This is a single-center, prospective, single-arm cohort study designed to estimate the incidence of postoperative cognitive decline (POCD) in patients aged 60 years and older with preexisting cognitive impairment undergoing elective non-cardiac surgery under general anesthesia.
To calculate the incidence of POCD, a normative population of cognitively normal individuals will be recruited to provide reference data for the expected range of cognitive change over time in the absence of surgery. This normative sample is not a control group for intervention comparison but is used to establish population-based norms for cognitive test score changes.
Study Procedures:
Preoperative assessment (1 day before surgery):
Patients will be screened using the Mini-Cog test. Those with scores ≤1 will be further evaluated with the Mini-Mental State Examination (MMSE) and Beijing version of the Montreal Cognitive Assessment (MoCA-B).（This threshold is based on the Chinese expert consensus scoring criteria, which differ slightly from the international standard. According to the Chinese algorithm, a score of ≤1 encompasses cases where the clock drawing is incorrect and one to two words are recalled, or when no words are recalled at all, aligning functionally with a score of ≤2 in the international system. Therefore, our inclusion criterion of ≤1 under the Chinese system is equivalent in diagnostic implication to the widely accepted international cutoff of ≤2, which indicates high likelihood of cognitive impairment.) Eligible participants will be enrolled after informed consent is obtained. Neuropsychiatric symptoms will be assessed using the Neuropsychiatric Inventory-Questionnaire (NPI-Q). Dementia severity and functional ability will be evaluated using the Clinical Dementia Rating (CDR), Basic Activities of Daily Living (BADL), and Instrumental Activities of Daily Living (IADL) scales.
Intraoperative data collection:
Information regarding surgery and anesthesia will be obtained from the electronic medical records.
Postoperative follow-up (at 3 and 12 months):
All participants will undergo cognitive assessments using the MMSE and MoCA-B. Neuropsychiatric symptoms, dementia severity (if applicable), and functional abilities will also be reassessed. Follow-up will be scheduled to minimize participant burden, and home-based assessments will be arranged when necessary. Remote assessments will be documented if used.
Normative Population:
Cognitively normal individuals will be recruited from inpatients and outpatients of Peking University Shenzhen Hospital. They will undergo cognitive assessments at baseline, 3 months, and 12 months to provide normative data for the calculation of standardized cognitive changes.
Primary Outcome:
The primary outcome is the incidence of POCD at 3 months after surgery. Following the nomenclature recommendations by Evered et al., POCD is defined as objectively measurable cognitive decline occurring between 3 months and 7.5 years after surgery. The diagnosis of POCD in this study will be based on the Reliable Change Index (RCI), a standardized statistical method that accounts for normal variability and practice effects over time.
RCI is calculated as:
RCI = (ΔX\_patient - Δμ\_normative) / σ\_normative
Where ΔX\_patient is the individual's change score, Δμ\_normative is the mean change in the normative group, and σ\_normative is the standard deviation of change scores in the normative group.
A patient will be considered to have POCD if either:
At least two cognitive domains show an RCI ≤ -1.96, or
The combined Z-score of all cognitive tests is ≤ -1.96.
Follow-up Management and Missing Data Handling:
Participants will be followed at 3 and 12 months. If a participant is unable to attend, alternative arrangements will be made. Loss to follow-up will be documented with reasons, and missing data will be addressed using sensitivity analyses.
