Skip to main content
Download PDF

Dissecting the clinical and pathological prognosis of MCI patients who reverted to normal cognition: a longitudinal study

BMC Medicine volume 23, Article number: 260 (2025) Cite this article

Abstract

Background

Controversy existed in the prognosis of reversion from mild cognitive impairment (MCI) to normal cognition (NC). We aim to depict the prognostic characteristics of cognition, neuroimaging, and pathology biomarkers, as well as the risk of Alzheimer’s disease (AD) dementia for MCI reverters.

Methods

A total of 796 non-demented participants (mean age = 73.3 years, female = 54.4%, MCI = 55.7%), who were divided into MCI reverters (n = 109), stable MCI (n = 334), and stable NC (n = 353) based on 2-year diagnosis changes, were subsequently followed up for 6 years. Cox proportional hazard regression models were applied to assess the AD dementia hazard. Linear mixed-effect models were used to evaluate the differences in changing rates of cognitive scores, brain volumes, brain metabolism, and AD biomarkers among three groups.

Results

The 2-year MCI reversion rate was 18.17%. MCI reversion was associated with an 89.6% lower risk of AD dementia (HR = 0.104, 95% confidence interval = [0.033, 0.335], p < 0.001) than stable MCI. No significant difference in incident AD risk was found between MCI reverters and stable NC (p = 0.533). Compared to stable MCI, reverters exhibited slower decreases in cognitive performance (false discovery rate corrected p value [FDR-Q] < 0.050), brain volumes (FDR-Q < 0.050), brain metabolism (FDR-Q < 0.001), and levels of cerebrospinal fluid β-amyloid1–42 (FDR-Q = 0.008). The above-mentioned differences were not found between MCI reverters and stable NC (FDR-Q > 0.050).

Conclusions

Reversion from MCI to NC predicts a favorable prognosis of pathological, neurodegenerative, and cognitive trajectory.

Peer Review reports

What is already known on this topic

Reversion from MCI to NC was not uncommon. However, the clinical and pathophysiologic trajectory after reversion was largely unstudied.

What this study adds

This study revealed that MCI reverters, similar to stable NC, had lower AD risks, as well as better prognosis trajectories in cognitive performance, neurodegeneration, and amyloid pathology compared to stable MCI.

How this study might affect research, practice or policy

Our study highlighted the heterogeneity of MCI and indicated that MCI reversion might predict a better prognosis trajectory. Our findings may serve as a valuable reference for future research and affect clinical decision-making for MCI management plans and targeted interventions.

Background

Mild cognitive impairment (MCI) is considered an intermediate state between normal cognition (NC) and dementia [1]. It is characterized by objective impairments in one or more cognition domains [1, 2]. MCI prevalence is approximately 6.7 to 25.2% in older populations, and it increases with age [2]. Studies have reported that MCI had heterogeneous transitions which could be roughly categorized into progression to dementia, stable MCI, or reversion to NC [3,4,5]. Meta-analyses uncovered that 18 to 30% of MCI individuals would revert to NC after 1 year or more [6, 7]. MCI reverters had decreased dementia risks compared to non-reverters, while it was inconsistent whether the risks differed between reverters and NC [8,9,10,11].

Cross-sectional studies revealed that MCI reverters had specific characteristics linked to cognitive prognosis, despite not all findings being consistent. MCI reverters had higher levels of cerebrospinal fluid (CSF) β-amyloid1–42 (Aβ1–42), and lower levels of CSF total tau (t-tau) and phosphorylated tau (p-tau) than stable MCI [11,12,13]. Compared to stable MCI, larger hippocampus volume, higher [18F] fluorodeoxyglucose (FDG) metabolism, and lower Aβ burden were detected among reverters [12,13,14]. It was elucidated that Alzheimer’s disease (AD) biomarker profiles and imaging characteristics of reverters were more similar to NC than MCI [5, 14, 15]. However, the longitudinal changes in cognition and AD pathology for MCI reverters remain largely unstudied. To our best knowledge, only one study focused on changes in cortical thickness and positron emission tomography (PET) imaging of Aβ, tau, and FDG in MCI reverters [16]. It found that reverters had slower decreases in cortical thickness and glucose metabolism than non-reverters [16].

In this study, we aimed to explore the longitudinal clinicopathologic characteristics of MCI reverters compared to stable MCI and stable NC in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database.

Methods

Study design and participants

Data were obtained from ADNI (http://adni.loni.usc.edu/). It began in 2004 and has been continuously collecting data across various phases. We used data from ADNI-1, ADNI-GO, ADNI-2, and ADNI-3, with collection dates up to 2022. The ADNI cohort incorporates over 2200 participants with longitudinal follow-up, including individuals with NC, MCI, and early AD dementia. The participants’ information and samples were collected, involving brain magnetic resonance imaging (MRI), PET, CSF, clinical information, and neuropsychological assessments. The institutional review boards at each participating institution approved this study. All participants or authorized representatives wrote informed consent.

The MCI diagnosis was based on Petersen criteria: [17, 18] (1) Subjective memory complaint. (2) Objective cognitive impairment. It is evaluated by a composite neuropsychological score with relevant cognitive domains, including memory, executive function, and language function. It means the participants have impaired composite score (Memory Function [MEM], Executive Function [EXF], or Language Function [LAN] were described in detail later), defined as > 1 SD below the age, education, and gender-adjusted normative mean, within at least one cognitive domain [19]. The mean and standard deviation (SD) of composites score were calculated based on stable CN population at baseline. (3) Relative independent daily function, which is quantified as Functional Assessment Questionnaire scores < 9 [20]. (4) Non-demented. The dementia diagnosis referenced published criteria [18, 21, 22]. Subjects who did not meet the criteria for MCI or dementia were considered CN.

Based on the above-mentioned criteria, a total of 600 MCI and 442 CN subjects who were followed for a maximum of 8 years were included. The participants were divided into three groups based on the first 2-year cognitive diagnosis changes, including MCI reverters who reverted from MCI to NC, stable MCI, and stable NC whose cognition was continuously normal across 2 years. Afterward, the second year was redefined as the new baseline. Figure 1 delineates the process of the study.

Fig. 1
figure 1

Flowchart of study design. The research overview summarizes the selection and analysis process of the population. Abbreviations: ADNI, Alzheimer’s Disease Neuroimaging Initiative; FAQ, Functional assessment questionnaire; MCI, Mild cognitive impairment; NC, Normal cognition; AD, Alzheimer’s diseases; MRI, Magnetic resonance imaging; PET, Positron emission tomography; CSF, Cerebral spinal fluid

Full size image

Cognitive measurements

Overall cognitive status was evaluated by Alzheimer’s Disease Assessment Scale-Cognitive 13 (ADAS-cog 13). The composite score of 13 items ranged from 0 to 85, and participants with higher scores represented worse cognition. MEM, a composite memory score, was obtained from the longitudinal Rey Auditory Verbal Learning Test, ADAS-cog 13, Mini-Mental State Examination, and Logical Memory data [23]. EXF, a useful composite measure of executive function, was developed from WAIS-R Digit Symbol Substitution, Digit Span Backwards, Trails A and B, Category Fluency, and Clock Drawing [24]. LAN, a composite language score, was calculated by the Boston Naming Test, Animal and Vegetable fluency, Mini-Mental State Examination, ADAS-cog 13, and Montreal Cognitive Assessment [25]. The co-calibrated composites score of MEM, EXF, and LAN were calculated by modern psychometric approaches based on the neuropsychological battery mentioned above [23,24,25]. More details on the calculation and co-calibration of the composite scores could be found from published studies [23,24,25] and ADNI websites (https://ida.loni.usc.edu/explore/jsp/search/search.jsp?project=ADNI#studyFiles). Including multiple indicators in a single score minimized the impact of measurement error due to idiosyncratic single items or subdomains [23]. Therefore, MEM, EXF, and LAN could reliably represent the memory, executive, and language functions, respectively.

MRI for brain volume

The procedure of MRI acquisition in ADNI has been described previously [26]. The MRI protocol for ADNI1 focused on consistent longitudinal structural imaging on 1.5T scanners. In ADNI-GO/ADNI2/ADNI3, imaging was performed at 3T scanners. Scanners from the three largest MRI vendors (GE, Philips, and Siemens) are supported at the ADNI imaging center. The range of scanners being qualified included GE 750/750W, Siemens Prisma/ Prisma Fit/Skyra/Verio/Trio/TIM, and Philips Achieva/Ingenia 3 T CX (https://adni.loni.usc.edu/help-faqs/adni-documentation/). FreeSurfer were used to ensure the most accurate and reliable morphometric measurements (https://surfer.nmr.mgh.harvard.edu/). In this study, the hippocampus, middle temporal cortex, and entorhinal cortex were considered as regions of interest.

PET for amyloid deposition and brain metabolism

Brain Aβ burden was estimated using [18F] florbetapir PET. Summary florbetapir standard uptake value ratios were acquired by averaging uptake ratios across cortical regions (frontal, anterior/posterior cingulate, lateral parietal, lateral temporal regions) normalized by the reference region (whole cerebellum) [27]. FDG PET was applied to identify important hypometabolic regions (metaROIs) associated with AD pathological metabolic change. The pre-defined regions of interest were developed based on literature review including angular, temporal, and posterior cingulate regions. The provided metaROIs mean was normalized by dividing the mean of the top 50% of pons/vermis reference region [14, 28]. For further details, please refer to “ADNI_UCBerkeley_AmyloidPET_methods_April2023.pdf” and “ADNI_UC_Berkeley_FDG_Methods_20220323.pdf” on the ADNI website.

CSF measurements of AD core biomarkers

CSF standard procedures in ADNI were described [29]. CSF samples were randomly distributed in plates and then were measured in duplicate. The antibodies and plates were from the same lot to avoid variability between batches. The mean intraplate variation coefficient was 2.2%; all duplicate measurements had variation coefficient below 15% [29, 30]. CSF samples were measured using the Elecsys β-amyloid (1–42) CSF, and the Elecsys phosphotau (181P) CSF and Elecsys total-tau CSF immunoassays on a cobase 601 analyzer at the Biomarker Research Laboratory, University of Pennsylvania, USA, according to the preliminary kit manufacturer’s instructions and as described in previous studies [31, 32].

Covariate measurements

All participants had baseline data on age, sex, and years of education. Apolipoprotein E (APOE4) status was determined according to rs7412 and rs429358, which were genotyped separately by an APOE genotyping kit to define the APOE ε2/ε3/ε4 isoforms. Comorbidities of hypertension, diabetes, coronary heart disease, stroke, depression, anxiety, obstructive sleep apnea syndrome, and insomnia were collected from self-reporting to whether the subjects had been diagnosed or treated for these diseases.

Statistical analysis

Population baseline characteristics were compared between MCI reverters and stable MCI or stable NC using Student’s t test or Mann–Whitney U test (for continuous variables) and the χ2 test or Fisher exact test (for categorical variables).

Cox proportional hazard regression models were applied to assess the associations of different MCI transition status with incident AD dementia risk. Participants were censored at their last follow-up visit. Hazard ratios (HR) and 95% confidence intervals (CI) were reported. Kaplan–Meier method was used to measure the cumulative AD incidence for each group and log-rank test was used to calculate the differences. Schoenfeld’s residuals were used to test the proportional hazards assumption. Stratified analyses were conducted by APOE4 status and sex. Next, linear mixed-effects model (LMEM) with random intercepts and slopes, termed time × group interaction, was applied to explore the differences in longitudinal change rates of cognition, imaging degeneration, and CSF biomarkers in different groups. Additional analyses with time × group × APOE4 status interaction and time × group × sex interaction were incorporated in LMEM to assess whether the MCI transition interacted with APOE4 status or sex in longitudinal changes of relevant markers. The overall significance of three-way interaction terms was evaluated by likelihood ratio test comparing the full model and nested model that did not incorporate three-way interaction terms. LMEM diagnostics were conducted to meet necessary assumptions: model residuals and random effects distributed normally and did not exhibit heteroscedasticity. Therefore, the data of CSF Aβ1–42, t-tau, p-tau, Amyloid PET, and FDG PET were normalized by z-score, and the outliers situated outside ± 3 SD were excluded. All models were adjusted for age, sex, APOE4 status, and years of education. The LMEM further adjusted the cognitive performance, imaging degeneration, and the level of CSF biomarkers at the new baseline respectively. Intracranial volume was adjusted in analyses involving cortical volume. We additionally adjusted the comorbidities in sensitivity analysis.

A two-tailed p < 0.05 was considered statistically significant, and false discovery rate (FDR) correction was used when needed. All analyses were performed using R version 4.3.1 for Windows with “car”, “survival”, “survminer”, “magrittr”, “nlme”, “arm”, “lme4”, “lmerTest”, “aod”, “ggplot2”, “ggsci”, and “ggpubr” packages.

Results

Characteristics of participants at new baseline

After a 2-year follow-up of 1042 non-demented adults (including 442 NC and 600 MCI), 89 (20.14%) NC, and 157 (26.17%) MCI participants were excluded due to progression. Other 353 (79.86%) NC participants remained cognitively normal, 109 (18.17%) MCI participants reverted to NC, and 334 (55.67%) remained MCI (Fig. 1). Of these, 673 (84.55%) individuals with continuous follow-up were further included for subsequent analyses. The characteristics for three groups at year 2 (called “new baseline”) were given in Table 1.

Table 1 Population characteristics at new baseline
Full size table

Compared to stable MCI, MCI reverters were more likely to be young, APOE 4 non-carrier, and well-educated (p < 0.050). They had better performance in general cognition, memory, and executive domains (p < 0.001). They had larger hippocampus and entorhinal volume (p < 0.001), lower brain Aβ burden by PET (p = 0.003), and higher levels of brain metabolism (p < 0.001). However, the differences in CSF biomarker profiles of Aβ1–42 and tau proteins were non-significant between reverters and stable MCI (p > 0.050). The Aβ positivity rates were not significantly different between the MCI reverters (43.3%) and the stable MCI (52.8%) (p = 0.071) (Table 1). Compared to stable NC, MCI reverters were older, tended to be female, and had fewer schooling years (p < 0.050). They had worse performance in general cognition, memory, and executive domains (p < 0.001) than stable NC. No significant difference was detected in brain volume, PET burden of amyloid, brain metabolism, CSF AD biomarkers, and the proportion of participants who were Aβ positive between MCI reverters and stable NC (p > 0.050) (Table 1).

The population characteristics for three groups at original baseline were displayed in Additional file: Table 1. Different from the characteristics at new baseline, MCI reverters were less likely to be Aβ positive than stable MCI at original baseline (p = 0.001) (Additional file: Table 1). Moreover, compared to cognitive scores at original baseline, MCI reverters’ ADAS13-cog scores decreased by an average of 1.5 points, MEM scores increased by an average of 0.22 points, and EXF scores increased by an average of 0.32 points at new baseline (Table 1 & Additional file: Table 1).

Risk of AD dementia or cognitive decline

Compared to stable MCI, stable NC had a 90.9% decreased risk of AD (HR = 0.091, 95% CI = [0.042, 0.198], p < 0.001) (Fig. 2 and Table 2). Consistent findings were found between MCI reverters and stable MCI. Kaplan-Meier analyses showed that reverters had lower cumulative incidence rates of AD than stable MCI (log-rank p < 0.001) (Fig. 2). MCI rverters had an 89.6% decreased AD hazard (HR = 0.104, 95% CI = [0.033, 0.335], p < 0.001) compared with stable MCI (Table 2). Interestingly, there were no significant differences in AD risks between MCI reverters and stable NC (p > 0.050) (Fig. 2 and Table 2). Nonetheless, MCI reverters had a higher risk of cognitive decline (progressed to MCI or AD) than stable NC (log-rank p < 0.001) (Additional file: Fig. S1), and the HR was 3.355 (95% CI = [2.396, 4.696], p < 0.001) (Additional file: Table 2). The results in subgroup analyses were similar to the main results and no stratified effects were detected by sex or APOE4 status (Table 2 & Additional file: Table 2). In sensitivity analysis adjusting comorbidities, the results were consistent to the primary analysis (Additional file: Table 3).

Fig. 2
figure 2

Kaplan-Meier plot of the cumulative probability of AD risk in MCI Reverters, Stable MCI, and Stable NC group. Abbreviations: MCI, Mild cognitive impairment; NC, Normal cognition

Full size image
Table 2 Cox proportional hazards regression model to estimate the risk of AD at the last visit
Full size table

Longitudinal changes of cognition, brain imaging, and AD biomarkers

Longitudinal changes of cognitive scores

Compared with stable MCI, stable NC had a slower increase in scores of ADAS-cog 13 (β = -0.636, SE = 0.079, FDR-Q < 0.001) and a slower decrease in scores of MEM (β = 0.026, SE = 0.006, FDR-Q < 0.001) and EXF (β = 0.019, SE = 0.007, FDR-Q = 0.016) (Additional file: Table 4). MCI reverters also had a slower increase in ADAS-cog 13 scores (β = -0.443, SE = 0.116, FDR-Q = < 0.001) and a slower decrease in MEM scores (β = 0.025, SE = 0.008, FDR-Q = 0.003) and EXF scores (β = 0.035, SE = 0.011, FDR-Q = 0.002) than stable MCI (Fig. 3A–C and Additional file: Table 4). Compared with stable NC, reverters had faster increase in scores of ADAS-cog 13 (β = 0.281, SE = 0.089, FDR-Q = 0.024). There were no significant differences in longitudinal change rates of MEM and EXF scores between MCI reverters and stable NC (FDR-Q > 0.050) (Additional file: Table 4). The results on verbal memory (assessed by Rey Auditory Verbal Learning Test immediate and forgetting) and specific executive function (assessed by Trails A and B) were consistent with the ones obtained using the MEM and EXF composites. Detailed descriptions are displayed in Additional file: Supplementary Results & Table 4.

Fig. 3
figure 3

A–H MRI, PET, CSF, and neuropsychological measures trajectories estimated using linear mixed-effects models in the MCI Reverters and Stable MCI group. Abbreviations: MCI, Mild cognitive impairment; MRI, Magnetic resonance imaging; PET, Positron emission tomography; CSF, Cerebral spinal fluid; SE, Standard error; ADAS-cog, Alzheimer’s Disease Assessment Scale-Cognitive 13; MEM, Memory function; EXF, Executive function; FDG, Fluorodeoxyglucose; Aβ1-42, β-amyloid 1-42; t-tau, total tau; p-tau, phosphorylation tau; FDR-Q, False discovery rate corrected p value

Full size image

Longitudinal changes of brain atrophy

Compared with stable MCI, stable NC had a slower reduction in hippocampus (β = 0.092, SE = 0.011, FDR-Q < 0.001), entorhinal (β = 0.052, SE = 0.018, FDR-Q = 0.007), and mid-temporal volume (β = 0.119, SE = 0.044, FDR-Q = 0.012) (Additional file: Table 5). Similarly, MCI reverters had slower atrophy in hippocampus (β = 0.085, SE = 0.018, FDR-Q < 0.001), entorhinal cortex (β = 0.108, SE = 0.027, FDR-Q < 0.001), and mid-temporal cortex (β = 0.143, SE = 0.058, FDR-Q = 0.020) than stable MCI (Fig. 3D–F and Additional file: Table 5). There were no significant differences in longitudinal change rates of brain volume between MCI reverters and stable NC (FDR-Q > 0.050) (Additional file: Table 5).

Longitudinal changes of brain FDG metabolism and Aβ burden

Compared with stable MCI, stable NC had a slower decrease in brain FDG metabolism (β = 0.029, SE = 0.003, FDR-Q < 0.001) (Additional file: Table 6). MCI transition status was associated with longitudinal change rates FDG-PET (β = 0.027, SE = 0.005, FDR-Q < 0.001). MCI reverters had a slower decrease in brain metabolism than stable MCI (Fig. 3G and Additional file: Table 6). There were no significant differences in longitudinal change rates of FDG-PET between MCI reverters and stable NC (β = -0.005, SE = 0.006, FDR-Q = 0.765) (Additional file: Table 6). Significant differences in longitudinal change rates of brain Aβ burden were not found among MCI reverters, stable MCI, and stable NC (FDR-Q > 0.050) (Additional file: Table 6).

Longitudinal changes of CSF biomarkers

Stable MCI had a faster decrease in CSF Aβ1–42 compared with stable NC (β = 0.025, SE = 0.011, FDR-Q = 0.037). No differences were detected in longitudinal change rates of CSF t-tau and p-tau between stable MCI and stable NC (FDR > 0.050) (Additional file: Table 7). MCI transition status was not associated with longitudinal change rates of CSF biomarkers (FDR-Q > 0.050) except CSF Aβ1–42 (β = 0.049, SE = 0.017, FDR-Q = 0.008). Stable MCI had a faster decrease in CSF Aβ1–42 than MCI reverters (Fig. 3H and Additional file: Table 7). There were no significant differences in longitudinal change rates of CSF Aβ1–42, t-tau, and p-tau between MCI reverters and stable NC (FDR-Q > 0.050) (Additional file: Table 7).

The interaction effect of MCI transition with APOE4 status or sex in longitudinal changes of cognition, brain imaging, and AD biomarkers

The three-way interaction of time × group × APOE4 status accounted for a significant amount of variance in Trails A and B Time to Complete (FDR-Q < 0.05). Further analysis stratified by APOE4 status indicated that MCI transition status was only associated with longitudinal change rates of Trails A and B Time to Complete in APOE4 carriers (Additional file: Table 8). The three-way interaction of time × group × APOE4 status accounted for a significant amount of variance in entorhinal volume (FDR-Q = 0.003). Further analysis stratified by APOE4 status indicated that MCI transition status was only associated with longitudinal change rates of entorhinal volume in APOE4 carriers (p < 0.001) (Additional file: Table 8).

The likelihood ratio test indicated that the three-way interaction of time × group × sex accounted for a significant amount of variance in Trails A and B Time to Complete (FDR-Q < 0.05). Further analyses stratified by sex status indicated that MCI transition status was associated with longitudinal change rates of Trails A and B Time to Complete in women and men independently, but the differences were larger in men (Additional file: Table 9). The likelihood ratio test indicated that the three-way interaction of time × group × sex accounted for significant amounts of variance on FDG-PET (FDR-Q = 0.029). Further analyses stratified by sex status indicated that MCI transition status was associated with longitudinal change rates of FDG-PET in women (p < 0.001), men (p < 0.001) independently, but the differences were larger in men (Additional file: Table 9).

Longitudinal changes of cognition, brain imaging, and AD biomarkers in sensitivity analysis

After adjusting the comorbidities, the results were basically consistent with the primary analysis in longitudinal changes of cognition, brain imaging, and AD biomarkers. The detailed results are displayed in Additional file: Table 10–13.

Discussion

The present study focused on the clinicopathologic prognosis of MCI reverters. Our study verified that MCI reverters, similar to stable NC, had lower AD risks than stable MCI. MCI reverters had a better trajectory of cognitive performance, neurodegeneration, and amyloid pathology than stable MCI. The cognitive, neurodegenerative, and amyloid pathological trajectories were similar between MCI reverters and stable NC.

Reversion from MCI to NC is not uncommon. Our findings that nearly one-fifth of MCI participants would revert to NC was consistent with previous studies [3, 6, 9, 11]. Recent meta-analyses revealed reversion rates ranging from 4 to 59% among all incorporated studies [6, 33]. The reversion rates had relatively high heterogeneity that could relate to discrepancies in study settings (community or clinic), population size, follow-up duration, and diagnosis criteria [6, 33]. Additionally, some previous studies considered MCI reversion as a random event in MCI to dementia trajectories [10]. However, our study suggested that MCI reverters had specific cognitive trajectories and dementia risk different from stable MCI. MCI reverters and stable MCI represented distinct pathophysiological processes. Our findings were consistent with those reported by Li Z [16].

MCI reverters, similar to stable NC, represented slower cognition decline and predicted lower AD risks than stable MCI. Consistent with our findings, research based on 3 longitudinal cohorts found that reverters had 59 to 88% reduced dementia hazards than non-reverters [16]. Another study found that MCI without reversion had 2.5 times dementia risks compared to reverters [8]. Some studies also focused on the prognosis of reverters compared to those without MCI history. Consistent with a few previous research, our results described that reverters had higher risks of cognitive decline but not higher risks of developing AD than stable CN [9, 11]. That means the occurrence of MCI events still had negative effects on individuals’ long-term cognitive health. The MCI reversion events also had positive effects on MCI patients. It is important to explore the risk factors of MCI incidence and predictive factors of MCI reversion [34, 35]. Nevertheless, few studies also described that reverters had increased dementia risks than NC [8, 10]. These controversial findings may be related to population source, sample size, and duration of follow-up. Further studies are needed.

Better cognitive trajectories in MCI reverters may be ascribed to slower degeneration in brain structure and metabolism. Specific brain atrophy has been proven to be neurodegenerative biomarkers for AD [36]. It was reported that hippocampus, entorhinal, and temporal volume loss were correlated with higher risk for MCI progression and less probability for MCI reversion [37,38,39,40]. Brain FDG hypometabolism could also predict the conversion from MCI to AD [41]. Furthermore, longitudinal studies showed that the decline of brain metabolism was associated with concurrent cognitive deterioration in MCI and early AD [42, 43]. Our study verified and complemented the findings that reverters had slower brain atrophy and metabolism decreasing than non-reverters in Mayo Clinic Study of Aging [16].

Slower deposition of Aβ and tau protein in reverters might correlate with slowing cognitive decline and lower AD risks. As the major hallmarks of AD, higher burden of Aβ and tau accumulation levels in MCI could lead to higher AD dementia risks [12, 15, 44]. Although brain amyloid burden was lower in reverters at baseline in this study, we did not detect distinctions in brain Aβ accumulation rates between reverters and stable MCI. Nonetheless, we found that reverters and stable NC had a significantly slower CSF Aβ1–42 decrease than stable MCI. Previous research suggested that CSF Aβ1–42 became abnormal in the early stages of AD, before amyloid PET, which could explain our findings [45]. A study published in 2022 also did not find significant differences in the accumulation rates of Aβ and tau based on longitudinal data from less than 100 participants [16]. The literature about pathophysiologic trajectories after reversion is scarce [3, 33]. The longitudinal changes of neurodegeneration and amyloid pathology in reverters require further studies based on larger population and longer follow-up duration.

MCI reverters, independent from stable MCI, had similar neurodegenerative and amyloid pathological trajectories and AD risks to stable NC. This might induce a doubt that why participants who reverted were initially diagnosed with MCI. Aside from irreversible neurodegenerative diseases, malnutrition, physical frailty, mild depressive symptoms, and sleep disorders are common causes of MCI [6]. Under these circumstances, AD pathology may be normal and corrections of these medical conditions could benefit cognitive improvement [46]. Another potential reason was that the false positives for MCI diagnosis might still exist though the reversion rate in our study was consistent with the findings of meta-analyses. To some extent, these indicated that reverters might attain finite benefits from needless antidementia treatments while facing risks of probably detrimental consequences, such as discrimination, overmedication, and side effects [8]. In broader clinical perspectives, as new treatments for latent neurodegeneration are invented, it will be important to choose participants for clinical trials who are really in the early stages of dementia to avoid bias resulting from MCI reversion [6, 8]. Therefore, further high-quality studies are warranted for reliable prediction of MCI reversion and progression to assist clinical decisions.

Our study had several strengths. Firstly, the participants came from ADNI with a high proportion of follow-up, holistic neuropsychological evaluations, reliable CSF biomarkers examinations, and comprehensive imaging data. Secondly, this study clarified the characteristics of longitudinal changes in cognition, brain atrophy, brain Aβ burden and metabolism, and CSF biomarkers among MCI reverters. To our best knowledge, this was the first study to find that clinical and pathological trajectories of reverters were different from stable MCI but similar to stable NC in 6-year follow-up.

Our study had some limitations. Firstly, the observation of MCI reversion in previous studies was 1 to 5 years [11, 47]. Although participants in our study were observed for 2 years to determine whether reversion occurred, the reversion rate was consistent with the findings of meta-analyses [6, 33]. Secondly, the potential effects of other covariates and modifiable factors, especially the comorbidities, were not fully investigated. We did not consider and differentiate the etiology of MCI at baseline. Future studies are needed to understand the natural trajectory of different types of MCI, particularly MCI due to AD. Thirdly, the representation of ADNI is limited due to the participants being volunteers with relatively high education rather than the community population. The ADNI’s sample size is limited and it decreases with the extension of follow-up time, especially for longitudinal data of CSF biomarkers. It might limit the interpretability of results. Thus, further research in other populations and centers with larger sample sizes and longer follow-ups was needed to verify our results. Lastly, practice effects might manifest as improvements in cognitive tests due to repetitive evaluation [48, 49]. Nevertheless, our participants underwent the assessments at 6 to 12 month intervals while practice effect studies repeated cognitive tests after 1 week [50]. Practice effects are intricate phenomena worthy of further study.

Conclusions

Our study revealed the clinicopathologic prognosis trajectories of MCI reverters. MCI reverters, similar to stable NC, had a favorable prognosis of pathology, neurodegeneration, and cognition than stable MCI. This may help clinicians to better comprehend MCI reversion and provide valuable references for future studies.

Data availability

 All data are available upon reasonable request or can be obtained from the Alzheimer's Disease Neuroimaging Initiative database (adni.loni.usc.edu).

Abbreviations

Aβ1-42 :

β-Amyloid1-42

AD:

Alzheimer’s disease

ADNI:

Alzheimer’s Disease Neuroimaging Initiative

ADAS-cog 13:

Alzheimer’s Disease Assessment Scale-Cognitive 13

APOE:

Apolipoprotein E

CI:

Confidence intervals

CSF:

Cerebrospinal fluid

EXF:

Executive function

FDG:

Fluorodeoxyglucose

FDR:

False discovery rate

HR:

Hazard ratios

LAN:

Language function

LMEM:

Linear mixed-effects model

MCI:

Mild cognitive impairment

MEM:

Memory function

MRI:

Magnetic resonance imaging

NC:

Normal cognition

PET:

Positron emission tomography

p-tau:

Phosphorylated tau

SD:

Standard deviation 

SE:

 Standard error

t-tau:

Total tau

References

  1. Alzheimer’s disease facts and figures. Alzheimer’s Dementia. 2023;19(4):1598–695.

    Article  Google Scholar 

  2. Petersen RC, Lopez O, Armstrong MJ, Getchius TSD, Ganguli M, Gloss D, et al. Practice guideline update summary: Mild cognitive impairment: Report of the Guideline Development, Dissemination, and Implementation Subcommittee of the American Academy of Neurology. Neurology. 2018;90(3):126–35.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Qin Y, Han H, Li Y, Cui J, Jia H, Ge X, et al. Estimating Bidirectional Transitions and Identifying Predictors of Mild Cognitive Impairment. Neurology. 2023;100(3):e297–307.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Langa KM, Levine DA. The diagnosis and management of mild cognitive impairment: a clinical review. JAMA. 2014;312(23):2551–61.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Knopman DS. Alzheimer disease biomarkers and insights into mild cognitive impairment. Neurology. 2013;80(11):978–80.

    Article  PubMed  Google Scholar 

  6. Canevelli M, Grande G, Lacorte E, Quarchioni E, Cesari M, Mariani C, et al. Spontaneous Reversion of Mild Cognitive Impairment to Normal Cognition: A Systematic Review of Literature and Meta-Analysis. J Am Med Dir Assoc. 2016;17(10):943–8.

    Article  PubMed  Google Scholar 

  7. Wood H. Alzheimer disease: Meta-analysis finds high reversion rate from MCI to normal cognition. Nat Rev Neurol. 2016;12(4):189.

    Article  PubMed  Google Scholar 

  8. Aerts L, Heffernan M, Kochan NA, Crawford JD, Draper B, Trollor JN, et al. Effects of MCI subtype and reversion on progression to dementia in a community sample. Neurology. 2017;88(23):2225–32.

    Article  PubMed  Google Scholar 

  9. Koepsell TD, Monsell SE. Reversion from mild cognitive impairment to normal or near-normal cognition: risk factors and prognosis. Neurology. 2012;79(15):1591–8.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Roberts RO, Knopman DS, Mielke MM, Cha RH, Pankratz VS, Christianson TJ, et al. Higher risk of progression to dementia in mild cognitive impairment cases who revert to normal. Neurology. 2014;82(4):317–25.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Thomas KR, Edmonds EC, Eppig JS, Wong CG, Weigand AJ, Bangen KJ, et al. MCI-to-normal reversion using neuropsychological criteria in the Alzheimer’s Disease Neuroimaging Initiative. Alzheimer’s Dementia. 2019;15(10):1322–32.

    Article  PubMed  Google Scholar 

  12. Park MH, Han C. Is there an MCI reversion to cognitively normal? Analysis of Alzheimer’s disease biomarkers profiles. Int Psychogeriatr. 2015;27(3):429–37.

    Article  PubMed  Google Scholar 

  13. Moradi K, Faghani S, Abdolalizadeh A, Khomeijani-Farahani M, Ashraf-Ganjouei A. Biological Features of Reversion from Mild Cognitive Impairment to Normal Cognition: A Study of Cerebrospinal Fluid Markers and Brain Volume. Journal of Alzheimer’s disease reports. 2021;5(1):179–86.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Ashraf-Ganjouei A, Moradi K, Faghani S, Abdolalizadeh A, Khomeijani-Farahani M, Fatehi F, et al. The Transition of Mild Cognitive Impairment Over Time: An AV45- and FDG-PET Study of Reversion and Conversion Phenomena. Curr Alzheimer Res. 2021;18(9):721–31.

    Article  CAS  PubMed  Google Scholar 

  15. Vermunt L, van Paasen AJL, Teunissen CE, Scheltens P, Visser PJ, Tijms BM. Alzheimer disease biomarkers may aid in the prognosis of MCI cases initially reverted to normal. Neurology. 2019;92(23):e2699–705.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Li Z, Heckman MG, Kanekiyo T, Martens YA, Day GS, Vassilaki M, et al. Clinicopathologic Factors Associated With Reversion to Normal Cognition in Patients With Mild Cognitive Impairment. Neurology. 2022;98(20):e2036–45.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Petersen RC. Mild cognitive impairment as a diagnostic entity. J Intern Med. 2004;256(3):183–94.

    Article  CAS  PubMed  Google Scholar 

  18. Angevaare MJ, Vonk JMJ, Bertola L, Zahodne L, Watson CW, Boehme A, et al. Predictors of Incident Mild Cognitive Impairment and Its Course in a Diverse Community-Based Population. Neurology. 2022;98(1):e15–26.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Albert MS, DeKosky ST, Dickson D, Dubois B, Feldman HH, Fox NC, et al. The diagnosis of mild cognitive impairment due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer’s Dementia. 2011;7(3):270–9.

    Article  PubMed  Google Scholar 

  20. Bondi MW, Edmonds EC, Jak AJ, Clark LR, Delano-Wood L, McDonald CR, et al. Neuropsychological criteria for mild cognitive impairment improves diagnostic precision, biomarker associations, and progression rates. Journal of Alzheimer’s disease : JAD. 2014;42(1):275–89.

    Article  PubMed  Google Scholar 

  21. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology. 1984;34(7):939–44.

    Article  CAS  PubMed  Google Scholar 

  22. Association AP. Diagnostic and statistical manual of'mental disorders (4th ed.). 1994.

  23. Crane PK, Carle A, Gibbons LE, Insel P, Mackin RS, Gross A, et al. Development and assessment of a composite score for memory in the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Brain Imaging Behav. 2012;6(4):502–16.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Gibbons LE, Carle AC, Mackin RS, Harvey D, Mukherjee S, Insel P, et al. A composite score for executive functioning, validated in Alzheimer’s Disease Neuroimaging Initiative (ADNI) participants with baseline mild cognitive impairment. Brain Imaging Behav. 2012;6(4):517–27.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Choi SE, Mukherjee S, Gibbons LE, Sanders RE, Jones RN, Tommet D, et al. Development and validation of language and visuospatial composite scores in ADNI. Alzheimer’s & dementia (New York, N Y). 2020;6(1): e12072.

    Article  Google Scholar 

  26. Jack CR Jr, Bernstein MA, Fox NC, Thompson P, Alexander G, Harvey D, et al. The Alzheimer’s Disease Neuroimaging Initiative (ADNI): MRI methods. Journal of magnetic resonance imaging : JMRI. 2008;27(4):685–91.

    Article  PubMed  Google Scholar 

  27. Shen XN, Li JQ, Wang HF, Li HQ, Huang YY, Yang YX, et al. Plasma amyloid, tau, and neurodegeneration biomarker profiles predict Alzheimer’s disease pathology and clinical progression in older adults without dementia. Alzheimer’s & dementia (Amsterdam, Netherlands). 2020;12(1): e12104.

    Google Scholar 

  28. Yu JT, Li JQ, Suckling J, Feng L, Pan A, Wang YJ, et al. Frequency and longitudinal clinical outcomes of Alzheimer’s AT(N) biomarker profiles: A longitudinal study. Alzheimer’s Dementia. 2019;15(9):1208–17.

    Article  PubMed  Google Scholar 

  29. Shaw LM, Vanderstichele H, Knapik-Czajka M, Clark CM, Aisen PS, Petersen RC, et al. Cerebrospinal fluid biomarker signature in Alzheimer’s disease neuroimaging initiative subjects. Ann Neurol. 2009;65(4):403–13.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Xu W, Tan CC, Cao XP, Tan L. Neuroinflammation modulates the association of PGRN with cerebral amyloid-β burden. Neurobiol Aging. 2021;103:52–9.

    Article  CAS  PubMed  Google Scholar 

  31. Hansson O, Seibyl J, Stomrud E, Zetterberg H, Trojanowski JQ, Bittner T, et al. CSF biomarkers of Alzheimer’s disease concord with amyloid-β PET and predict clinical progression: A study of fully automated immunoassays in BioFINDER and ADNI cohorts. Alzheimer’s Dementia. 2018;14(11):1470–81.

    Article  PubMed  Google Scholar 

  32. Bittner T, Zetterberg H, Teunissen CE, Ostlund RE Jr, Militello M, Andreasson U, et al. Technical performance of a novel, fully automated electrochemiluminescence immunoassay for the quantitation of β-amyloid (1–42) in human cerebrospinal fluid. Alzheimer’s Dementia. 2016;12(5):517–26.

    Article  PubMed  Google Scholar 

  33. Malek-Ahmadi M. Reversion From Mild Cognitive Impairment to Normal Cognition: A Meta-Analysis. Alzheimer Dis Assoc Disord. 2016;30(4):324–30.

    Article  PubMed  Google Scholar 

  34. Yu HH, Tan CC, Huang SJ, Zhang XH, Tan L, Xu W. Predicting the reversion from mild cognitive impairment to normal cognition based on magnetic resonance imaging, clinical, and neuropsychological examinations. J Affect Disord. 2024;353:90–8.

    Article  PubMed  Google Scholar 

  35. O’Bryant SE, Johnson L, Reisch J, Edwards M, Hall J, Barber R, et al. Risk factors for mild cognitive impairment among Mexican Americans. Alzheimer’s Dementia. 2013;9(6):622–31.e1.

    Article  PubMed  Google Scholar 

  36. Pini L, Pievani M, Bocchetta M, Altomare D, Bosco P, Cavedo E, et al. Brain atrophy in Alzheimer’s Disease and aging. Ageing Res Rev. 2016;30:25–48.

    Article  PubMed  Google Scholar 

  37. Prosser L, Macdougall A, Sudre CH, Manning EN, Malone IB, Walsh P, et al. Predicting Cognitive Decline in Older Adults Using Baseline Metrics of AD Pathologies, Cerebrovascular Disease, and Neurodegeneration. Neurology. 2023;100(8):e834–45.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Li JQ, Tan L, Wang HF, Tan MS, Tan L, Xu W, et al. Risk factors for predicting progression from mild cognitive impairment to Alzheimer’s disease: a systematic review and meta-analysis of cohort studies. J Neurol Neurosurg Psychiatry. 2016;87(5):476–84.

    Article  PubMed  Google Scholar 

  39. Huang K, Lin Y, Yang L, Wang Y, Cai S, Pang L, et al. A multipredictor model to predict the conversion of mild cognitive impairment to Alzheimer’s disease by using a predictive nomogram. Neuropsychopharmacology : official publication of the American College of Neuropsychopharmacology. 2020;45(2):358–66.

    Article  PubMed  Google Scholar 

  40. Sachdev PS, Lipnicki DM, Crawford J, Reppermund S, Kochan NA, Trollor JN, et al. Factors predicting reversion from mild cognitive impairment to normal cognitive functioning: a population-based study. PLoS ONE. 2013;8(3): e59649.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Pagani M, Nobili F, Morbelli S, Arnaldi D, Giuliani A, Öberg J, et al. Early identification of MCI converting to AD: a FDG PET study. Eur J Nucl Med Mol Imaging. 2017;44(12):2042–52.

    Article  CAS  PubMed  Google Scholar 

  42. Caminiti SP, De Francesco S, Tondo G, Galli A, Redolfi A, Perani D. FDG-PET markers of heterogeneity and different risk of progression in amnestic MCI. Alzheimer's & dementia : the journal of the Alzheimer's Association. 2023.

  43. Landau SM, Harvey D, Madison CM, Koeppe RA, Reiman EM, Foster NL, et al. Associations between cognitive, functional, and FDG-PET measures of decline in AD and MCI. Neurobiol Aging. 2011;32(7):1207–18.

    Article  PubMed  Google Scholar 

  44. Trojanowski JQ, Vandeerstichele H, Korecka M, Clark CM, Aisen PS, Petersen RC, et al. Update on the biomarker core of the Alzheimer’s Disease Neuroimaging Initiative subjects. Alzheimer’s Dementia. 2010;6(3):230–8.

    Article  CAS  PubMed  Google Scholar 

  45. Palmqvist S, Mattsson N, Hansson O. Cerebrospinal fluid analysis detects cerebral amyloid-β accumulation earlier than positron emission tomography. Brain : a journal of neurology. 2016;139(Pt 4):1226–36.

    Article  PubMed  Google Scholar 

  46. Cruz-Oliver DM, Malmstrom TK, Roegner M, Tumosa N, Grossberg GT. Cognitive deficit reversal as shown by changes in the Veterans Affairs Saint Louis University Mental Status (SLUMS) examination scores 7.5 years later. J American Med Direct Assoc. 2014;15(9):687-e5-10.

    Google Scholar 

  47. Sanz-Blasco R, de Sánchez León JM, Ávila-Villanueva M, Valentí-Soler M, Gómez-Ramírez J, Fernández-Blázquez MA. Transition from mild cognitive impairment to normal cognition: Determining the predictors of reversion with multi-state Markov models. Alzheimers Dement. 2022;18(6):1177–85.

    Article  PubMed  Google Scholar 

  48. Calamia M, Markon K, Tranel D. Scoring higher the second time around: meta-analyses of practice effects in neuropsychological assessment. Clin Neuropsychol. 2012;26(4):543–70.

    Article  PubMed  Google Scholar 

  49. Duff K, Hammers DB, et al. Practice effects in mild cognitive impairment: A validation of Calamia,( 2012). The Clinical neuropsychologist. 2022;36(3):571–83.

    Article  PubMed  Google Scholar 

  50. Duff K, Callister C, Dennett K, Tometich D. Practice effects: a unique cognitive variable. Clin Neuropsychol. 2012;26(7):1117–27.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

The authors thank contributors, including the staff at Alzheimer’s Disease Centers who collected samples used in this study, patients, and their families whose help and participation made this work possible. Data collection and sharing for this project were funded by the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH-12-2-0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer’s Association; Alzheimer’s Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol-Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health (www.fnih.org). The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer’s Therapeutic Research Institute at the University of Southern California. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California.

The data used in preparation for this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in the analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf.

Funding

This study was supported by grants from the National Natural Science Foundation of China (82001136) and the Taishan Scholar Project (NO.tsqn202211375).

Author information

Authors and Affiliations

  1. Department of Neurology, Qingdao Municipal Hospital, Qingdao University, Donghai Middle Road, No.5, Qingdao, China

    Hai-Hong Yu, Lan Tan, Chen-Chen Tan & Wei Xu

  2. Medical College, Qingdao University, Qingdao, China

    Hai-Hong Yu, Meng-Jiao Jiao, Yi-Ju Lv & Xin-Hao Zhang

Authors
  1. Hai-Hong Yu
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Lan Tan
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Meng-Jiao Jiao
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Yi-Ju Lv
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Xin-Hao Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Chen-Chen Tan
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Wei Xu
    View author publications

    You can also search for this author inPubMed Google Scholar

Consortia

for the Alzheimer’s Disease Neuroimaging Initiative

Contributions

Prof. W Xu: conceptualization and design of the study and revision of the manuscript. Dr. H-H Yu: analysis of the data, drafting and revision of the manuscript, and prepared all the figures. Prof. L T: drafting and revision of the manuscript. Prof C-C T, Dr X-H Z, Dr. Y-J L, and Dr. M-J J: revision of the manuscript.

Corresponding author

Correspondence to Wei Xu.

Ethics declarations

Ethics approval and consent to participate

The ADNI was approved by the Institutional Review Boards of all participating centers, and written informed consent was obtained from all participants or authorized representatives according to the 1975 Declaration of Helsinki. More information is available at the ADNI website (https://adni.loni.usc.edu/).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

12916_2025_4092_MOESM1_ESM.doc

Additional file 1: Supplementary Results: Detailed descriptions of the results on verbal memory and specific executive function. Table 1. Population characteristics at original baseline. Table 2. Hazard ratios for progression of MCI Reverters to MCI or AD compared to Stable NC. Table 3. Cox proportional hazards regression model to estimate the risk of cognitive decline at the last visit in sensitivity analysis. Table 4. Linear mixed effects model results for longitudinal rates of change in neuropsychological measures. Table 5. Linear mixed effects model results for longitudinal rates of change in Magnetic Resonance Imaging measures. Table 6. Linear mixed effects model results for longitudinal rates of change in Positron Emission Tomography measures. Table 7. Linear mixed effects model results for longitudinal rates of change in Cerebral Spinal Fluid biomarkers. Table 8. APOE4 status interacted with longitudinal rates of change in MRI, PET, CSF and neuropsychological measures by MCI Reverters or Stable MCI. Table 9. Sex interacted with longitudinal rates of change in MRI, PET, CSF and neuropsychological measures by MCI Reverters or Stable MCI. Table 10. Linear mixed effects model results for longitudinal rates of change in neuropsychological measures in sensitivity analysis. Table 11. Linear mixed effects model results for longitudinal rates of change in Magnetic Resonance Imaging measures in sensitivity analysis. Table 12. Linear mixed effects model results for longitudinal rates of change in Positron Emission Tomography measures in sensitivity analysis. Table 13. Linear mixed effects model results for longitudinal rates of change in Cerebral Spinal Fluid biomarkers in sensitivity analysis.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Yu, HH., Tan, L., Jiao, MJ. et al. Dissecting the clinical and pathological prognosis of MCI patients who reverted to normal cognition: a longitudinal study. BMC Med 23, 260 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12916-025-04092-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12916-025-04092-0

Keyword

BMC Medicine

ISSN: 1741-7015

Contact us