Participants
A total of 61 participants were enrolled in this study, including 40 patients with LOD and 21 healthy volunteers without a history of psychiatric disease. Forty patients with LOD (5 males and 35 females) were recruited from the outpatient clinic of the Depression Center at Samsung Medical Center between May 2016 and January 2018. All patients were clinically referred. They maintained their psychotropic medication, including antidepressants, mood stabilizers, benzodiazepines, and zolpidem, during the study.
Inclusion criteria for patients were: being 65 years of age or older, experiencing a current unipolar major depressive episode verified by the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria for major depressive disorder (MDD), with onset after 60 years of age (30). The baseline minimum 17-item Hamilton Scale for Depression (HAM-D) (31) score required for enrollment was 16. Exclusion criteria were: presence of psychotic disorders (e.g., schizophrenia or delusional disorder), bipolar affective disorder, neurological illnesses including significant cognitive impairment and Parkinson’s disease, mental retardation, significant medical conditions, epilepsy, history of alcohol or drug dependence, personality disorders, brain injury, and individuals currently taking or requiring antipsychotics, choline esterase inhibitors, memantine, Ginkgo biloba, or acetyl-L-carnitine.
The control group comprised 21 healthy volunteers aged 65 years or older. They were recruited through advertisement. Volunteers with a positive family history of mood disorders were excluded. Screening assessments were conducted to rule out individuals with a history of current or past depressive episodes.
The study protocol was approved by the Ethics Review Board of Samsung Medical Center, Seoul, Korea. All research procedures were performed in accordance with relevant guidelines. Signed informed consent was obtained from all participants.
Clinical Evaluation
The Korean version of the Mini-Mental Status Examination (K-MMSE) (32) and the Korean version of the Montreal Cognitive Assessment (MoCA-K) (33) were applied to test eligibility of each participant for this study. To assess ‘cognitive normality’, clinical psychiatrists interviewed participants regarding their daily life activities following K-MMSE and MoCA-K assessments, considering the impact of depression on cognition. Psychiatrists with more than three years of clinical experience evaluated past psychiatric and physical history of participants to confirm their eligibility for this study.
After enrollment, participants were evaluated twice: at 1 month and 3 months after the initial assessment. During these follow-ups, a professional clinical psychologist at the Depression Center performed the Mini International Neuropsychiatric Interview (MINI) (34) to evaluate psychiatric comorbidities. The psychologist also assessed participants’ severity of depression and anxiety using the 17-item Hamilton Depression Rating Scale (HAM-D) and the Hamilton Anxiety Rating Scale (HAM-A) (35). In addition to the K-MMSE and MoCA-K, the Clinical Global Impression Scale (CGI-S) and the Seoul Neuropsychological Screening Battery (SNSB-II) (36) were used to assess the overall state and cognitive function of participants. The Massachusetts General Hospital Cognitive and Physical Functioning Questionnaire (MGH-CPFQ) was used to assess clinically relevant cognitive and physical symptoms associated with depression as a subjective cognitive domain (37).
The LOD group was treated according to evidence-based treatment guidelines for Korean MDD, initiating antidepressant treatment at the first visit. If augmentation was needed, mood stabilizers, benzodiazepine, and zolpidem were allowed to be used. However, antipsychotics, choline esterase inhibitors, memantine, Ginkgo biloba, and acetyl L-carnitine were not allowed.
One month after the initial visit, all participants were evaluated using the HAM-D and CGI-S. After 3 months, they were re-evaluated with the HAM-D, CGI-S, SNSB-II, and MoCA-K.
MRI Data Acquisition
Structural images of brains of all 61 participants were acquired at the Samsung Medical Center using a 3.0T Achieva MRI scanner (Philips, Amsterdam, The Netherlands) within one month of the second visit. T1-weighted MRI data were recorded using the following imaging parameters: sagittal slice thickness of 1 mm, contiguous slices with 50% overlap, no gap, repetition time (TR) of 9.9 ms, echo time (TE) of 4.6 ms, flip angle of 8°, and matrix size of 240 × 240 pixels. Images were reconstructed to 480 × 480 pixels over a 240 mm field of view.
PET Imaging Analyses
Participants underwent 18F-florbetaben PET scanning at Samsung Medical Center using a Discovery STe PET/Computed tomography (CT) scanner (GE Medical Systems, Milwaukee, WI, USA) (47 slices of 3.3 mm in thickness, spanning the entire brain). For attenuation correction, CT images were acquired using a 16-slice helical CT (140 KeV, 80 mA; 3.75 mm section width). Ninety minutes after injecting 300 MBq ± 20% 18F-florbetaben, a 20-minute emission PET scan in a dynamic mode (consisting of 4 × 5 min frames) was performed. Using the ordered-subsets expectation maximization algorithm (iteration = 4 and subset = 20), 3-dimensional PET images were reconstructed in a 128 × 128 × 48 matrix with a 2 × 2 × 3.27 mm voxel size.
Image Processing Steps
T1-weighted MR images were processed to obtain anatomical parcellations based on the Desikan-Killiany atlas (38) using the FreeSurfer software package (version 5.1.0, http://surfer.nmr.mgh.harvard.edu). Each florbetaben image was co-registered to the T1 image through affine coregistration (FSL-FLIRT) and normalized by the mean value of cerebellar gray matter to measure standardized uptake value ratios (SUVRs) in 61 participants. SUVR values were then measured as means of each Desikan-based region of interest (ROI). In the present study, amyloid deposition data were analyzed using ROI and metaROI approaches (39). Group comparisons utilized both ROI and metaROI approaches. Based on group comparison results, correlational analysis also used metaROI approach. This method comprised 13 metaROIs (medial temporal lobe, anterior cingulate, middle frontal lobe, medial occipital lobe, lateral occipital lobe, inferior parietal lobe, posterior cingulate, orbitofrontal cortex, lateral temporal lobe, inferior frontal lobe, precuneus, superior frontal lobe, and superior parietal lobe). Applying bilateral metaROIs could reduce the number of estimated correlations.
Statistical Analysis
SPSS Statistics 24 (SPSS Inc., Chicago, IL, USA) was used for all statistical analyses, with a significance level set at p < 0.05. The false discovery rate (FDR) procedure was applied to multiple comparison.
Clinical and demographic profiles of the 61 participants are presented as either categorical or continuous variables, as appropriate. Categorical variables are presented as frequencies and proportions, while continuous variables are presented as mean ± standard deviation (SD). Student’s t-test, Mann-Whitney test, or Fisher’s exact test was applied after evaluating the normality of distributions. To compare neurocognitive functions between the two groups, SNSB-II and MoCA-K results were subjected to multiple analyses of covariance, adjusting for age, sex, and years of education.
We used permutation-based analysis of covariance (ANCOVA) to compare amyloid depositions in cortical and subcortical structures between the LOD group and the healthy control (HC) group, controlling for the effects of age, sex, and education. Significance level of the group difference was estimated as the fraction of permutations whose F values were not less than those from the original dataset. We used 5,000 permutations. We performed correlation tests between amyloid deposition in each brain structure and depressive symptom severity as well as each item of the depression scale (HAM-D) in LOD individuals. Since amyloid deposition often does not follow a normal distribution, we used Spearman partial correlation to control for effects of age, sex, and years of education. Additionally, we evaluated the correlation of amyloid deposition with baseline neurocognitive assessments and cognitive changes following antidepressant treatment after adjusting for the severity of depression, age, sex, and years of education.