Table 1.
Study | n | Subjects | Age | Male (%) | Education (years) | Design | Nuisance | Corrected for | Outcome | Effect |
---|---|---|---|---|---|---|---|---|---|---|
Quantitative assessment | ||||||||||
Mori, 1997* [41] | 60 | Mild to moderate AD | 70.2 (7.1) | 38.3 | 8.9 (2.3) | Cross-sectional | Age, sex, education | Atrophy | ADAS-Coga | −0.12 |
WAIS-R Full IQe | 0.40 | |||||||||
Staff, 2004 [58] | 98 | Unknown | 79 | 57.6 | 9.8 (1.6) | Cross-sectional | Childhood IQ, sex | WMH | AVLT Memoryb | 0.00 |
RPMe | 0.01 | |||||||||
Christensen, 2009 [15] | 416 | Unknown | 62.6 (1.4) | 52 | 14 (2.6) | Longitudinal, 4-year change | Age, sex, education | Atrophy and WMH | SDMTc | −0.22 |
CVLT: Immediateb | −0.39 | |||||||||
CVLT: Delayedb | −1.45 † | |||||||||
Farias, 2012 [18] | 401 | Mixed HC, MCI and dementia | 75 (6.9) | 37.3 | 12.3 (4.8) | Cross-sectional | Sex, height | TBV, hippocampal volume, and WMH | SENAS: Semantic memoryb | 0.20 |
Episodic memoryb | 0.08 | |||||||||
Executive functionc | 0.27 | |||||||||
Spatial abilityd | 0.16 | |||||||||
Royle, 2013 [50] | 327 | HC | 72.5 (0.7) | 100 | – | Cross-sectional | Age | TBV | Composite scorea | 0.19 |
GM and WM | 0.22 | |||||||||
297 | 72.6 (0.73) | 0 | – | TBV | 0.21 | |||||
GM and WM | 0.23 | |||||||||
Groot, 2018 [22] | 201 | Aβ+ preclinical and prodromal AD | 66.6 (7.5) | 53 | 10–11** | Cross-sectional | Age, sex, education, scanner | Atrophy | Memoryb | 0.12 |
Attentionc | 0.06 | |||||||||
Executivec | 0.18 | |||||||||
Languagef | −0.03 | |||||||||
Visuospatiald | 0.14 | |||||||||
MMSEa | 0.16 | |||||||||
462 | Aβ+ probable AD | 66.1 (7.4) | 47 | 10–11** | Memoryb | 0.10 | ||||
Attentionc | 0.14 | |||||||||
Executivec | 0.15 | |||||||||
Languagef | 0.05 | |||||||||
Visuospatiald | 0.13 | |||||||||
MMSEa | 0.15 | |||||||||
Categorical assessment | ||||||||||
Wolf, 2004 [69] | 73 | HC, MCI | 79.1 | 49.3 | 11.3 | Cross-sectional | Education | Left hippocampus | Predicting HC vs MCI (OR) | 1.04*** |
70 | MCI, dementia | 78.7 | 34.3 | 10.8 | Age | RBV | Predicting MCI vs dementia (OR) | 1.05*** | ||
Wolf, 2004 [70] | 167 | HC, MCI, AD, VaD | 60.7 (9.9) | 43 | – | Cross-sectional | Age, sex | Hippocampal atrophy (visual assessment) | HC vs cognitive impairment (OR compared with smallest quartile) | 2.9 |
Silbert, 2009 [56] | 49 | HC (at baseline) | 84.1 (6.2) | 47 | 14.5 (2.7) | Longitudinal, 10-year change | Age, MMSE, APOEe4 status. | ΔWMH, ΔvCSF, and hippocampal, vCSF and WMH | Persistent cognitive decline (HR) | 1.0 |
Negash, 2013 [44] | 54 | Aβ+ HC and AD | 72.7 | 42.6 | 14.4 | Cross-sectional | Age, sex, education, APOEe4 status | MTL volume | Resilience (normal despite Aβ+; OR) | 1.01 |
Hippocampal and posterior cingulate volume | 1.01 |
Bold effects are reported to be significant according to study-specific statistical thresholds
Aβ amyloid-beta, AD Alzheimer’s disease, ADAS-Cog Amsterdam dementia assessment scale—cognitive subscale, APOEε4 apolipoprotein ε4, AVLT auditory verbal learning test, CVLT California verbal learning test, GM gray matter, HC healthy controls, HR hazard ratio, ICV intracranial volume, IQ intelligence quotient, MCI mild cognitive impairment, MMSE Mini-Mental State Examination, MTL medial temporal lobe, OR odds ratio, RBV relative brain volume (brain volume to ICV ratio), RPM Raven’s progressive matrices, SDMT symbol-digit modalities test, SENAS Spanish-English neuropsychological assessment scale, TBV total brain volume, VaD vascular dementia, vCSF ventricular cerebrospinal fluid, WAIS-R Wechsler adult intelligence scale-revised, WM white matter, WMH white matter hyperintensity
*Premorbid brain volume calculated as regression coefficient of (age + sex + ICV = whole brain volume) multiplied by ICV + constant
**Categorization according to the Verhage scale [65] converted into years
***Odds ratios calculated from β coefficients using e^(β)
†This effect was considered an outlier and was not included in the meta-analysis
aGlobal cognition
bMemory
cAttention/executive functions
dVisuospatial ability
eIntelligence
fLanguage