the accumulation of Ab observed in our study population is a manifestation of early, incipient AD.

While a global estimate of Ab levels may be achieved by analyzing circulating CSF, a small frontal brain biopsy sample may reflect only a limited brain region that may not be representative of the global tissue levels of Ab. Car-bon 11–labeled Pittsburgh compound B ([11C]PiB) PET and its analog fluorine-18 [18F] flutemetamol PET may quantify insoluble fibrillar Ab deposits in vivo.12,17,25 In iNPH, [11C]PiB and [18F]flutemetamol uptake seems cor-related to immunohistochemical Ab aggregates.19,40 How-ever, the PET technique cannot estimate soluble Ab spe-cies in brain tissue that was analyzed in the present study. Consistent with results of previous studies, patients with Ab pathology on brain biopsy tended to have lower inter-stitial fluid Ab42 levels,10,11 likely due to accumulation of Ab42 from the interstitial fluid into insoluble aggregates. In contrast, we observed the highest levels of soluble Ab in patients with immunohistochemical evidence of insoluble Ab aggregates, and the levels of Ab40 and Ab42 were cor-related to the levels of soluble Ab aggregates in the form of oligomers/protofibrils. The findings from our study sug-gest that in iNPH patients with a tendency for formation of insoluble Ab aggregates, a large reservoir of soluble Ab, including toxic oligomers and protofibrils, may exist.

In the present study, 12 of 20 of patients exhibited in-soluble Ab aggregates, although there were no significant differences between Ab+ and Ab- patients in terms of preoperative clinical symptoms and CSF levels of Ab42, total tau, or HPtau. The CSF circulation helps to clear the interstitial fluid space from toxic metabolites, includ-ing Ab. Thus, the deposition of Ab in iNPH may be the result of increased CSF outflow resistance,34 explaining the increase in soluble Ab in CSF following shunt pro-cedures.13,23 Amyloid-b deposits are common findings in normal aging and exist in 39%–82% of elderly nonde-mented brains on postmortem examination.4 Nevertheless, we found a significant difference between the Ab+ and Ab- groups in cognitive symptoms at 3 months postsur-

gery. While improvement of gait symptoms was observed, these data suggest that Ab pathology in iNPH restricts resolution of cognitive symptoms.14 In the preoperative evaluation process, care was taken to exclude patients with signs of AD. Patients were not included in the study if the onset of the cognitive dysfunction preceded the gait dis-turbance, if there were clinical signs of cortical dementia, or if there were other typical AD features such as cortical atrophy on brain imaging. However, in patients with iNPH, concurrent AD as comorbidity cannot be completely ruled out, and we cannot exclude that some iNPH patients in our study will proceed to develop classic AD.

Interestingly, in the Ab- group, gliosis was a prevalent finding, seen in 5 of 8 biopsy samples. Gliosis is a common observation in association to vascular alterations. There are few postmortem studies carried out on iNPH subjects, one being the postmortem findings in 10 patients with pre-sumed normal pressure hydrocephalus.18 In that study, it was reported that in a substantial number of iNPH patients with cognitive impairment, but without Ab pathology, vas-cular alterations were seen.

Following CSF diversion, a favorable outcome is seen in 50%–80% of iNPH patients.16,36 The response to shunt treatment may be influenced by inadequate preoperative evaluation, technical aspects, and complications of the shunting procedure, including obstruction, catheter mis-placement, or postoperative infection.26 One important trial in the workup to select patients suitable for shunt in-sertion, the high-volume CSF tap test, has a high predictive value for a favorable shunt response. However, the negative predictive value is low, and a lack of improvement follow-ing the test does not render the patient ineligible for CSF diversion.21 Possibly, a lumbar drain for continuous CSF drainage could yield a higher specificity. Since a lumbar drain requires hospital admission and is associated with the risk of complications such as radicular pain or menin-gitis,22 this procedure is not routinely used in our depart-ment. In our study, the decision to insert a shunt in patients who did not experience a clinical improvement following the CSF tap test was based on clinical and radiological factors as well as the results from the infusion tests. Nev-ertheless, 3 of 5 patients with no improvement following the CSF tap test eventually had a favorable outcome after shunt insertion. In addition, the incidence of postoperative complications was low.

The relation between insoluble Ab aggregates in brain biopsy samples and the response to shunt placement has been previously documented28 and was confirmed in our report. Previous studies have also shown an association between CSF levels of Ab42 and outcome following sur-gery,24,27 which was not confirmed in our study, possibly due to the small sample size. However, we observed an associa-tion between a favorable shunt response and a low brain tis-sue Ab42 level, indicating that brain tissue Ab42 levels are also related to the response to shunting. Furthermore, there was a clear trend for an association between Ab oligomers/protofibrils levels and a favorable shunt response. On the other hand, neither brain tissue Ab42 nor Ab oligomer/pro-tofibril levels were correlated with patients’ age, preopera-tive symptoms, or CSF levels of tau or HPtau.

This study has a number of limitations. First, due to




the rather small patient cohort, the findings need to be confirmed in larger future studies. Second, neither the temporal nor the topographic distribution of brain tissue Ab deposits could be assessed, and we can therefore not ascertain whether the measured Ab species are manifes-tations of early AD and/or decreased CSF Ab clearance. Preferably, future studies should measure long-term out-come beyond the 3-month period and include longitudinal assessment of insoluble Ab aggregates in the CSF. Finally, additional studies using antibody-based PET ligands to de-tect insoluble Ab aggregates could help determine tempo-ral changes in Ab deposition in iNPH.32

ConclusionsWe conclude that brain tissue levels of Ab42 are associ-

ated with the postoperative outcome following shunt inser-tion in iNPH patients and may be an important indicator of the likelihood of a favorable outcome after CSF diversion.

AcknowledgmentsWe thank Sravani Musunuri for technical assistance with

homogenization of brain tissue samples.The study was supported by funds from the Swedish Brain

Foundation and Uppsala University Hospital ALF funds.

References 1. Brody DL, Magnoni S, Schwetye KE, Spinner ML, Esparza

TJ, Stocchetti N, et al: Amyloid-b dynamics correlate with neurological status in the injured human brain. Science 321:1221–1224, 2008

2. Daou B, Klinge P, Tjoumakaris S, Rosenwasser RH, Jabbour P: Revisiting secondary normal pressure hydrocephalus: does it exist? A review. Neurosurg Focus 41(3):E6, 2016

3. Elobeid A, Laurell K, Cesarini KG, Alafuzoff I: Correlations between mini-mental state examination score, cerebrospinal fluid biomarkers, and pathology observed in brain biopsies of patients with normal-pressure hydrocephalus. J Neuropathol Exp Neurol 74:470–479, 2015

4. Elobeid A, Libard S, Leino M, Popova SN, Alafuzoff I: Al-tered proteins in the aging brain. J Neuropathol Exp Neurol 75:316–325, 2016

5. Evans WA Jr: AN encephalographic ratio for estimating ventricular enlargement and cerebral atrophy. Arch Neurol Psychiatry 47:931–937, 1942

6. Golomb J, Wisoff J, Miller DC, Boksay I, Kluger A, Weiner H, et al: Alzheimer’s disease comorbidity in normal pressure hydrocephalus: prevalence and shunt response. J Neurol Neurosurg Psychiatry 68:778–781, 2000

7. Hardy J, Selkoe DJ: The amyloid hypothesis of Alzheimer’s disease: progress and problems on the road to therapeutics. Science 297:353–356, 2002

8. Hellström P, Klinge P, Tans J, Wikkelsø C: A new scale for assessment of severity and outcome in iNPH. Acta Neurol Scand 126:229–237, 2012

9. Hellström P, Scharin M: Stroop, the bewildering effect. Svensk Neuropsykologi 13:12–15, 2001 (Swedish)

10. Herukka SK, Rummukainen J, Ihalainen J, von Und Zu Fraunberg M, Koivisto AM, Nerg O, et al: Amyloid-b and tau dynamics in human brain interstitial fluid in patients with suspected normal pressure hydrocephalus. J Alzheimers Dis 46:261–269, 2015

11. Hong S, Quintero-Monzon O, Ostaszewski BL, Podlisny DR, Cavanaugh WT, Yang T, et al: Dynamic analysis of amyloid

b-protein in behaving mice reveals opposing changes in ISF versus parenchymal Ab during age-related plaque formation. J Neurosci 31:15861–15869, 2011

12. Ikonomovic MD, Klunk WE, Abrahamson EE, Mathis CA, Price JC, Tsopelas ND, et al: Post-mortem correlates of in vivo PiB-PET amyloid imaging in a typical case of Alzhei-mer’s disease. Brain 131:1630–1645, 2008

13. Jeppsson A, Zetterberg H, Blennow K, Wikkelsø C: Idio-pathic normal-pressure hydrocephalus: pathophysiology and diagnosis by CSF biomarkers. Neurology 80:1385–1392, 2013

14. Kazui H, Kanemoto H, Yoshiyama K, Kishima H, Suzuki Y, Sato S, et al: Association between high biomarker probability of Alzheimer’s disease and improvement of clinical outcomes after shunt surgery in patients with idiopathic normal pres-sure hydrocephalus. J Neurol Sci 369:236–241, 2016

15. Kazui H, Miyajima M, Mori E, Ishikawa M: Lumboperitone-al shunt surgery for idiopathic normal pressure hydrocepha-lus (SINPHONI-2): an open-label randomised trial. Lancet Neurol 14:585–594, 2015

16. Klinge P, Hellström P, Tans J, Wikkelsø C: One-year outcome in the European multicentre study on iNPH. Acta Neurol Scand 126:145–153, 2012

17. Leinonen V, Alafuzoff I, Aalto S, Suotunen T, Savolainen S, Någren K, et al: Assessment of b-amyloid in a frontal cortical brain biopsy specimen and by positron emission tomography with carbon 11-labeled Pittsburgh Compound B. Arch Neu-rol 65:1304–1309, 2008

18. Leinonen V, Koivisto AM, Savolainen S, Rummukainen J, Sutela A, Vanninen R, et al: Post-mortem findings in 10 patients with presumed normal-pressure hydrocephalus and review of the literature. Neuropathol Appl Neurobiol 38:72–86, 2012

19. Leinonen V, Rinne JO, Virtanen KA, Eskola O, Rummu-kainen J, Huttunen J, et al: Positron emission tomography with [18F]flutemetamol and [11C]PiB for in vivo detection of cerebral cortical amyloid in normal pressure hydrocephalus patients. Eur J Neurol 20:1043–1052, 2013

20. Martín-Láez R, Caballero-Arzapalo H, López-Menéndez LA, Arango-Lasprilla JC, Vázquez-Barquero A: Epidemiology of idiopathic normal pressure hydrocephalus: a systematic review of the literature. World Neurosurg 84:2002–2009, 2015

21. Mihalj M, Dolić K, Kolić K, Ledenko V: CSF tap test—obso-lete or appropriate test for predicting shunt responsiveness? A systemic review. J Neurol Sci 362:78–84, 2016

22. Mori E, Ishikawa M, Kato T, Kazui H, Miyake H, Miyajima M, et al: Guidelines for management of idiopathic normal pressure hydrocephalus: second edition. Neurol Med Chir (Tokyo) 52:775–809, 2012

23. Moriya M, Miyajima M, Nakajima M, Ogino I, Arai H: Im-pact of cerebrospinal fluid shunting for idiopathic normal pressure hydrocephalus on the amyloid cascade. PLoS One 10:e0119973, 2015

24. Nakajima M, Miyajima M, Ogino I, Akiba C, Sugano H, Hara T, et al: Cerebrospinal fluid biomarkers for prognosis of long-term cognitive treatment outcomes in patients with idiopathic normal pressure hydrocephalus. J Neurol Sci 357:88–95, 2015

25. Nelissen N, Van Laere K, Thurfjell L, Owenius R, Vanden-bulcke M, Koole M, et al: Phase 1 study of the Pittsburgh compound B derivative 18F-flutemetamol in healthy volun-teers and patients with probable Alzheimer disease. J Nucl Med 50:1251–1259, 2009

26. Nigim F, Critchlow JF, Schneider BE, Chen C, Kasper EM: Shunting for hydrocephalus: analysis of techniques and fail-ure patterns. J Surg Res 191:140–147, 2014

27. Patel S, Lee EB, Xie SX, Law A, Jackson EM, Arnold SE, et al: Phosphorylated tau/amyloid beta 1-42 ratio in ventricular




cerebrospinal fluid reflects outcome in idiopathic normal pressure hydrocephalus. Fluids Barriers CNS 9:7, 2012

28. Pomeraniec IJ, Bond AE, Lopes MB, Jane JA Sr: Concur-rent Alzheimer’s pathology in patients with clinical normal pressure hydrocephalus: correlation of high-volume lumbar puncture results, cortical brain biopsies, and outcomes. J Neurosurg 124:382–388, 2016

29. Relkin N, Marmarou A, Klinge P, Bergsneider M, Black PM: Diagnosing idiopathic normal-pressure hydrocephalus. Neu-rosurgery 57 (3 Suppl):S4–S16, ii–v, 2005

30. Savolainen S, Hurskainen H, Paljärvi L, Alafuzoff I, Va-palahti M: Five-year outcome of normal pressure hydroceph-alus with or without a shunt: predictive value of the clinical signs, neuropsychological evaluation and infusion test. Acta Neurochir (Wien) 144:515–523, 2002

31. Sehlin D, Englund H, Simu B, Karlsson M, Ingelsson M, Nikolajeff F, et al: Large aggregates are the major soluble Ab species in AD brain fractionated with density gradient ultra-centrifugation. PLoS One 7:e32014, 2012

32. Sehlin D, Fang XT, Cato L, Antoni G, Lannfelt L, Syvänen S: Antibody-based PET imaging of amyloid beta in mouse mod-els of Alzheimer’s disease. Nat Commun 7:10759, 2016

33. Sehlin D, Söllvander S, Paulie S, Brundin R, Ingelsson M, Lannfelt L, et al: Interference from heterophilic antibodies in amyloid-b oligomer ELISAs. J Alzheimers Dis 21:1295–1301, 2010

34. Silverberg GD, Mayo M, Saul T, Rubenstein E, McGuire D: Alzheimer’s disease, normal-pressure hydrocephalus, and se-nescent changes in CSF circulatory physiology: a hypothesis. Lancet Neurol 2:506–511, 2003

35. Stenh C, Englund H, Lord A, Johansson AS, Almeida CG, Gellerfors P, et al: Amyloid-b oligomers are inefficiently measured by enzyme-linked immunosorbent assay. Ann Neurol 58:147–150, 2005

36. Sundstrom N, Malm J, Laurell K, Lundin F, Kahlon B, Cesa-rini KG, et al: Incidence and outcome of surgery for adult hy-drocephalus patients in Sweden. Br J Neurosurg 31:21–27, 2017

37. Tarnaris A, Toma AK, Chapman MD, Keir G, Kitchen ND,

Watkins LD: Use of cerebrospinal fluid amyloid-b and total tau protein to predict favorable surgical outcomes in patients with idiopathic normal pressure hydrocephalus. J Neurosurg 115:145–150, 2011

38. Tucker S, Möller C, Tegerstedt K, Lord A, Laudon H, Sjödahl J, et al: The murine version of BAN2401 (mAb158) selec-tively reduces amyloid-b protofibrils in brain and cerebrospi-nal fluid of tg-ArcSwe mice. J Alzheimers Dis 43:575–588, 2015

39. Weller RO, Djuanda E, Yow HY, Carare RO: Lymphatic drainage of the brain and the pathophysiology of neurological disease. Acta Neuropathol 117:1–14, 2009

40. Wolk DA, Grachev ID, Buckley C, Kazi H, Grady MS, Trojanowski JQ, et al: Association between in vivo fluorine 18-labeled flutemetamol amyloid positron emission tomog-raphy imaging and in vivo cerebral cortical histopathology. Arch Neurol 68:1398–1403, 2011

DisclosuresThe authors report no conflict of interest concerning the materi-als or methods used in this study or the findings specified in this paper.

Author ContributionsConception and design: Marklund, Cesarini. Acquisition of data: Abu Hamdeh, Sehlin. Analysis and interpretation of data: Marklund, Abu Hamdeh, Virhammar, Sehlin, Alafuzoff. Drafting the article: Marklund, Abu Hamdeh. Critically revising the article: Marklund, Virhammar, Alafuzoff. Reviewed submitted version of manuscript: Marklund. Approved the final version of the manu-script on behalf of all authors: Marklund. Statistical analysis: Abu Hamdeh. Study supervision: Marklund.

CorrespondenceNiklas Marklund: Uppsala University Hospital, Uppsala, Sweden. [email protected].


brain tissue aβ42 levels are linked to shunt response in ...the number of steps and seconds needed...

Documents