Skip to main content

Table 3 Regression modelling of biomarkers predicting MMSE scores over 26 pre- and postoperatively and gait velocity improvement over 20% postoperatively

From: Cerebrospinal fluid biomarkers that reflect clinical symptoms in idiopathic normal pressure hydrocephalus patients

MMSE ≥ 26 preoperative

Univariate

Odds

p-value

OR

C.I. (OR)

Predicted

R-squared

Model p-value

P-tau < 27.5

1.23

.001*

3.41

1.62–7.20

   

T-tau < 211

1.31

.001*

3.69

1.66–8.24

   

NfL < 1317

0.71

.069

2.02

0.95–4.33

   

Aβ38 < 1692

0.74

.068

2.10

0.95–4.64

   

Aβ40 < 4461

0.67

.111

1.95

0.86–4.43

   

Aβ42 < 404

0.41

.270

1.51

0.73–3.13

   

sAPPα  < 241

0.59

.238

1.80

0.68–4.78

   

sAPPβ  < 317

− 0.37

.308

0.69

0.34–1.40

   

MCP1 < 493

− 0.39

.281

0.68

0.33–1.38

   

Multivariate: P-tau, NfL, Age, Sex, APOE

 

69.8

.183

.011*

 P-tau < 27.5

1.22

.011*

3.38

1.32–8.70

   

 NfL < 1317

0.73

.165

2.07

0.74–5.78

   

 APOE ε4 absent

0.26

.608

1.29

0.48–3.47

   

 Age

− 0.03

.348

0.97

0.92–1.04

   

 Sex (male)

0.15

.741

1.16

0.47–2.87

   

MMSE ≥ 26 postoperative

P-tau < 27.5

1.50

.000*

4.49

2.17–9.31

   

T-tau < 213.5

2.09

.000*

8.05

3.58–18.11

   

NfL < 1045

1.31

.000*

3.69

1.82–7.46

   

Aβ38 < 1710

0.92

.016*

2.51

1.19–5.30

   

Aβ40 < 3714

0.78

.024*

2.18

1.11–4.28

   

Aβ42 < 407

0.30

.397

1.35

0.68–2.68

   

sAPPα  < 435

0.37

.288

1.44

0.74–2.82

   

sAPPβ  < 260

0.45

.194

1.56

0.80–3.07

   

MCP1 < 515

0.42

.249

1.52

0.75–3.12

   

Multivariate: T-tau, NfL, Aβ38, Age, Sex

 

72.9

.321

.000*

 T-tau < 213.5

2.61

.000*

13.61

3.37–54.99

   

 NfL < 1045

0.88

.049*

2.40

1.01–5.75

   

 Aβ38 < 1710

-0.94

.172

0.39

0.10–1.50

   

 Age

0.03

.424

1.03

0.96–1.10

   

 Sex (male)

0.21

.623

1.23

0.54–2.82

   

Gait improvement > 20%

P-tau < 27.5

0.65

.118

1.91

0.85–4.27

   

T-tau < 206

1.01

.016*

2.75

1.20–6.27

   

NfL < 1050

1.00

.017*

2.73

1.20–6.22

   

Aβ38 < 1255

− 0.64

.116

0.53

0.24–1.17

   

Aβ40 < 3188

− 0.36

.382

0.70

0.31–1.57

   

Aβ42 < 301

− 0.86

.034*

0.43

0.19–0.94

   

sAPPα < 271

− 0.80

.099

0.45

0.17–1.16

   

sAPPβ < 239

− 0.94

.024*

0.39

0.17–0.89

   

MCP1 < 504

0.48

.242

1.61

0.73–3.56

   

Multivariate: T-tau, sAPPβ, Age, Sex, APOE

    

75.8

.221

.008*

 T-tau < 206

1.14

.043*

3.13

1.04–9.43

   

 APOE ε4 absent

1.35

.008*

3.85

1.43–10.36

   

 sAPPβ < 239

− 1.01

.076

0.37

0.12–1.11

   

 Age

0.02

.654

1.02

0.95–1.09

   

 Sex (male)

0.25

.624

1.28

0.47–3.48

   
  1. Significant results are highlighted with asterisk (*). Biomarkers were transformed to dichotomous variables with ROC derived cut-offs for biomarker concentration in CSF (used cut-off presented as ng/l and after each biomarker). Odds and odds ratios are calculated for low concentration group predicting the MMSE score over 26 and gait velocity improvement over 20%. The model prediction accuracy is presented in predicted column. Explained variance is evaluated by Nagelkerke R2 and presented in R-squared column. Age and sex were used as covariates for analysis. The multivariate models are the best combination of variables to explain the variance and predict the MMSE performances and gait improvement. APOE dichotomized by presence of allele ε4. Results are presented for patients with no APOE allele ε4. OR: odds ratio; MMSE: mini mental state examination; Gait: gait velocity in m/s; T-tau: total tau protein; P-tau: phosphorylated at threonine 181 tau protein; NfL: neurofilament-light; Aβ38: Amyloid-β 38; Aβ40: Amyloid-β 40; Aβ42: Amyloid-β 42; sAPPα: soluble amyloid precursor protein α; sAPPβ: soluble amyloid precursor protein β; MCP1: monocyte chemoattractant protein 1; APOE ε4: apolipoprotein E allele ε4
Back to article page

Fluids and Barriers of the CNS

ISSN: 2045-8118