Clinical parameters predicted the progression to dementia in oldest old patients with mild cognitive impairment (MCI).

Background: This study intends to assess to what extent instruments commonly used in clinical practice, as well as plasma p-tau-181, can predict the progression from MCI to dementia. The usefulness of a disease progression model (DPM) is also explored.

Methods: A longitudinal, prospective nested case-control study was conducted with patients from the Geriatrics outpatient clinics who met the MCI International Working Group criteria. The patients had a first clinical interview and two follow-ups after 12 and 24 months. Validated Spanish instruments were used for assessment, including the Mini-Mental State Examination (MMSE), the clock test, verbal fluency, the EURO-D depression scale, Barthel's Index, and Lawton's Index. P-tau-181 analysis was performed with SIMOA (Single MOlecule Array). A robust parametric disease progression model (RPDPM) was developed.

Results: Fifty-nine patients fulfilled the inclusion criteria. The median age was 82.7 + /-8.7 years, 93 % had amnestic MCI and 45.8 % progressed to dementia (ICD-11 criteria) in two years. P-tau-181 was not prognostic. An RPDPM with the MMSE, clock test, and Lawton's Index could predict progression to dementia with an AUC of 0.945.

Conclusion: A combination of the MMSE, clock test, and Lawton's Index in a DPM model predicted progression from MCI to dementia best. P-tau and other blood biomarkers did not predict progression. Our results highlight the strength of clinical variables to predict the progression of MCI.