A total of 102 MCI (34 naMCI and age-, sex-, and education matched 68 aMCI) patients were recruited from the pool of older adults who visited the Dementia and Age-Associated Cognitive Decline Clinic of the Seoul National University Hospital (SNUH) to get outpatient clinical services. All participants met the current consensus criteria for MCI: 1) cognitive impairment confirmed by an informant, 2) objective cognitive impairment, 3) preserved global cognitive function, 4) independence in functional activities, and 5) not demented [13,14]. aMCI and naMCI subjects met different criterion 2). For aMCI, the age-, sex-, and education-adjusted z-score for at least one of the four episodic memory tests (i.e., Word List Memory, Word List Recall, Word List Recognition, and Constructional Recall tests) in the Korean version of the Consortium to Establish a Registry for Alzheimer’s disease (CERAD-K) neuropsychological battery [15,16] was below -1.5. In contrast, for naMCI, the age-, sex-, and education-adjusted z-score for all the four episodic memory tests were identical to or above -1.5 and at least one of four non-memory tests (i.e., Boston Naming, Visuospatial Construction, and Sematic Fluency in the CERAD-K neuropsychological battery, and Stroop Color Word test) was below -1.5. All participants had an overall clinical dementia rating (CDR) [17] of 0.5. Exclusion criteria for all subjects were any serious medical or neurological diseases present that could affect mental function, evidence of focal brain lesions on magnetic resonance imaging, the presence of severe behavioral or communication problems that would make a clinical examination difficult, and the absence of a reliable informant. Individuals with minor physical abnormalities (e.g., diabetes with no serious complications, essential hypertension, and mild hearing loss) were included. This study was approved by the Institutional Review Boards of the SNUH (IRB No. 1905-110-1035).

All participants underwent comprehensive clinical and neuropsychological evaluation based on the CERAD-K assessment packet [15,16]. Clinical diagnosis and CDR rating was made through consensus case conference in which more than four psychiatrists with expertise for dementia research participated. The presence or absence of six cerebrovascular risk factors, including stroke, diabetes, dyslipidemia, transient ischemic attack, hypertension, and coronary artery disease, was systematically assessed from the subjects and subjects’ histories provided by the informant as well as a review of pertinent medical records. To calculate an overall measure of cerebrovascular burden, we created a composite score, i.e., the vascular risk score (VRS), which was the sum of the factors present, ranging from 0 to 6 and reported as a percentage [18]. Apolipoprotein E (APOE) genotyping was implemented for subset of participants (n=81).

All participant received [¹⁸F] florbetaben positron emission tomography (PET) scan. PET scans were performed using the ECAT EXACT47scanner (Siemens-CTI; Knoxville, TN, USA). The participants were classified as either Aβ positive or negative according to the Brain Amyloid Plaque Load (BAPL) score [19]. The BAPL is a three-grade scoring system, which is determined by visual assessment of the degree of Aβ accumulation. If the BAPL score is 2 or 3 points in at least one of the four regions of Interest (ROIs), i.e., the lateral temporal lobes, frontal lobes, posterior cingulate/precuneus and parietal lobes, the patient is classified to Aβ positive. A trained psychiatric resident (SHL) rated the BAPL score for PET scan of every participant. In order to verify the reliability of the rating, the agreement between ratings of the rater (SHL) and clinical ratings provided by nuclear medicine physicians at the SNUH was evaluated by calculating kappa value. More than substantial agreement was observed: Kappa value was 0.729 for aMCI, 1.000 for naMCI, and 0.822 for overall MCI cases. For Aβ positive participants, Aβ accumulation pattern was also determined by visual inspection by the same rater (SHL). The accumulation pattern was classified into A-pattern and B-pattern according to the guideline described in the previous report [11]. A-pattern has predominant amyloid retention in the frontal, anterior cingulate, lateral temporal, and basal ganglia, while B-pattern shows generalized amyloid retention over the entire cerebral cortex (Figure 1).

All statistical analyses were conducted using SPSS (version 24, IBM Corp., Armonk, NY, USA). For the comparison between aMCI and naMCI, Student’s t-test or chi-square test was used depending on the type of variables. The threshold for statistical significance was p<0.05 for all the analyses.

Table 1 shows the demographic and clinical characteristics of the participants by MCI subtype. Since the participants of aMCI and naMCI were matched by age, sex and years of education, there was no significant difference in the demographic factors. The two MCI subtypes also did not differ in regard of age of onset, duration of illness, APOE ε4 carrier state, and VRS. The mean CDR sum of box (SOB) score of aMCI was significantly higher than that of naMCI. As expected, z-scores of four episodic memory tests in the CERAD-K neuropsychological batter were significantly lower in aMCI than in naMCI.

The Aβ positivity rate of aMCI and naMCI was 64.7% and 26.5%, respectively (Table 2, Figure 2). The rate of naMCI was significantly lower than that of aMCI.

There was no significant difference in the proportion of amyloid accumulation patterns between Aβ positive aMCI and Aβ positive naMCI (Table 3, Figure 2).