Table 9.
Neuroimaging.
| Reference | Population | Treatment | Diagnosis | Response criteria | Follow-up | Predictor | Risk of bias |
|---|---|---|---|---|---|---|---|
| Cheng et al. (2015) | 353 AD patients | Donepezil, galantamine, rivastigmine | Clinical (probable AD, NINCDS-ADRDA) | Longer TMSE decline-free survival (Survival analysis, using TMSE decline of 3+ points as endpoint) | Survival analysis, median 46.6 months | Lower total MTA score and older age | Low |
| Connelly et al. (2005a,b) | 140 AD patients (160 enrolled) | Donepezil, galantamine, rivastigmine | Clinical (probable AD, NINCDS-ADRDA) | At least two of gain of 2+ points on MMSE, improvement of the combined score IALD and social behavior subscales of NOSGER (or maintenance of maximum score), positive global change as defined as a tripartite agreement amongst doctor, subject, and carer (global impression) at 6 months | 3 years (actually 6 months) | Less medial temporal lobe atrophy at baseline | Moderate |
| Um et al. (2017) | 64 AD patients | Donepezil | Clinical (probable AD, NINCDS-ADRDA) | 2+ points improvement in MMSE | 24 weeks | Higher left hippocampal total volume and left CA1 volume (no cut-offs provided) | Moderate |
| Csernansky et al. (2005) | 39 AD patients (37 completed the study) | Donepezil | Clinical (NINCDS-ADRDA)—doesn't state whether it's probable or possible | Rate of change in ADAS-Cog (primary outcome); CDR, MMSE, NPI (secondary outcome)—Fiala et al. (1998) and Krstic et al. (2012) | 2 years | Smaller hippocampal volumes associated with deformation of the lateral and inferomedial surface (inward deformation of the hippocampal surface in proximity to the CA1 subfield and the subiculum) are negative predictor | High |
| Graff-Radford et al. (2012) | 54 DLB patients | Donepezil, galantamine, rivastigmine | Clinical (probable DLB, McKeith et al., 2004) | Reliable change in Dementia Rating Scale: reliable improvement with 9+ points increase in <15 months and 10+ >15 months, reliable decline 6+ decrease in <15 months, and 7+ decrease >15 months, stability in between | 1 year | Larger hippocampal volume | Low |
| Teipel et al. (2016) | 215 suspected prodromal AD patients | Donepezil | Clinical (progressive hippocampal amnestic syndrome: free recall ≤ 17 or total recall <40 on FCSRT) | Change in ADAS-Cog-MCI, CDR, TMT-A/B, CVLT | 12 months + 6 months OLE | Higher hippocampal volume | Some concerns—Low |
| Visser et al. (2005) | 121 AD patients | Rivastigmine | Clinical (probable AD, NINCDS-ADRDA, and DSM-IV) | 2+ points improvement on MMSE or 4+ points improvement on ADAS-Cog | 26 weeks | Absence of medial temporal atrophy (Giacobini et al., 2022) in patients receiving low dose of rivastigmine; medial temporal lobe atrophy in general is not a predictor | Low |
| Salib Tony Sheridan Mark Allington (2001) | 59 AD patients | Donepezil 5 mg | Clinical (NINCDS-ADRDA, not stated if possible included) | Unclear: “improvement or cognitive stability,” but non-response is defined as deterioration with MMSE <10 during the study period—probably means that non-responders are those who had to discontinue treatment | 2 years | Lower bicaudate span and bicaudate ratio | High |