Genetic and Clinical Correlates of AI-Based Brain Aging Patterns in Cognitively Unimpaired Individuals

Importance

Brain aging elicits complex neuroanatomical changes influenced by multiple age-related pathologies. Understanding the heterogeneity of structural brain changes in aging may provide insights into preclinical stages of neurodegenerative diseases.

Objective

To derive subgroups with common patterns of variation in participants without diagnosed cognitive impairment (WODCI) in a data-driven manner and relate them to genetics, biomedical measures, and cognitive decline trajectories.

Design, Setting, and Participants

Data acquisition for this cohort study was performed from 1999 to 2020. Data consolidation and harmonization were conducted from July 2017 to July 2021. Age-specific subgroups of structural brain measures were modeled in 4 decade-long intervals spanning ages 45 to 85 years using a deep learning, semisupervised clustering method leveraging generative adversarial networks. Data were analyzed from July 2021 to February 2023 and were drawn from the Imaging-Based Coordinate System for Aging and Neurodegenerative Diseases (iSTAGING) international consortium. Individuals WODCI at baseline spanning ages 45 to 85 years were included, with greater than 50 000 data time points.

Exposures

Individuals WODCI at baseline scan.

Main Outcomes and Measures

Three subgroups, consistent across decades, were identified within the WODCI population. Associations with genetics, cardiovascular risk factors (CVRFs), amyloid β (Aβ), and future cognitive decline were assessed.

Results

In a sample of 27 402 individuals (mean [SD] age, 63.0 [8.3] years; 15 146 female [55%]) WODCI, 3 subgroups were identified in contrast with the reference group: a typical aging subgroup, A1, with a specific pattern of modest atrophy and white matter hyperintensity (WMH) load, and 2 accelerated aging subgroups, A2 and A3, with characteristics that were more distinct at age 65 years and older. A2 was associated with hypertension, WMH, and vascular disease-related genetic variants and was enriched for Aβ positivity (ages ≥65 years) and apolipoprotein E (APOE) ε4 carriers. A3 showed severe, widespread atrophy, moderate presence of CVRFs, and greater cognitive decline. Genetic variants associated with A1 were protective for WMH (rs7209235: mean [SD] B = -0.07 [0.01]; P value = 2.31 × 10-9) and Alzheimer disease (rs72932727: mean [SD] B = 0.1 [0.02]; P value = 6.49 × 10-9), whereas the converse was observed for A2 (rs7209235: mean [SD] B = 0.1 [0.01]; P value = 1.73 × 10-15 and rs72932727: mean [SD] B = -0.09 [0.02]; P value = 4.05 × 10-7, respectively); variants in A3 were associated with regional atrophy (rs167684: mean [SD] B = 0.08 [0.01]; P value = 7.22 × 10-12) and white matter integrity measures (rs1636250: mean [SD] B = 0.06 [0.01]; P value = 4.90 × 10-7).

Conclusions and Relevance

The 3 subgroups showed distinct associations with CVRFs, genetics, and subsequent cognitive decline. These subgroups likely reflect multiple underlying neuropathologic processes and affect susceptibility to Alzheimer disease, paving pathways toward patient stratification at early asymptomatic stages and promoting precision medicine in clinical trials and health care.

Plasma Biomarkers as Predictors of Progression to Dementia in Individuals with Mild Cognitive Impairment

Background

Blood-based biomarkers (BBMs) are of growing interest in the field of Alzheimer's disease (AD) and related dementias.

Objective

This study aimed to assess the ability of plasma biomarkers to 1) predict disease progression from mild cognitive impairment (MCI) to dementia and 2) improve the predictive ability of magnetic resonance imaging (MRI) and cerebrospinal fluid (CSF) measures when combined.

Methods

We used data from the Alzheimer's Disease Neuroimaging Initiative. Machine learning models were trained using the data from participants who remained cognitively stable (CN-s) and with Dementia diagnosis at 2-year follow-up visit. The models were used to predict progression to dementia in MCI individuals. We assessed the performance of models with plasma biomarkers against those with CSF and MRI measures, and also in combination with them.

Results

Our models with plasma biomarkers classified CN-s individuals from AD with an AUC of 0.75±0.03 and could predict conversion to dementia in MCI individuals with an AUC of 0.64±0.03 (17.1% BP, base prevalence). Models with plasma biomarkers performed better when combined with CSF and MRI measures (CN versus AD: AUC of 0.89±0.02; MCI-to-AD: AUC of 0.76±0.03, 21.5% BP).

Conclusions

Our results highlight the potential of plasma biomarkers in predicting conversion to dementia in MCI individuals. While plasma biomarkers could improve the predictive ability of CSF and MRI measures when combined, they also show the potential to predict non-progression to AD when considered alone. The predictive ability of plasma biomarkers is crucially linked to reducing the costly and effortful collection of CSF and MRI measures.

Association of Stages of Objective Memory Impairment With Incident Symptomatic Cognitive Impairment in Cognitively Normal Individuals

BACKGROUND AND OBJECTIVES

Increasing evidence indicates that a subset of cognitively normal individuals has subtle cognitive impairment at baseline. We sought to identify them using the Stages of Objective Memory Impairment (SOMI) system. Symptomatic cognitive impairment was operationalized by a Clinical Dementia Rating (CDR) ≥0.5. We hypothesized that incident impairment would be higher for participants with subtle retrieval impairment (SOMI-1), higher still for those with moderate retrieval impairment (SOMI-2), and highest for those with storage impairment (SOMI-3/4) after adjusting for demographics and APOE ε4 status. A secondary objective was to determine whether including biomarkers of β-amyloid, tau pathology, and neurodegeneration in the models affect prediction. We hypothesized that even after adjusting for in vivo biomarkers, SOMI would remain a significant predictor of time to incident symptomatic cognitive impairment.

METHODS

Among 969 cognitively normal participants, defined by a CDR = 0, from the Knight Alzheimer Disease Research Center, SOMI stage was determined from their baseline Free and Cued Selective Reminding Test scores, 555 had CSF and structural MRI measures and comprised the biomarker subgroup, and 144 of them were amyloid positive. Cox proportional hazards models tested associations of SOMI stages at baseline and biomarkers with time to incident cognitive impairment defined as the transition to CDR ≥0.5.

RESULTS

Among all participants, the mean age was 69.35 years, 59.6% were female, and mean follow-up was 6.36 years. Participants in SOMI-1-4 had elevated hazard ratios for the transition from normal to impaired cognition in comparison with those who were SOMI-0 (no memory impairment). Individuals in SOMI-1 (mildly impaired retrieval) and SOMI-2 (moderately impaired retrieval) were at nearly double the risk of clinical progression compared with persons with no memory problems. When memory storage impairment emerges (SOMI-3/4), the hazard ratio for clinical progression increased approximately 3 times. SOMI stage remained an independent predictor of incident cognitive impairment after adjusting for all biomarkers.

DISCUSSION

SOMI predicts the transition from normal cognition to incident symptomatic cognitive impairment (CDR ≥0.5). The results support the use of SOMI to identify those cognitively normal participants most likely to develop incident cognitive impairment who can then be referred for biomarker screening.

Predictors of treatment-response to caffeine combination products, acetaminophen, acetylsalicylic acid (aspirin), and nonsteroidal anti-inflammatory drugs in acute treatment of episodic migraine

OBJECTIVE

To identify predictors of acute treatment optimization for migraine with "over-the-counter" (OTC) or prescription nonsteroidal anti-inflammatory drugs (NSAIDs) as well as other widely used OTCs including acetaminophen, caffeine combination products (CCP), and acetylsalicylic acid (ASA, aspirin) among people with episodic migraine and to develop models that predict treatment response to each class of OTCs.

BACKGROUND

Efficacy of acute OTC medications for migraine varies greatly. Identifying predictors of treatment response to particular classes of medication is a step toward evidence-based personalized therapy.

METHODS

For this prediction model development study, we used data from 2224 participants from the American Migraine Prevalence and Prevention (AMPP) study who were aged ≥18 years, met criteria for migraine, had <15 monthly headache days, and reported being on monotherapy for acute migraine attacks with one of the following classes medications: CCP (N = 711), acetaminophen (N = 643), ASA (N = 110), and prescription or OTC NSAIDs (N = 760). The primary outcome measures of treatment optimization were adequate 2-h pain freedom (2hPF) and adequate 24-h pain relief (24hPR), which were defined by responses of half the time or more to the relevant items on the Migraine Treatment Optimization Questionnaire-6.

RESULTS

The mean (SD) age of the participants was 46.2 (13.1) years, 79.4% (1765/2224) were female, 43.7% (972/2224) reported adequate 2hPF, and 46.1% (1025/2224) reported adequate 24hPR. Those taking CCP had better 2hPF and 24PR outcomes. For those taking NSAIDs, better outcomes were associated with lower average pain intensity (2hPF: odds ratio [OR] 0.89, 95% confidence interval [CI] 0.80-0.99; 24PR: OR 0.86, 95% CI 0.77-0.96), cutaneous allodynia (2hPF: OR 0.92, 95% CI 0.89-0.96; 24PR: OR 0.91, 95% CI 0.87-0.95), depressive symptoms (2hPF: OR 0.95, 95% CI 0.92-0.98; 24PR: OR 0.95, 95% CI 0.91-0.99), and Migraine Disability Assessment Scale (MIDAS) grade (2hPF: OR 0.76, 95% CI 0.64-0.90; 24PR: OR 0.79, 95% CI 0.65-0.95). Adequate 2hPF for those taking CCP was associated with male gender (OR 1.83, 95% CI 1.21-2.77), lower average pain intensity (OR 0.80, 95% CI 0.70-0.91), lower cutaneous allodynia (OR 0.94, 95% CI 0.90-0.97), and lower Migraine Symptom Severity Scale Score (MSSS; OR 0.91, 95% CI 0.86-0.97). Adequate 24hPR for those taking CCP was associated with lower average pain intensity (OR 0.85, 95% CI 0.75-0.96), lower cutaneous allodynia (OR 0.92, 95% CI 0.89-0.96), and lower MIDAS grade (OR 0.81, 95% CI 0.68-0.96). Participants who were married (OR 1.51, 95% CI 1.05-2.19), had lower average pain intensity (OR 0.79, 95% CI 0.70-0.89), lower MSSS (OR 0.93, 95% CI 0.88-0.99), less depression (OR 0.96, 95% CI 0.93-0.99), and lower MIDAS grade (OR 0.72, 95% CI 0.59-0.87) had adequate 2hPF after taking acetaminophen. Participants who were married (OR 1.50, 95% CI 1.02-2.21), had lower pain intensity (OR 0.78, 95% CI 0.69-0.88), less depression (OR 0.95, 95% CI 0.91-0.98) and lower MIDAS grade (OR 0.53, 95% CI 0.42-0.67) had higher 24hPR following use of acetaminophen. A lower MSSS was the only factor associated with higher 2hPF and 24PR after using ASA (OR 0.78, 95% CI 0.67-0.92 and OR 0.79, 95% CI 0.67-0.93). Predictive models had modest performance in identifying responders to each class of OTC.

CONCLUSION

A large subgroup of people with migraine had an inadequate response to their usual acute OTC migraine treatment 2- and 24-h after dosing. These findings suggest a need to improve OTC treatment for some and to offer prescription acute medications for others. Predictive models identified several factors associated with better treatment-response in each OTC class. Selecting OTC treatment based on factors predictive of treatment optimization might improve patient outcomes.

Tracking cognition with the T-MoCA in a racially/ethnically diverse older adult cohort

Introduction

We investigated the utility of the Telephone-Montreal Cognitive Assessment (T-MoCA) to track cognition in a diverse sample from the Einstein Aging Study.

Methods

Telephone and in-person MoCA data, collected annually, were used to evaluate longitudinal cognitive performance. Joint models of T-MoCA and in-person MoCA compared changes, variance, and test-retest reliability measured by intraclass correlation coefficient by racial/ethnic group.

Results

There were no significant differences in baseline performance or longitudinal changes across three study waves for both MoCA formats. T-MoCA performance improved over waves 1-3 but declined afterward. Test-retest reliability was lower for the T-MoCA than for the in-person MoCA. In comparison with non-Hispanic Whites, non-Hispanic Blacks and Hispanics performed worse at baseline on both MoCA formats and showed lower correlations between T-MoCA and in-person versions.

Conclusions

The T-MoCA provides valuable information on cognitive change, despite racial/ethnic disparities and practice effects. We discuss implications for health disparity populations.

Highlights

We assessed the comparability of Telephone-Montreal Cognitive Assessment (T-MoCA) and in-person MoCA for tracking cognition.Changes within 3 years in T-MoCA were similar to that for the in-person MoCA.T-MoCA is subject to practice effects and shows difference in performance by race/ethnicity.Test-retest reliability of T-MoCA is lower than that for in-person MoCA.

Associations of Stages of Objective Memory Impairment with Cerebrospinal Fluid and Neuroimaging Biomarkers of Alzheimer's Disease

OBJECTIVE

To investigate cerebrospinal fluid (CSF) and neuroimaging correlates of Stages of Objective Memory Impairment (SOMI) based on Free and Cued Selective Reminding Test (FCSRT) performance, and to evaluate the effect of APOE ε4 status on this relationship.

METHODS

Data from 586 cognitively unimpaired individuals who had FCSRT, CSF, and volumetric magnetic resonance imaging (MRI) measures available was used. We compared CSF measures of β-amyloid (Aβ42/Aβ40 ratio), phosphorylated tau (p-Tau181), total tau (t-Tau), hippocampal volume, and PIB-PET mean cortical binding potential with partial volume correction (MCBP) among SOMI groups in the whole sample and in subsamples stratified by APOE ε4 status.

RESULTS

Participants had a mean age of 67.4 (SD=9.1) years, had 16.1 (SD=2.6) years of education, 57.0% were female, and 33.8% were APOE ε4 positive. In the entire sample, there was no significant difference between SOMI stages in Aβ42/Aβ40 ratio, p-Tau181, t-Tau, or PIB-PET MCBP when adjusted for age, sex, and education. However, higher SOMI stages had smaller hippocampal volume (F=3.29, p=0.020). In the stratified sample based on APOE ε4 status, in APOE ε4 positive individuals, higher SOMI stages had higher p-Tau181 (F=2.94, p=0.034) higher t-Tau (F=3.41, p=0.019), and smaller hippocampal volume (F=5.78, p<0.001). There were no significant differences in CSF or imaging biomarkers between SOMI groups in the APOE ε4 negative subsample.

CONCLUSION

Cognitively normal older individuals with higher SOMI stages have higher in-vivo tau and neurodegenerative pathology only in APOE ε4 carriers. These original results indicate the potential usefulness of the SOMI staging system in assessing of tau and neurodegenerative pathology.

Association of Alcohol Consumption with Cognition in Older Population: The A4 Study

BACKGROUND

Alcohol use disorders have been categorized as a 'strongly modifiable' risk factor for dementia.

OBJECTIVE

To investigate the cross-sectional association between alcohol consumption and cognition in older adults and if it is different across sexes or depends on amyloid-β (Aβ) accumulation in the brain.

METHODS

Cognitively unimpaired older adults (N = 4387) with objective and subjective cognitive assessments and amyloid positron emission tomography (PET) imaging were classified into four categories based on their average daily alcohol use. Multivariable linear regression was then used to test the main effects and interactions with sex and Aβ levels.

RESULTS

Individuals who reported no alcohol consumption had lower scores on the Preclinical Alzheimer Cognitive Composite (PACC) compared to those consuming one or two drinks/day. In sex-stratified analysis, the association between alcohol consumption and cognition was more prominent in females. Female participants who consumed two drinks/day had better performance on PACC and Cognitive Function Index (CFI) than those who reported no alcohol consumption. In an Aβ-stratified sample, the association between alcohol consumption and cognition was present only in the Aβ- subgroup. The interaction between Aβ status and alcohol consumption on cognition was not significant.

CONCLUSION

Low or moderate consumption of alcohol was associated with better objective cognitive performance and better subjective report of daily functioning in cognitively unimpaired individuals. The association was present only in Aβ- individuals, suggesting that the pathophysiologic mechanism underlying the effect of alcohol on cognition is independent of Aβ pathology. Further investigation is required with larger samples consuming three or more drinks/day.

Trans-Tubular Translaminar Microscopic-Assisted Nucleotomy for Lumbar Disc Herniations in the Hidden Zone

STUDY DESIGN

A prospective cohort study in a high-flow spine center in Germany.

OBJECTIVES

This study aimed to evaluate clinical outcomes and complications of the trans-tubular translaminar microscopic-assisted percutaneous nucleotomy in cases of cranially migrated lumbar disc herniations (LDH).

METHODS

Between January 2013 and January 2018, 66 consecutive patients with cranio-laterally migrated LDH were operated upon. The following outcome measures were evaluated: (1) Visual Analog Scale (VAS) for leg and back pain; (2) Oswestry Disability Index (ODI) and Macnab´s criteria. All patients were operated upon with trans-tubular Translaminar Microscopic-assisted Percutaneous Nucleotomy (TL-MAPN). Perioperative radiographic and clinical evaluations were reported. The mean follow-up period was 32 months.

RESULTS

The mean age was 59 years. L4/L5 was the commonest affected level (27 patients). The mean preoperative VAS for leg pain was 6.44 (±2.06), improved to 0,35 (±0.59) postoperatively. Dural injury occurred in 1 patient, treated with dural patch. Improved neurological function was reported in 41/44 Patients (neurological improvement rate of 93%) at the final follow up. There was a significant improvement in the mean ODI values, from 50.19 ± 4.92 preoperatively to 10.14 ± 2.22 postoperatively (P < 0.001). Sixty four out of 66 patients (96%) showed an excellent or good functional outcome according to Macnab´s criteria. No recurrent herniations were observed.

CONCLUSION

The translaminar approach is a viable minimal invasive technique for cranially migrated LDH. The preservation of the flavum ligament is one of the main advantages of this technique. It is an effective, safe and reproducible minimally invasive surgical alternative in treatment of cranially migrated LDHs.

Predicting Amyloid Positivity in Cognitively Unimpaired Older Adults: A Machine Learning Approach Using A4 Data

BACKGROUND AND OBJECTIVES

To develop and test the performance of the Positive Aβ Risk Score (PARS) for prediction of β-amyloid (Aβ) positivity in cognitively unimpaired individuals for use in clinical research. Detecting Aβ positivity is essential for identifying at-risk individuals who are candidates for early intervention with amyloid targeted treatments.

METHODS

We used data from 4,134 cognitively normal individuals from the Anti-Amyloid Treatment in Asymptomatic Alzheimer's (A4) Study. The sample was divided into training and test sets. A modified version of AutoScore, a machine learning-based software tool, was used to develop a scoring system using the training set. Three risk scores were developed using candidate predictors in various combinations from the following categories: demographics (age, sex, education, race, family history, body mass index, marital status, and ethnicity), subjective measures (Alzheimer's Disease Cooperative Study Activities of Daily Living-Prevention Instrument, Geriatric Depression Scale, and Memory Complaint Questionnaire), objective measures (free recall, Mini-Mental State Examination, immediate recall, digit symbol substitution, and delayed logical memory scores), and APOE4 status. Performance of the risk scores was evaluated in the independent test set.

RESULTS

PARS model 1 included age, body mass index (BMI), and family history and had an area under the curve (AUC) of 0.60 (95% CI 0.57-0.64). PARS model 2 included free recall in addition to the PARS model 1 variables and had an AUC of 0.61 (0.58-0.64). PARS model 3, which consisted of age, BMI, and APOE4 information, had an AUC of 0.73 (0.70-0.76). PARS model 3 showed the highest, but still moderate, performance metrics in comparison with other models with sensitivity of 72.0% (67.6%-76.4%), specificity of 62.1% (58.8%-65.4%), accuracy of 65.3% (62.7%-68.0%), and positive predictive value of 48.1% (44.1%-52.1%).

DISCUSSION

PARS models are a set of simple and practical risk scores that may improve our ability to identify individuals more likely to be amyloid positive. The models can potentially be used to enrich trials and serve as a screening step in research settings. This approach can be followed by the use of additional variables for the development of improved risk scores.

CLASSIFICATION OF EVIDENCE

This study provides Class II evidence that in cognitively unimpaired individuals PARS models predict Aβ positivity with moderate accuracy.

Predictive models for determining treatment response to nonprescription acute medications in migraine: Results from the American Migraine Prevalence and Prevention Study

OBJECTIVE

To identify predictors of acute treatment response for nonprescription (over-the-counter [OTC]) medications among people with migraine and develop improved models for predicting treatment response.

BACKGROUND

Pain freedom and sustained pain relief are important priorities in the acute treatment of migraine. OTC medications are widely used for migraine; however, it is not clear which treatment works best for each patient without going through the trial and error process.

METHODS

A prediction model development study was completed using the 2006 American Migraine Prevalence and Prevention Study survey, from participants who were aged ≥18, met criteria and headache day frequency for episodic migraine, did not take prescription medication for migraine, and used ≥1 of the following acute migraine medication classes: acetaminophen, aspirin, NSAIDs, or caffeine containing combination products (CCP). Two items from the Migraine Treatment Optimization Questionnaire were used to evaluate treatment response, adequate 2-h pain freedom (2hPF) and 24-h pain relief (24hPR), which were defined by a response to treatment ≥half the time at 2 h and 24 h post treatment, respectively. We identified predictors of adequate treatment response and developed models to predict probability of treatment response to each medication class.

RESULTS

The sample included 3852 participants (3038 [79.0%] females) with an average age of 45.0 years (SD = 12.8). Only 1602/3852 (41.6%) and 1718/3852 (44.6%) of the participants reported adequate 2hPF and 24hPR, respectively. Adequate treatment-response was significantly predicted by lower average headache pain intensity, less cutaneous allodynia, and lower depressive symptom scores. Lower migraine symptom severity was predictive of adequate 2hPF and fewer monthly headache days was predictive of adequate 24hPR. Among participants reporting OTC monotherapy (n = 2168, 56.3%) individuals taking CCP were more likely to have adequate 2hPF (OR = 1.55, 95% CI 1.23-1.95) and 24hPR (OR = 1.79, 95% CI 1.18-1.88) in comparison with those taking acetaminophen. Predictive models were modestly predictive of responders to OTC medications (c-statistics = 0.65; 95% CI 0.62-0.68).

CONCLUSION

These results show that response to acute migraine treatments is not optimized in the majority of people with migraine treating with OTC medications. Predictive models can improve our ability to choose the best therapeutic option for individuals with episodic migraine and increase the proportion of patients with optimized response to treatments.

Comparing Performance of Different Predictive Models in Estimating Disease Progression in Alzheimer Disease

BACKGROUND

Automatic classification techniques provide tools to analyze complex data and predict disease progression.

METHODS

A total of 305 cognitively normal; 475 patients with amnestic mild cognitive impairment (aMCI); and 162 patients with dementia were included in this study. We compared the performance of 3 different methods in predicting progression from aMCI to dementia: (1) index-based model; (2) logistic regression (LR); and (3) ensemble linear discriminant (ELD) machine learning models. LR and ELD models were trained using data from cognitively normal and dementia subgroups, and subsequently were applied to aMCI subgroup to predict their disease progression.

RESULTS

Performance of ELD models were better than LR models in prediction of conversion from aMCI to Alzheimer dementia at all time frames. ELD models performed better when a larger number of features were used for prediction.

CONCLUSION

Machine learning models have substantial potential to improve the predictive ability for cognitive outcomes.

Associations of Stages of Objective Memory Impairment With Amyloid PET and Structural MRI: The A4 Study

BACKGROUND AND OBJECTIVES

The goal of this work was to investigate the neuroimaging correlates of the Stages of Objective Memory Impairment (SOMI) system operationalized with the Free and Cued Selective Reminding Test (FCSRT), a widely used episodic memory measure.

METHODS

The FCSRT begins with a study phase in which items (e.g., grapes) are identified in response to unique semantic cues (e.g., fruit) that are used in the test phase to prompt recall of items not retrieved by free recall. There are 3 test trials of the 16 items (maximum 48). Data from 4,484 cognitively unimpaired participants from the Anti-Amyloid Treatment in Asymptomatic Alzheimer's (A4) study were used. All participants had amyloid PET imaging, and a subset of 1,262 β-amyloid (Aβ)-positive had structural MRIs. We compared the Aβ mean cortical standardized uptake value ratio (SUVR) and volumetric measures of hippocampus, parahippocampal gyrus, entorhinal cortex, and inferior temporal cortex between the 5 SOMI stages.

RESULTS

Participants had a mean age of 71.3 (SD 4.6) years; 40.6% were male; and 34.6% were APOE ε4 positive. Half had no memory impairment; the other half had retrieval deficits, storage limitations, or both. Analysis of covariance in the entire sample while controlling for age, sex, education, and APOE ε4 showed that individuals in higher SOMI stages had higher global amyloid SUVR (p < 0.001). Both SOMI-4 and -3 subgroups had higher amyloid SUVR than SOMI-0 and SOMI-1 subgroups. Individuals in higher SOMI stages had smaller hippocampal volume (p = 0.003), entorhinal cortex (p < 0.05), and inferior temporal lobes (p < 0.05), but there was no difference between parahippocampal gyrus volume of different SOMI stages. Pairwise comparison of SOMI subgroups showed that the SOMI-4, -3, and -2 subgroups had smaller hippocampal volume than the SOMI-0 and -1 subgroup. The SOMI-4 subgroup had significantly smaller entorhinal cortex and smaller inferior temporal lobe compared to all other groups.

DISCUSSION

Presence of Alzheimer disease pathology is closely related to memory impairment according to SOMI stages in the cognitively unimpaired sample of A4. Results from structural MRIs suggest that memory storage impairment (SOMI-3 and -4) is present when there is widespread medial temporal lobe atrophy.

TRIAL REGISTRATION INFORMATION

ClinicalTrials.gov identifier: NCT02008357.

CLASSIFICATION OF EVIDENCE

This study provides Class I evidence that, in normal older individuals, higher stages of memory impairment assessed with FCSRT were associated with higher amyloid imaging burden and lower volume of hippocampus, entorhinal cortex, and inferior temporal lobes.

Impact of sex and APOE ε4 on the association of cognition and hippocampal volume in clinically normal, amyloid positive adults

INTRODUCTION

Cognitive decline follows pathological changes including neurodegeneration on the Alzheimer's disease continuum. However, it is unclear which cognitive domains first become affected by neurodegeneration in amyloid-positive individuals and if sex or apolipoprotein (APOE) ε4 status differences affect this relationship.

METHODS

Data from 1233 cognitively unimpaired, amyloid-positive individuals 65 to 85 years of age were studied to assess the effect of hippocampal volume (HV) on cognition and to evaluate differences due to sex and APOE ε4 status.

RESULTS

Lower HV was linked with worse performance on measures of memory (free recall, total recall, logical memory delayed recall, Mini-Mental State Examination [MMSE]), executive functioning (digit symbol substitution, DSS), and the Preclinical Alzheimer's Cognitive Composite (PACC). Among both women and APOE ε4+ individuals, all cognitive measures, except MMSE, were associated with HV. DSS and PACC had the largest effect sizes in differentiating early and intermediate stage neurodegeneration.

DISCUSSION

Despite all cognitive measures being associated with HV, cognitive tests show differences in detecting early or late signs of neurodegeneration. Differences exist in association between cognition and neurodegeneration based on sex and APOE ε4 status.

Anterior cervical corpectomy and fusion with stand-alone cages in patients with multilevel degenerative cervical spine disease is safe

Background

In case of spinal cord compression behind the vertebral body, anterior cervical corpectomy and fusion (ACCF) proves to be a more feasible approach than cervical discectomy. The next step was the placement of an expandable titanium interbody in order to restore the vertebral height. The need for additional anterior plating with ACCF has been debatable and such technique has been evaluated by very few studies. The objective of the study is to evaluate radiographic and clinical outcomes in patients with multilevel degenerative cervical spine disease treated by stand-alone cages for anterior cervical corpectomy and fusion (ACCF).

Methods

Thirty-one patients (66.5 ± 9.75 years, range 53–85 years) were analyzed. Visual Analog Scale (VAS) and the 10-item Neck Disability Index (NDI) were assessed preoperatively and during follow-up on a regular basis after surgery and after one year at least. Assessment of radiographic fusion, subsidence, and lordosis measurement of Global cervical lordosis (GCL); fusion site lordosis (FSL); the anterior interbody space height (ant. DSH); the posterior interbody space height (post. DSH); the distance of the cage to the posterior wall of the vertebral body (CD) were done retrospectively. Mean clinical and radiographic follow-up was 20.0 ± 4.39 months.

Results

VAS-neck (p = 0.001) and VAS-arm (p < 0.001) improved from preoperatively to postoperatively. The NDI improved at the final follow-up (p < 0.001). Neither significant subsidence of the cages nor significant loss of lordotic correction were seen. All patients showed a radiographic union of the surgically addressed segments at the last follow up.

Conclusions

Application of a stand-alone expandable cage in the cervical spine after one or two-level ACCF without additional posterior fixation or anterior plating is a safe procedure that results in fusion. Neither significant subsidence of the cages nor significant loss of lordotic correction were seen.

Trial registration

Retrospectively registered. According to the Decision of the ethics committee, Jena on 25th of July 2018, that this study doesn’t need any registration. https://www.laek-thueringen.de/aerzte/ethikkommission/registrierung/.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12891-021-04883-5.

Association of Depressive Symptoms and Cognition in Older Adults Without Dementia Across Different Biomarker Profiles

BACKGROUND

Depression is a late-life risk factor for cognitive decline. Evidence suggests an association between Alzheimer's disease (AD) associated pathologic changes and depressive symptoms.

OBJECTIVE

To investigate the influence of AT(N) biomarker profile (amyloid-β [A], p-tau [T], and neurodegeneration [N]) and gender on cross-sectional associations between subclinical depressive symptoms and cognitive function among older adults without dementia.

METHODS

Participants included 868 individuals without dementia from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Depressive symptoms were measured using the Geriatric Depression Scale (GDS). ADNI neuropsychological composite scores assessed memory and executive function (EF). PET, cerebrospinal fluid, and MRI modalities classified the study sample into biomarker profiles: normal biomarkers (A-T-N-), AD continuum (A+T±N±), and suspect non-AD pathology (SNAP; A-T±N-or A-T-N±). Multivariate regression models were used to investigate associations between GDS and cognitive domains.

RESULTS

GDS was negatively associated with memory (β= -0.156, p < 0.001) and EF (β= -0.147, p < 0.001) in the whole sample. When classified by biomarker profile, GDS was negatively associated with memory and EF in AD continuum (memory: β= -0.174, p < 0.001; EF: β= -0.129 p = 0.003) and SNAP (memory: β= -0.172, p = 0.005; EF: β= -0.197, p = 0.001) subgroups. When stratified by sex, GDS was negatively associated with memory (β= -0.227, p < 0.001) and EF (β= -0.205, p < 0.001) in men only.

CONCLUSION

The association between subclinical depressive symptoms and cognitive function is highly influenced by the AT(N) biomarker profile.

Application of predictive models in boosting power of Alzheimer's disease clinical trials: A post hoc analysis of phase 3 solanezumab trials

Background

The ideal participants for Alzheimer's disease (AD) clinical trials would show cognitive decline in the absence of treatment (i.e., placebo arm) and would also respond to the therapeutic intervention.

Objective

To investigate if predictive models can be an effective tool for identifying and excluding people unlikely to show cognitive decline as an enrichment strategy in AD trials.

Method

We used data from the placebo arms of two phase 3, double-blind trials, EXPEDITION and EXPEDITION2. Patients had 18 months of follow-up. Based on the longitudinal data from the placebo arm, we classified participants into two groups: one showed cognitive decline (any negative slope) and the other showed no cognitive decline (slope is zero or positive) on the Alzheimer's Disease Assessment Scale-Cognitive subscale (ADAS-cog). We used baseline data for EXPEDITION to train regression-based classifiers and machine learning classifiers to estimate probability of cognitive decline. Models were applied to EXPEDITION2 data to assess predicted performance in an independent sample. Features used in predictive models included baseline demographics, apolipoprotein E ε4 genotype, neuropsychological scores, functional scores, and volumetric magnetic resonance imaging.

Result

In EXPEDITION, 46.3% of placebo-treated patients showed no cognitive decline and the proportion was similar in EXPEDITION2 (45.6%). Models had high sensitivity and modest specificity in both the training (EXPEDITION) and replication samples (EXPEDITION2) for detecting the stable group. Positive predictive value of models was higher than the base prevalence of cognitive decline, and negative predictive value of models were higher than the base rate of participants who had stable cognition.

Conclusion

Excluding persons with AD unlikely to decline from the active and placebo arms of clinical trials using predictive models may boost the power of AD trials through selective inclusion of participants expected to decline.

Neuroimaging correlates of Stages of Objective Memory Impairment (SOMI) system

INTRODUCTION

To assess the relationship between memory performance defined by the Stages of Objective Memory Impairment (SOMI) system and the Alzheimer's disease (AD) ATN (amyloid beta [A], pathologic tau [T], and neurodegeneration [N]) biomarker system.

METHODS

We used data from the Harvard Aging Brain Study cohort to estimate the level of ATN biomarkers: amyloid beta (C-Pittsburgh compound B-positron emission tomography [PET]), tau (F-18-flortaucipir [FTP] PET), and neurodegeneration (magnetic resonance imaging volumetrics). We assessed the cross-sectional relationship of SOMI classification with global amyloid levels, entorhinal and inferior temporal tau deposition, and hippocampal atrophy.

RESULTS

Participants with both memory storage and retrieval deficits (SOMI-3, -4) had smaller hippocampal volumes and higher entorhinal and inferior temporal tau burden than participants with no memory impairment (SOMI-0) or mild retrieval difficulty (SOMI-1). Amyloid burden did not differ among SOMI stages.

DISCUSSION

This pilot supports the close relationship between tau pathology and memory impairment across the AD continuum.  SOMI may be useful to determine eligibility for randomized controlled trials prior to the assessment of biomarker status.

Predictive value of ATN biomarker profiles in estimating disease progression in Alzheimer's disease dementia

We aimed to evaluate the value of ATN biomarker classification system (amyloid beta [A], pathologic tau [T], and neurodegeneration [N]) for predicting conversion from mild cognitive impairment (MCI) to dementia. In a sample of people with MCI (n = 415) we assessed predictive performance of ATN classification using empirical knowledge-based cut-offs for each component of ATN and compared it to two data-driven approaches, logistic regression and RUSBoost machine learning classifiers, which used continuous clinical or biomarker scores. In data-driven approaches, we identified ATN features that distinguish normals from individuals with dementia and used them to classify persons with MCI into dementia-like and normal groups. Both data-driven classification methods performed better than the empirical cut-offs for ATN biomarkers in predicting conversion to dementia. Classifiers that used clinical features performed as well as classifiers that used ATN biomarkers for prediction of progression to dementia. We discuss that data-driven modeling approaches can improve our ability to predict disease progression and might have implications in future clinical trials.

Leveraging machine learning predictive biomarkers to augment the statistical power of clinical trials with baseline magnetic resonance imaging

A key factor in designing randomized clinical trials is the sample size required to achieve a particular level of power to detect the benefit of a treatment. Sample size calculations depend upon the expected benefits of a treatment (effect size), the accuracy of measurement of the primary outcome, and the level of power specified by the investigators. In this study, we show that radiomic models, which leverage complex brain MRI patterns and machine learning, can be utilized in clinical trials with protocols that incorporate baseline MR imaging to significantly increase statistical power to detect treatment effects. Akin to the historical control paradigm, we propose to utilize a radiomic prediction model to generate a pseudo-control sample for each individual in the trial of interest. Because the variability of expected outcome across patients can mask our ability to detect treatment effects, we can increase the power to detect a treatment effect in a clinical trial by reducing that variability through using radiomic predictors as surrogates. We illustrate this method with simulations based on data from two cohorts in different neurologic diseases, Alzheimer's disease and glioblastoma multiforme. We present sample size requirements across a range of effect sizes using conventional analysis and models that include a radiomic predictor. For our Alzheimer's disease cohort, at an effect size of 0.35, total sample size requirements for 80% power declined from 246 to 212 for the endpoint cognitive decline. For our glioblastoma multiforme cohort, at an effect size of 1.65 with the endpoint survival time, total sample size requirements declined from 128 to 74. This methodology can decrease the required sample sizes by as much as 50%, depending on the strength of the radiomic predictor. The power of this method grows with increased accuracy of radiomic prediction, and furthermore, this method is most helpful when treatment effect sizes are small. Neuroimaging biomarkers are a powerful and increasingly common suite of tools that are, in many cases, highly predictive of disease outcomes. Here, we explore the possibility of using MRI-based radiomic biomarkers for the purpose of improving statistical power in clinical trials in the contexts of brain cancer and prodromal Alzheimer's disease. These methods can be applied to a broad range of neurologic diseases using a broad range of predictors of outcome to make clinical trials more efficient.

Predicting Amyloid-β Levels in Amnestic Mild Cognitive Impairment Using Machine Learning Techniques

BACKGROUND

Amyloid-β positivity (Aβ+) based on PET imaging is part of the enrollment criteria for many of the clinical trials of Alzheimer's disease (AD), particularly in trials for amyloid-targeted therapy. Predicting Aβ positivity prior to PET imaging can decrease unnecessary patient burden and costs of running these trials.

OBJECTIVE

The aim of this study was to evaluate the performance of a machine learning model in estimating the individual risk of Aβ+ based on gold-standard of PET imaging.

METHODS

We used data from an amnestic mild cognitive impairment (aMCI) subset of the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort to develop and validate the models. The predictors of Aβ status included demographic and ApoE4 status in all models plus a combination of neuropsychological tests (NP), MRI volumetrics, and cerebrospinal fluid (CSF) biomarkers.

RESULTS

The models that included NP and MRI measures separately showed an area under the receiver operating characteristics (AUC) of 0.74 and 0.72, respectively. However, using NP and MRI measures jointly in the model did not improve prediction. The models including CSF biomarkers significantly outperformed other models with AUCs between 0.89 to 0.92.

CONCLUSIONS

Predictive models can be effectively used to identify persons with aMCI likely to be amyloid positive on a subsequent PET scan.

Radiofrequency ablation and balloon kyphoplasty for palliation of painful spinal metastases

PURPOSE

This study was designed with an aim to assess the safety and early postoperative outcomes of the combined Radiofrequency ablation (RFA) and Balloon Kyphoplasty (BKP) used for the treatment of painful neoplastic spinal lesions palliatively.

PATIENTS AND METHODS

Between December 2015 and December 2018, 60 patients (35 men and 25 women) with spinal metastases were operated using RFA and BKP at our institution. Transpedicular biopsy was performed in all cases. Patients' demographics, lesion characteristics, concurrent palliative therapies and complications were recorded. All patients were clinically (Pain score VAS 0-10) and radiologically evaluated pre- and postoperatively. Retrospective analysis of data for this cohort was performed.

RESULTS

Seventy-five painful spinal metastases (46 in the lumbar spine and 29 in the thoracic region) in 60 patients were operated [transpedicular RFA alone in 5 lesions, and in combination with BKP in 70 lesions (93%)]. The mean pre-procedure and post-procedure VAS for back pain was 7.2/10 and 2.7/10, respectively (p value = 0.0001). No neurological complications related to RFA were found and no cement extravasation into the spinal canal was observed. In two patients, asymptomatic leaks into the needle track, in two patients into draining veins and in one patient into the disk space were detected.

CONCLUSION

Combined RFA and BKP appears to be a safe, practical, effective and reproducible palliative treatment for painful spinal osteolytic metastasis. In carefully indicated cases, it relieves pain and maintains stability in a minimal invasive way without adding significant surgical trauma or complications.

White matter hyperintensities and cognition across different Alzheimer's biomarker profiles

BACKGROUND/OBJECTIVES

To examine the association between white matter hyperintensities (WMH) and cognitive domains such as memory and executive function (EF) across different clinical and biomarker categories of Alzheimer's disease (AD).

DESIGN

Cross-sectional study.

SETTING

Alzheimer's Disease Neuroimaging Initiative.

PARTICIPANTS

A total of 216 cognitively normal (CN) participants and 407 participants with mild cognitive impairment (MCI) from the Alzheimer's Disease Neuroimaging Initiative (ADNI) at baseline.

MEASUREMENTS

Based on the 2018 research framework, participants were classified using AT(N) (amyloid-β deposition [A], pathologic tau [T], and neurodegeneration [(N)]) biomarkers into one of three categories: biomarker negative [A - T- (N)-], amyloid negative but other biomarker positive [A - T ± (N)+ or A - T + (N)±] or amyloid positive [A + T ± (N)±]. Linear regression models were then used to examine the association between WMH and memory composite scores and EF composite scores.

RESULTS

Higher WMH burden was associated with worse EF in both CN and MCI subgroups while a significant association between WMH and memory was only found in the MCI subgroup. Furthermore, WMH was associated with EF in the group with A - T ± (N)+ or A - T + (N)± biomarker category, but not for A - T - (N)- (normal biomarker) and A + T ± (N) ± (AD pathology). The association between higher WMH and worse memory was independent of amyloid levels in individuals with MCI with evidence of AD pathology.

CONCLUSION

Vascular disease, as indexed by WMH, independent of AD pathology affects cognitive function in both CN and MCI subgroups. Future studies using the AT(N) research framework should consider white matter lesions as a key biomarker contributing to the clinical presentation of AD.

Detecting biological heterogeneity patterns in ADNI amnestic mild cognitive impairment based on volumetric MRI

There is substantial biological heterogeneity among older adults with amnestic mild cognitive impairment (aMCI). We hypothesized that this heterogeneity can be detected solely based on volumetric MRI measures, which potentially have clinical implications and can improve our ability to predict clinical outcomes. We used latent class analysis (LCA) to identify subgroups among persons with aMCI (n = 696) enrolled in the Alzheimer's Disease Neuroimaging Initiative (ADNI), based on baseline volumetric MRI measures. We used volumetric measures of 10 different brain regions. The subgroups were validated with respect to demographics, cognitive performance, and other AD biomarkers. The subgroups were compared with each other and with normal and Alzheimer's disease (AD) groups with respect to baseline cognitive function and longitudinal rate of conversion. Four aMCI subgroups emerged with distinct MRI patterns: The first subgroup (n = 404), most similar to normal controls in volumetric characteristics and cognitive function, had the lowest incidence of AD. The second subgroup (n = 230) had the most similar MRI profile to early AD, along with poor performance in memory and executive function domains. The third subgroup (n = 36) had the highest global atrophy, very small hippocampus and worst overall cognitive performance. The fourth subgroup (n = 26) had the least amount of atrophy, however still had poor cognitive function specifically in in the executive function domain. Individuals with aMCI who were clinically categorized within one group other showed substantial heterogeneity based on MRI volumetric measures.

Machine Learning Predictive Models Can Improve Efficacy of Clinical Trials for Alzheimer's Disease

BACKGROUND

The ideal participants for Alzheimer's disease (AD) clinical trials would show cognitive decline in the absence of treatment (i.e., placebo arm) and also would be responsive to the therapeutic intervention being studied (i.e., drug arm). One strategy to boost the power of trials is to enroll individuals who are more likely to progress targeted using data-driven predictive models.

OBJECTIVE

To investigate if machine learning (ML) models can effectively predict clinical disease progression (cognitive decline) in mild-to-moderate AD patients during the timeframe of a phase III clinical trial.

METHODS

Data from 202 participants with a diagnosis of AD at baseline from the Alzheimer's Disease Neuroimaging Initiative (ADNI) was used to train ML classifiers that can differentiate between individuals who had declining cognitive function (DC) and individuals with stable cognitive function (SC). DC was defined as any downward change in the Alzheimer's Disease Assessment Scale cognitive subscale (ADAS-cog) score over 12 months of follow-up. SC was defined by the absence of decline in ADAS-cog. Trained models were applied to data from 77 participants from the placebo arm of the phase III trial of Semagacestat (LFAN study) to identify subgroups of SC versus DC.

RESULTS

Only 74.8% of ADNI participants and 63.6% of LFAN participants had cognitive decline after one year of follow up. K-nearest neighbors (kNN) classifier had an accuracy of 68.3%, sensitivity of 80.1%, and specificity of 33.3% for identifying decliners in ADNI (training sample). In LFAN (validation sample), the model showed an overall accuracy of 61.3%, sensitivity of 65.5%, and specificity of 47.0% in identifying decliners at the 12 months of follow-up. The model had a positive predictive value of 80.8%, which was 17.2% more than the base prevalence of decliners.

CONCLUSIONS

Machine learning predictive models can be effectively used to boost the power of clinical trials by reducing the sample size.

Effect of Mentally Challenging Occupations on Incident Dementia Differs Between African Americans and Non-Hispanic Whites

BACKGROUND

Engaging in mentally challenging activities may protect against dementia in late life. However, little is known whether the association between mentally challenging activities and dementia risk varies with race/ethnicity.

OBJECTIVE

The current study investigates whether having jobs with higher mental stimulation is differentially associated with a decreased risk of dementia between African Americans (AAs) and non-Hispanic Whites (nHWs).

METHODS

The sample consisted of 1,079 individuals (66% nHWs, 28% AAs; age = 78.6±5.3) from the longitudinal Einstein Aging Study. Occupation information of each participant was collected retrospectively at baseline and was linked to the substantive complexity of work score from the Dictionary of Occupational Titles. Cox proportional hazards models were used to evaluate the associations of occupational complexity with risk of dementia.

RESULTS

Individuals whose jobs had moderate-to-high levels of complexity, compared to those with the lowest complexity, were at modestly decreased risk for incident dementia. When stratified by race, moderate-to-high levels of occupational complexity were significantly associated with lower risk of developing dementia for AAs (HR = 0.35). When risk of dementia was evaluated based on the combinations of race×occupational complexity, AAs with lowest occupational complexity showed the highest risk of developing dementia, while other combinations exhibited lower risk of developing dementia (HRs = 0.36~0.43).

CONCLUSION

Our results suggest that moderate-to-high levels of complexity at work are associated with a decreased risk of incident dementia in AAs. Understanding the differential effects of mentally challenging occupations across race/ethnicity may suggest important intervention strategies that could mitigate racial disparities in dementia rates.

Class-Specific Incidence of All-Cause Dementia and Alzheimer's Disease: A Latent Class Approach

Identifying preclinical Alzheimer's disease (AD) is an important step toward developing approaches to early treatment and dementia prevention. We applied latent class analysis (LCA) to 10 baseline neuropsychological assessments for 1,345 participants from Einstein Aging Study. Time-to-event models for all-cause dementia and AD were run examining events in 4-year intervals. Five classes were identified: Mixed-Domain Impairment (n = 107), Memory-Specific Impairment (n = 457), Average (n = 539), Frontal Impairment (n = 118), and Superior Cognition (n = 124). Compared to the Average class, the Mixed-Domain Impairment and Memory-Specific Impairment classes were at higher risk of incident all-cause dementia and AD in the first 4 years from baseline, while the Frontal Impairment class was associated with higher risk between 4 and 8 years of follow-up. LCA identified classes which differ in cross-sectional cognitive patterns and in risk of dementia over specific follow-up intervals.

The relationship between hippocampal volume, chronic pain, and depressive symptoms in older adults

We aimed to test the hypothesis that the effect of chronic pain on depressive symptoms is mediated through hippocampal volume (HV). Participants were 131 non-demented adults over the age of 70 years from the Einstein Aging Study. Smaller right and left HV were both associated with higher depressive symptoms, but only smaller right HV was associated with chronic pain. In mediation models, right HV was a significant mediator for the effect of chronic pain on depression. Our findings suggest presence of a shared brain substrates between chronic pain and depression as reflected by right HV.

Depressive Symptoms Predict Incident Dementia in a Community Sample of Older Adults: Results From the Einstein Aging Study

BACKGROUND:

There is increasing evidence that depressive symptoms are associated with increased risk of cognitive impairment and dementia in older adults. In current study, we aimed to investigate the effect of depressive symptoms on incident Alzheimer disease and all-cause dementia in a community sample of older adults.

METHODS:

Participants were 1219 older adults from the Einstein Aging Study, a longitudinal cohort study of community-dwelling older adults in Bronx County, New York. The Geriatric Depression Scale (GDS, 15-item) was used as a measure of depressive symptoms. The primary outcome was incident dementia diagnosed using the Diagnostic and Statistical Manual, Fourth Edition, criteria. Cox proportional hazard models were used to estimate the risk of incident dementia as a function of GDS score for the whole population and also for 2 different time intervals, <3 years and ≥3 years after baseline assessment.

RESULTS:

Among participants, 132 individuals developed dementia over an average 4.5 years (standard deviation [SD] = 3.5) of follow-up. Participants had an average age of 78.3 (SD = 5.4) at baseline, and 62% were women. Among all participants, after controlling for demographic variables and medical comorbidities, a 1-point increase in GDS was associated with higher incidence of dementia (hazard ratio [HR] = 1.11, P = .007). After up to 3 years of follow-up, depressive symptoms were not significantly associated with dementia incidence (HR = 1.09; P = .070). However, after more than 3 years, GDS score was a significant predictor of incident dementia (HR = 1.13, P = .028).

CONCLUSIONS:

Our results suggest that depressive symptoms are associated with an increased risk of incident dementia in older adults.

Optimizing Machine Learning Methods to Improve Predictive Models of Alzheimer's Disease

BACKGROUND

Predicting clinical course of cognitive decline can boost clinical trials' power and improve our clinical decision-making. Machine learning (ML) algorithms are specifically designed for the purpose of prediction; however. identifying optimal features or algorithms is still a challenge.

OBJECTIVE

To investigate the accuracy of different ML methods and different features to classify cognitively normal (CN) individuals from Alzheimer's disease (AD) and to predict longitudinal outcome in participants with mild cognitive impairment (MCI).

METHODS

A total of 1,329 participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI) were included: 424 CN, 656 MCI, and 249 AD individuals. Four feature-sets at baseline (hippocampal volume and volume of 47 cortical and subcortical regions with and without demographics and APOE4) and six machine learning methods (decision trees, support vector machines, K-nearest neighbor, ensemble linear discriminant, boosted trees, and random forests) were used to classify participants with normal cognition from participants with AD. Subsequently the model with best classification performance was used for predicting clinical outcome of MCI participants.

RESULTS

Ensemble linear discriminant models using demographics and all volumetric magnetic resonance imaging measures as feature-set showed the best performance in classification of CN versus AD participants (accuracy = 92.8%, sensitivity = 95.8%, and specificity = 88.3%). Prediction accuracy of future conversion from MCI to AD for this ensemble linear discriminant at 6, 12, 24, 36, and 48 months was 63.8% (sensitivity = 74.4, specificity = 63.1), 68.9% (sensitivity = 75.9, specificity = 67.8), 74.9% (sensitivity = 71.5, specificity = 76.3), 75.3%, (sensitivity = 65.2, specificity = 79.7), and 77.0% (sensitivity = 59.6, specificity = 86.1), respectively.

CONCLUSIONS

Machine learning models trained for classification of CN versus AD can improve our prediction ability of MCI conversion to AD.

Subtypes Based on Neuropsychological Performance Predict Incident Dementia: Findings from the Rush Memory and Aging Project

BACKGROUND

In a previous report, we used latent class analysis (LCA) to identify natural subgroups of older adults in the Einstein Aging Study (EAS) based on neuropsychological performance. These subgroups differed in demographics, genetic profile, and prognosis. Herein, we assess the generalizability of these findings to an independent sample, the Rush Memory and Aging Project (MAP), which used an overlapping, but distinct neuropsychological battery.

OBJECTIVE

Our aim was to identify the association of natural subgroups based on neuropsychological performance in the MAP cohort with incident dementia and compare them with the associations identified in the EAS.

METHODS

MAP is a community-dwelling cohort of older adults living in the northeastern Illinois, Chicago. Latent class models were applied to baseline scores of 10 neuropsychological measures across 1,662 dementia-free MAP participants. Results were compared to prior findings from the EAS.

RESULTS

LCA resulted in a 5-class model: Mixed-Domain Impairment (n = 71, 4.3%), Memory-specific-Impairment (n = 274, 16.5%), Average (n = 767, 46.1%), Frontal Impairment (n = 222, 13.4%), and a class of Superior Cognition (n = 328, 19.7%). Similar to the EAS, the Mixed-Domain Impairment, the Memory-Specific Impairment, and the Frontal Impairment classes had higher risk of incident Alzheimer's disease when compared to the Average class. By contrast, the Superior Cognition had a lower risk of Alzheimer's disease when compared to the Average class.

CONCLUSIONS

Natural cognitive subgroups in MAP are similar to those identified in EAS. These similarities, despite study differences in geography, sampling strategy, and cognitive tests, suggest that LCA is capable of identifying classes that are not limited to a single sample or a set of cognitive tests.

The Temporal Relationship between Pain Intensity and Pain Interference and Incident Dementia

BACKGROUND

Chronic pain is common among older adults and is associated with cognitive dysfunction based on cross-sectional studies. However, the longitudinal association between chronic pain and incident dementia in community-based samples is unknown.

OBJECTIVE

We aimed to evaluate the association of pain intensity and pain interference with incident dementia in a community-based sample of older adults.

METHODS

Participants were 1,114 individuals 70 years of age or older from Einstein Aging Study (EAS), a longitudinal cohort study of community-dwelling older adults in the Bronx County, NY. The primary outcome measure was incident dementia, diagnosed using DSM-IV criteria. Pain intensity and interference in the month prior to first annual visit were measured using items from the SF-36 questionnaire. Pain intensity and pain interference were assessed as predictors of time to incident dementia using Cox proportionate hazards models while controlling for potential confounders.

RESULTS

Among participants, 114 individuals developed dementia over an average 4.4 years (SD=3.1) of follow-up. Models showed that pain intensity had no significant effect on time to developing dementia, whereas higher levels of pain interference were associated with a higher risk of dementia. In the model that included both pain intensity and interference as predictors of incident dementia, pain interference had a significant effect on incident dementia, and pain intensity remained non-significant.

CONCLUSION

As a potential remediable risk factor, the mechanisms linking pain interference to cognitive decline merit further exploration.

Kyphoplasty for osteoporotic vertebral fractures with posterior wall injury

BACKGROUND CONTEXT

Cement augmentation techniques are standard treatments for osteoporotic vertebral fractures. Compared with vertebroplasty, kyphoplasty is associated with lower rates of cement leak and better deformity correction; however, posterior wall fractures are relative, but not absolute; contraindications for both techniques and hence treatment practices vary among spine centers.

PURPOSE

The primary aim of this study was to assess our center's incidence of posterior cement leakage in osteoporotic vertebral fractures with posterior wall injury treated by balloon kyphoplasty (BKP). Secondarily, physiological results, pain relief, complication rates, and non-posterior cement leakage were also evaluated.

STUDY DESIGN

This is a prospective cohort study done in a high-volume spine center in Germany.

PATIENT SAMPLE

Eighty-two patients with 98 osteoporotic vertebral fractures with posterior wall cortical injury were studied from 2012 to 2016.

OUTCOME MEASURES

The following were the outcome measures: (1) physiological measures: standing plain x-rays (anteroposterior and lateral views), with the following parameters evaluated: cement leak behind the posterior vertebral body border, Cobb angle for local sagittal deformity, vertebral wedge angle, and anterior vertebral height; (2) cement volume injected in each vertebra; and (3) self-report measures: visual analog scale (VAS).

METHODS

All patients underwent BKP using a bipedicular approach. Preoperative clinical and neurologic evaluations were done. Radiological evaluations included plain X-ray images, computed tomography scans and magnetic resonance imaging. The average follow-up period was 18 months.

RESULTS

No cement leakage into the spinal canal occurred in any of the patients. Asymptomatic leakage into other sites was seen in 22 vertebrae (22.45%). There was significant improvement in the Cobb angle, the vertebral wedge angle, and the anterior vertebral height in all cases. The mean preoperative VAS was 8.1, and this improved to 2.3 on the third postoperative day.

CONCLUSION

Balloon kyphoplasty is a viable option for the treatment of osteoporotic vertebral fractures even with posterior wall involvement.

Extraforaminal microscopic assisted percutaneous nucleotomy for foraminal and extraforaminal lumbar disc herniations

BACKGROUND

Foraminal and extraforaminal lumbar disc herniations are uncommon. The main presentation is radicular pain related to the exiting nerve root at the affected level. Different approaches for surgical intervention have been described.

PURPOSE

This study aimed to evaluate the clinical outcome, complications recurrence, and reoperation rate of extraforaminal microscopic-assisted percutaneous nucleotomy, with literature review focusing on complications and recurrence rate.

STUDY DESIGN

This is a prospective cohort study done in a high-flow spine center in Germany.

PATIENT SAMPLE

Between October 2012 and October 2015, 76 patients (35 women and 41 men) with foraminal or extraforaminal lumbar disc prolapse were operated on.

OUTCOME MEASURES

The following were the outcome measures: (1) self-report measures: Visual Analogue Scale (VAS) for leg pain and back pain; (2) physiological measures: standing plain X-rays (anterioposterior, lateral, and dynamic views); and (3) functional measures: Oswestry Disability Index (ODI) (validated German version) and Odom's criteria.

METHODS

All patients were operated upon with trans-tubular extraforaminal microscopic-assisted percutaneous nucleotomy (EF-MAPN) technique. Preoperative clinical and neurologic evaluations were done. The mean follow-up period was 38 months (range 12-54). The study has not received funding for research from any organization. All authors do not have any conflict of interest.

RESULTS

The mean age was 54 years. The most commonly affected level was L4/L5 (34 patients). The mean preoperative VAS for leg pain was 7.6 (3-10), which improved to 1.4 (0-4) postoperatively. The average operative time was 57.5 minutes. There were no intraoperative complications. One patient had temporary postoperative quadriceps weakness (L4 radiculopathy) that was completely improved at 3 months' follow-up. Another patient had deep venous thrombosis after discharge. Two patients had recurrences that necessitated another operation within the first 6 months postoperatively. Both were followed up for 1 year without a second recurrence.

CONCLUSION

Trans-tubular percutaneous extraforaminal microscopic-assisted nucleotomy is effective for foraminal and extraforaminal disc herniations. It is a muscle-splitting minimally invasive approach with minimal morbidity. Complications, recurrence, and reoperation rate are not different compared with microsurgical open or endoscopic techniques.

Odontoid-sparing transnasal approach for drainage of craniocervical epidural abscess; a novel technique and review of the literature

BACKGROUND CONTEXT

Surgical approaches to the craniovertebral junction (CVJ) are challenging. Available approaches include posterior, transoral, endonasal, and anterior extended retropharyngeal approach. Resection of the odontoid process is necessary to gain access to the pathology posterior to it. The resultant cranio-atlanto-axial instability usually necessitates subsequent posterior stabilization.

PURPOSE

To describe a new odontoid-sparing approach to the spinal canal at the CVJ. This dens-sparing approach preserves occipito-atlanto-axial stability and avoids the need for occipitocervical stabilization that adds to the extent of surgery and its associated morbidity and mortality.

STUDY DESIGN

Describing a novel technique and reporting two cases.

PATIENT SAMPLE

Two patients that presented with infection at the CVJ with a retro-odontoid epidural abscess were operated on.

OUTCOME MEASURES

Self-reported measures: visual analog scale for neck pain. Physiologic measures: plain x-rays (anteroposterior and lateral views), magnetic resonance imaging with contrast, computed tomography scan, C-reactive protein, and leukocytic count. Functional measures: dynamic flexion-extension views of the cervical spine.

METHODS

Two patients were operated on using a combined endoscopic transnasal-transoral approach for drainage of a retro-odontoid epidural abscess and debridement without dens resection. A 4-mm, 30-degree rigid endoscope was used. Preoperative clinical and neurologic status was evaluated. The follow-up period was 12 months. The study received no funding from any organization. None of the authors has any relevant financial disclosures or conflict of interest.

RESULTS

Both patients improved clinically after the endonasal transoral abscess drainage. Follow-up contrast magnetic resonance imaging showed complete resolution of the abscess after 3 weeks. Culture-sensitivity tests were positive for Staphylococcus aureus in one patient. Antibiotic therapy with clindamycin and flucloxacillin was continued for 12 weeks postoperatively. There were no intraoperative or postoperative complications. There was no need for posterior occipitocervical stabilization in both cases.

CONCLUSION

This represents the first clinical report of accessing the spinal canal at the CVJ without resection of the odontoid or the anterior arch of the atlas. The addition of endoscopic-assisted supra-dental approach to the transoral one improved visibility, and allowed access to the most cranial part of spinal canal without the need for dens resection, a procedure that significantly compromises C0-1-2 stability necessitating stabilization. This novel odontoid-sparing approach showed a favorable outcome in our first two cases with retro-odontoid abscess; however, it would likely pose a high risk in other pathologies including tumors.

Transcranial Doppler and Lower Extremity Function in Older Adults: Einstein Aging Study

OBJECTIVES

To determine whether transcranial Doppler ultrasound (TCD) measures of mean blood flow velocity (MBFV) in the major cerebral arteries are associated with measures of lower extremity function in community-dwelling older adults.

DESIGN

Cross-sectional study.

SETTING

Community sample.

PARTICIPANTS

Individuals aged 70 and older (mean 79.5, 54% female) without dementia participating in the Einstein Aging Study (N = 200).

MEASUREMENTS

All participants underwent TCD assessments and tests of lower extremity function at an annual clinic visit. Average MBFV for anterior (left and right anterior and middle cerebral arteries (MCAs)) and posterior (vertebral (VA) and basilar (BA) artery) circulation was measured using a standardized TCD protocol. Lower extremity function was characterized according to gait speed (cm/s) measured using an instrumented walkway, balance according to unipedal stance time (UPST, seconds), and lower extremity strength according to timed repeated chair rise (seconds).

RESULTS

Multiple regression models adjusted for age, sex, race, education, and medical comorbidities showed that lower MBFV in the MCA was associated with slower gait speed and chair rise time but not with UPST. Ordinal regression models showed that lower MBFV in the VA and BA is associated with shorter UPST.

CONCLUSION

Low MBFV in the anterior and posterior cerebral circulation was associated with worse lower extremity function and balance in older adults. This might be indicative of the importance of age-related changes in cerebral hemodynamics in the function of brain regions involved in specific aspects of physical performance.

The association of visual memory with hippocampal volume

Background

In this study we investigated the role of hippocampal volume (HV) in visual memory.

Methods

Participants were a subsample of older adults (> = 70 years) from the Einstein Aging Study. Visual performance was measured using the Complex Figure (CF) copy and delayed recall tasks from the Repeatable Battery for the Assessment of Neuropsychological Status. Linear regressions were fitted to study associations between HV and visual tasks.

Results

Participants’ (n = 113, mean age = 78.9 years) average scores on the CF copy and delayed recall were 17.4 and 11.6, respectively. CF delayed recall was associated with total (β = .031, p = 0.001) and left (β = 0.031, p = 0.001) and right HVs (β = 0.24, p = 0.012). CF delayed recall remained significantly associated with left HV even after we also included right HV (β = 0.27, p = 0.025) and the CF copy task (β = 0.30, p = 0.009) in the model. CF copy did not show any significant associations with HV.

Conclusion

Our results suggest that left HV contributes in retrieval of visual memory in older adults.