Increased ASL-CBF in the right amygdala predicts the first onset of depression in healthy young first-degree relatives of patients with major depression

Abstract

Healthy first-degree relatives of patients with major depression are at an elevated risk of developing depression, and regional cerebral blood flow (CBF) alterations are observed in patients with depression. Therefore, in a 33-month follow-up study, we used arterial spin labeling-magnetic resonance imaging (ASL-MRI) to investigate quantitative CBF before and after the diagnosis of depression in healthy young adults with and without first-degree relatives with major depression (FH + and FH−, respectively). In cross-sectional and longitudinal CBF comparisons, CBF in the right amygdala was increased or decreased. Additionally, a significant correlation was observed between the altered CBF in the right amygdala and the scores on the 17-item Hamilton Depression Rating Scale (HDRS) in the FH + group. Furthermore, logistic regression and receiver operating characteristic curve analyses showed that increased CBF in the right amygdala at baseline predicted the subsequent onset of depression in the FH + group. Our results suggest that among healthy young adults with a familial risk of depression, those who exhibit increased CBF in the amygdala are susceptible to developing this disease.

Keywords

Amygdala arterial spin labeling cerebral blood flow depression family history prediction

Introduction

Depression, which accounts for almost half of all disability-adjusted life years lost to disease worldwide, is one of the most prevalent psychiatric illnesses and the 11th leading component of the global burden of disease, with an estimated global prevalence of 4.7%.^1–3 Depression is difficult to treat once it has developed, and its etiology remains poorly understood. Depressive disorder is a multifactorial disease, i.e. it might be caused by a wide range of genetic and environmental factors.⁴ In addition, family history is a major risk factor associated with the development of depression. According to Weissman et al., first-degree relatives of people with depression have a two- to four-fold increased risk of developing the disease compared with first-degree relatives of healthy control individuals.^5–7 Thus, given the potential severity of the depression and the high risk associated with family history, it is important to detect early predictors of the onset of depression among healthy first-degree relatives of patients with major depressive disorder (MDD).

Several studies have addressed the methods used to predict the onset of depression. For example, Foland-Ross et al.⁸ reported that cortical gray matter structure might predict the subsequent onset of depression during adolescence. Moreover, in a longitudinal community-based sample, Pan et al.⁹ indicated that aberrant left ventral striatum functional connectivity specifically predicts the onset of adolescent depressive disorder. Furthermore, according to LeMoult et al.,¹⁰ the interaction of elevated diurnal cortisol and negative life events predicts the first onset of depression in young girls. However, few studies have focused on predicting the onset of depression in participants with a family history of MDD.

Previous studies have found that healthy participants with a familial risk of depression show a range of abnormalities, and various neuroimaging studies have reported both anatomical and functional abnormalities. In terms of structural abnormalities, a familial risk of depression in healthy participants is associated with a thinned cerebral cortex and altered volumes in certain brain regions, e.g. an increased amygdala volume.^11–17 Functional findings, as noted via blood-oxygen-level changes, suggest that these high-risk individuals have an altered response pattern to positive and negative stimuli in certain brain areas including the amygdala, hippocampus, putamen, caudate, orbitofrontal cortex and cingulate cortex.^18,19 Furthermore, there are functional alterations in several regions related to emotional disturbances, particularly involving the regions of the prefrontal cortex and closely related limbic structures.^20–23 These functional changes were investigated using blood oxygen level-dependent-functional MRI (BOLD-fMRI), which provides a relative, rather than absolute, measure of neural activity.^24,25 By contrast, cerebral blood flow (CBF), an absolute measure of neural activity, reflects the energy demand in the brain more directly and effectively than fMRI and structural MRI indicators. In parallel, a meta-analysis revealed a large body of literature involving alterations in the CBF of people with depression.²⁶ To our knowledge, however, few studies have used CBF to examine the risk of depression among people with a family history of depression.

Therefore, this study employed arterial spin labeling (ASL) to test our hypotheses that (1) altered CBF would exist in one or more brain regions of healthy, never-depressed young adults with a family history of depression and susceptibility to depression, with their developmental stage marking the beginning of a peak period of risk for the emergence of major depression;²⁷ and (2) the specific alterations in CBF would be useful for predicting the onset of depression in these high-risk participants.

Materials and methods

The study was performed in accordance with the latest version of the Declaration of Helsinki and approved by the Ethical Committee of the Affiliated Hospital of Jiangsu University. Written informed consent was obtained from all participants after they had been given a detailed description of the study.

Participants and procedure

The overall study design is shown in Figure 1. From all young adult college students attending Jiangsu University (age range 18–22 years), 200 (females, 120; males, 80) healthy never-depressed first-degree relatives of MDD patients (FH+) and 100 (females, 60; males, 40) healthy never-depressed control subjects (FH−) without first-degree relatives suffering from MDD were recruited for the longitudinal study via notices and word of mouth. All healthy volunteers belonged to the Han Chinese population. All participants (both FH + and FH− participants) and their first-degree relatives (Table S1) were examined at baseline by an experienced psychiatrist or clinical psychologist using the Screening Interview from the Structured Clinical Interview of the DSM-IV (SCID) to assess the presence of current and past psychopathology.²⁸ The FH + group consisted of healthy first-degree relatives of patients with MDD under current or previous treatment, and the affected relatives were required to provide medical reports confirming their diagnosis of MDD. All participants met the following criteria: (1) no history of childhood maltreatment; (2) no history of psychiatric disorder (Axis I or Axis II) or use of psychotropic medication; (3) no history of severe neurological (e.g. concussion, stroke, neuroinflammatory diseases) or medical conditions (e.g. cancer, chronic inflammatory or autoimmune diseases); (4) no other conditions affecting CBF or metabolism such as hypertension, diabetes, or thyroid dysfunction; and (5) no contraindications for MRI scanning. Demographic variables, inclusion criteria, and clinical details were assessed using a standardized questionnaire, as well as through the DSM-IV Structured Clinical Interviews for psychiatric and personality disorders. Additionally, since previous research suggests that stressful life events are powerful risk factors for the development of major depression, participants reported at baseline on the stressful life events (e.g. death of family member/a good friend, parental separation, parent losing a job, breaking up with boyfriend/girlfriend) that they had experienced over the prior year and rated the effect of the negative life event on a four-point scale (1 to 4 scale) using the Life Events Scale for Students (LESS) to obtain a LESS total score by calculating the sum of the scores for all events.^29–32 In addition, to predict the onset of depression in the future, we administered the 17-item Hamilton Depression Rating Scale (HDRS) on the day of the MR examination.³³ An HDRS total score of less than or equal to 7 was applied as an inclusion criterion. Two participants in the FH + group were excluded due to a total HDRS score greater than 7, which left 298 participants (198 FH + and 100 FH− participants) who underwent an MRI scan following ASL and 3D T1-weighted (T1W) protocols at baseline (Time 1).

Figure 1.

Flow chart of our study procedure. Boxes with red edges indicate excluded individuals, the light gray boxes indicate the baseline and follow-up MRI scans, and the yellow box indicates the procedure of follow-up assessments.

Following the baseline MRI, all individuals were contacted every three months and reassessed by two psychiatrists using the SCID. We have conducted the assessments 11 times to date, and we will continue to do so. Of the original 198 FH + participants, 19 dropped out of the study, and 40 were eliminated from participation (36 met the criteria for an anxiety disorder and 4 were diagnosed with bipolar disorder). By contrast, 32 FH + participants (20 females, 12 males) were placed in the DD/FH + group because they were diagnosed with a depressive disorder; 20 of these participants (15 females, 5 males) met the DSM-IV diagnostic criteria for MDD with HDRS scores greater than 24, and 12 met the criteria for a depressive disorder with HDRS scores greater than 7 but less than or equal to 24. Importantly, all the depressive events included in the report represent the first time that the participants had experienced depression. Moreover, a second MRI scan for the DD/FH + participants was performed shortly after diagnosis and before treatment. The remaining 107 (60 females, 47 males) FH + participants who did not develop depression were placed in the HC/FH + group. Of the 100 baseline FH− participants, the data of 4 were lost, and those of 7 were excluded (3 were diagnosed with a depressive disorder and 4 were diagnosed with an anxiety disorder). Therefore, the final sample included 89 participants (54 females, 35 males) based on the inclusion criteria defined above. All healthy participants received a follow-up scan after the 11th assessment, and the HDRS was administered on the same day as the MRI. At each assessment, we also obtained a LESS total score for each participant by asking them the same questions from the LESS to assess the stressful life events that had occurred since the prior assessment. We then calculated the average life stress score for all assessments completed before the follow-up scan (Time 2).

MRI scanning

Imaging was performed on a Siemens 3.0T Trio MR scanner using an eight channel head coil. The resting state perfusion imaging was performed using a pulsed ASL sequence with the following parameters: repetition time = 3249 ms, echo time = 13 ms, TI₁ = 700 ms, TI₂ = 1800 ms, flip angle = 90°; voxel size = 4 mm×4 mm× 3 mm; field of view = 256 mm×256 mm; matrix = 64 ×64; slice thickness = 3 mm, gap = 0.75 mm; 31 axial slices; time = 4 min 20 s. Finally, 91 ASL volumes were acquired and the first was used as the M0 image with the remaining 90 volumes used as 45 control-label pairs. To allow for the processing of MRI data, a high-resolution T1-weighted anatomical image was acquired by using a 3D magnetization-prepared, rapid acquisition gradient echo (MPRAGE: 192 sagittal slices, TR/TE/flip angle = 2530 ms/2.26 ms/90°; voxel size = 1 mm×1 mm × 1 mm; FOV = 256 mm×256 mm, matrix = 256 × 256, slice thickness = 1 mm, gap =0 mm; time = 10 min 8 s). The head was stabilized with a cushion to minimize head motion, and earplugs were used to protect against scanner noise. During the scans, all subjects were instructed to keep their eyes closed, relax without movement, think of nothing in particular, and not fall asleep.

MRI data processing

Before preprocessing, two experienced radiologists checked the data directly for quality control after acquiring the data on an MR scanner, and participants were rescanned when motion artifacts caused by head movement were observed. ASL data were preprocessed using the ASL Perfusion MRI Signal Processing Toolbox (ASLtbx), which is based on SPM12 (http://www.fil.ion.ucl.ac.uk/spm/).^34,35 The preprocessing steps included image reorientation, motion correction, coregistration to the M0 image and then to the anatomical image, temporal filtering, spatial smoothing with an isotropic Gaussian kernel (full width half maximum = 6 mm), the removal of nonbrain tissue, CBF quantification, and normalization.

High-resolution structural magnetic resonance imaging was also used to determine the volume of the significantly changed brain areas in the CBF comparisons. Volume was calculated using FreeSurfer (version 5.0, http://surfer.nmr.mgh.harvard.edu/) as described by Grimm et al.³⁶ This processing step includes motion correction, the removal of non-brain tissue using a hybrid watershed/surface deformation procedure, automated Talairach transformation, intensity normalization, the tessellation of the gray/white matter boundary, automated topology correction, and surface deformation following intensity gradients to optimally place the gray/white and gray/cerebrospinal fluid borders at the location where the greatest shift in intensity defines the transition to the other tissue class. Quality control was performed by visual inspection as described in a recent protocol (http://enigma.loni.ucla.edu/protocols/) and checking for outliers. The volume measures of the significantly changed brain areas in the CBF comparisons were derived from the standard statistics directory using the Desikan Atlas.

Statistical analyses

A total of 228 participants in three groups (32 DD/FH+, 107 HC/FH+, and 89 HC/FH−) were included in the statistical analyses. The participants' demographic and clinical data were analyzed using SPSS 24.0 (IBM SPSS Statistics for Windows, Armonk, NY, USA), and all analyses were two-tailed with an alpha level of 5%. The Chi-square test was used to compare the sex distributions of the groups. Analyses of variance (ANOVAs) were used to compare the groups based on age, HDRS, LESS score and whole-brain global mean CBF. If the group-level test results were significant, then post hoc pairwise comparisons were additionally performed for DD/FH + versus HC/FH+, DD/FH + versus HC/FH−, and HC/FH + versus HC/FH−.

Voxel-based comparisons of the perfusion maps were performed among the three groups at baseline and follow-up using one-way ANOVA. Additionally, post hoc tests were used to reveal the source of any difference detected by ANOVA. Moreover, for the three groups, paired-samples t-tests were performed to compare the CBF maps between baseline and follow-up. In the analyses above, we regarded the global mean CBF as a confounding factor and used its value as a covariate. Importantly, the same analyses were repeated using the volumes of the brain areas with significantly changed CBF and the global mean CBF value as covariates.

Additionally, for the resulting statistical CBF maps, we corrected for multiple comparisons using a familywise error (FWE) threshold of P < 0.05 for the ANOVA and the AlphaSim method for the t-tests (P < 0.001 uncorrected, cluster extent ≥324 for cluster P < 0.05) based on the Correction Thresholds by AlphaSim module of Resting-State fMRI Data Analysis Toolkit V1.8 (http://www.restfmri.net/forum/REST_V1.8).

Then, we extracted the quantitative CBF values of the significantly changed brain areas from the quantitative CBF maps. We used a general linear model (GLM) to compare the CBF values between groups corrected for volume, with or without the global mean CBF as a covariate. Finally, the CBF values of the significantly changed brain areas were tested for associations with the HDRS scores using Pearson's correlation analysis. A logistic regression was used to assess the predictive value of the CBF in the right amygdala at baseline for the first onset of depression in the FH + group. Several covariates were included in the regression models: gender, age at baseline, LESS scores at baseline and follow-up, and HDRS score at baseline. A receiver operating characteristic (ROC) curve was generated to determine the optimal CBF threshold of the right amygdala associated with the first onset of depression among first-degree relatives of patients with MDD. A sensitivity/specificity analysis was performed on the CBF of the right amygdala.

Results

Participant characteristics

The demographic and clinical characteristics of the participants are presented in Table 1. No significant differences were found in age, gender, LESS scores or global mean CBF between groups at baseline or at follow-up. There was also no significant difference in the proportion of women between groups. The three groups showed no significant differences in HDRS scores at baseline (Time 1). As expected, the HDRS scores of the DD/FH + group were significantly higher than those of the HC/FH + or HC/FH− group, and the differences in the HDRS scores between the HC/FH + and HC/FH− groups were not significant at follow-up (Time 2). Moreover, no significant difference was found in the HDRS scores between men and women in any of the three groups at baseline or at follow-up.

Table 1.

Demographic and clinical comparison.

				P Value
Variable	DD/FH+ N = 32	HC/FH+ N = 107	HC/FH− N = 89	Group	DD/FH + vs. HC/FH+	DD/FH + vs. HC/FH−	HC/FH + vs. HC/FH−
Gender (F/M)	20 F/12 M	59 F/48 M	53 F/36 M	NS	NS	NS	NS
Baseline age(y)	19.63 ± 1.48	19.69 ± 1.40	19.36 ± 1.35	NS	NS	NS	NS
LESS scores Time1	11.13 ± 9.64	10.94 ± 8.78	10.66 ± 6.66	NS	NS	NS	NS
LESS scores Time2	6.31 ± 4.50	4.96 ± 4.44	5 ± 3.14	NS	NS	NS	NS
HDRS scores Time1	3.28 ± 1.49	3.14 ± 1.22	3.13 ± 1.38	NS	NS	NS	NS
HDRS scores Time2	19.34 ± 5.86	4.63 ± 1.29	4.16 ± 1.22	<0.001	<0.001	<0.001	NS
Time baseline to follow-up (months)	28.5 ± 4.31	33 ± 0	33 ± 0	…	…	…	…
Global mean CBF Time1 (mL/100 g/min)	55.55 ± 5.53	53.76 ± 6.17	54.26 ± 5.46	NS	NS	NS	NS
Global mean CBF Time2 (mL/100 g/min)	54.72 ± 5.09	53.56 ± 6.42	54.30 ± 5.43	NS	NS	NS	NS

Note: Values shown are the mean ± SD.

NS: not significant.

Voxel-based analysis

Group differences in CBF were corrected for the global mean CBF by using its value as a covariate, with or without the volume as a covariate. Compared with the HC/FH + and HC/FH− groups, the DD/FH + group showed significantly increased CBF in the right amygdala at baseline but reduced CBF at follow-up. Furthermore, compared with that of the HC/FH− group, the CBF of the right amygdala in the HC/FH + group showed no significant differences at baseline but was significantly decreased at follow-up. In addition, the DD/FH + and HC/FH + groups exhibited a significantly decreased CBF in the right amygdala at follow-up compared with baseline (Table S2). The differences in CBF between baseline and follow-up in the HC/FH− group were not significant. Importantly, the differences in CBF in the right amygdala remained significant when the volume of the right amygdala was included as a covariate in the between-group CBF comparisons (Figures 2(b), 3, and 4).

Figure 2.

(a) Perfusion map randomly selected from the DD/FH + group at baseline. Value of each voxel represents cerebral blood flow (CBF). (b) A voxel-based analysis showing the right amygdala with increased (red) CBF, corrected for global mean CBF and the volume in DD/FH + vs. HC/FH + vs. HC/FH− (F test, P < 0.05 FEW corrected) at baseline. RA: right amygdala.

Figure 3.

(a) A voxel-based analysis showing decreased (blue) CBF in the right amygdala and left middle frontal gyrus as well as increased (red) CBF in the left hippocampus, corrected for global mean CBF and volume in DD/FH + vs. HC/FH + vs. HC/FH− (F test, P < 0.05 FEW corrected) at follow-up. (b) The CBF difference between the HC/FH + and HC/FH− groups at follow-up showing decreased (blue) CBF in the right amygdala, corrected for global mean CBF and volume (T test, P < 0.001 uncorrected, cluster sizes >324 mm³ for cluster P < 0.05). RA: right amygdala; LH: left hippocampus; LMFG: left middle frontal gyrus.

Figure 4.

(a) The CBF changes between baseline and follow-up of the DD/FH + group showing decreased (blue) CBF in the right amygdala and left middle frontal gyrus, corrected for global mean CBF and volume. (b) The CBF changes between baseline and follow-up of the HC/FH + group showing decreased (blue) CBF in the right amygdala, corrected for global mean CBF and volume (T test, P < 0.001 uncorrected, cluster sizes > 324 mm³ for cluster P < 0.05). RA: right amygdala; LMFG: left middle frontal gyrus.

Moreover, the HC/FH− group exhibited decreased CBF in the left insula compared with both the DD/FH + and HC/FH + groups at baseline. However, this difference was no longer significant when the left insula volume was included as a covariate. In addition to the reduced CBF in the right amygdala, the DD/FH + group had a significantly reduced CBF in the left middle frontal gyrus and increased CBF in the bilateral hippocampus compared with the HC/FH + and HC/FH− groups at follow-up. DD/FH + participants displayed significantly reduced CBF in the right amygdala and the left middle frontal gyrus at follow-up compared with baseline (Table S2). The CBF in the right hippocampus showed no significant difference when the volume was included as a covariate, and the volumes of other significant brain regions had no effects on the CBF comparisons (Figures 3(a) and 4(a)).

GLM analysis

Group CBF differences in the significantly changed brain areas were corrected for volume, with or without the global mean CBF as a covariate. Interestingly, regardless of whether the whole brain average CBF was corrected, the CBF of the right amygdala in the DD/FH + group was increased at baseline and decreased at follow-up (Figure 5(a) and (b), Table S3). In addition, the global mean CBF showed no effects on the CBF comparisons in the left insula, left middle frontal gyrus, left hippocampus or right hippocampus (Figure 5(b) and Table S3).

Figure 5.

(a) The scatterplots show the results of the post hoc t-tests of the CBF differences among the DD/FH+, HC/FH + and HC/FH− groups, corrected for volume at baseline. The asterisk represents P < 0.05. (b) The scatterplots show the results of the post hoc t-tests of the CBF differences among the DD/FH+, HC/FH + and HC/FH− groups at follow-up, corrected for volume. The asterisk represents P < 0.05. (c) The scatterplot shows the correlation between the CBF (Time2) of the right amygdala and the HDRS scores at follow-up (Time2) for the FH + group (r = −0.191, P = 0.024). (d) The scatterplot shows the correlation between the CBF at baseline of the right amygdala and %HDRS scores at follow-up for the FH + group (r = 0.444, P < 0.001), %HDRS = (Time2 HDRS – Time1 HDRS) / Time1 HDRS. (e) The scatterplot shows the correlation between the CBF (Time1) of the right amygdala at baseline and the HDRS scores at follow-up (Time2) for the DD/FH + and HC/FH + group (r = 0.556, P < 0.001 and r = 0.323, P < 0.001, respectively). (f) The ROC curve of the CBF values in the right amygdala at baseline. An optimal CBF cutoff value of 40.50 mL/100 g/min was determined at a sensitivity of 87.5% and a specificity of 73.8%.

Correlation between altered CBF in the right amygdala and HDRS scores

The results of the correlation analysis are shown in Table S4. For the FH + group, the increased CBF of the right amygdala at baseline was not significantly correlated with the HDRS scores at baseline (r = 0.071, P = 0.403). However, the CBF of the right amygdala at follow-up was significantly and negatively correlated with the HDRS scores at follow-up (r = −0.191, P = 0.024; Figure 5(c)). Furthermore, the increased CBF in the right amygdala at baseline compared with follow-up was significantly correlated with the increase in HDRS scores (%HDRS = (t2 HDRS – t1 HDRS)/t1 HDRS; r = 0.444, P < 0.001; Figure 5 (d)). A significant correlation was found between the initial CBF of the right amygdala and t2 HDRS scores in the DD/FH + and HC/FH + groups (r = 0.556, P < 0.001 and r = 0.323, P < 0.001, respectively; Figure 5(e)). By contrast, the correlation between the t2 HDRS scores and the t1 CBF of the right amygdala of the HC/FH− participants (N = 89) was not significant (r = 0.038, P = 0.711).

In addition, none of the HDRS scores were significantly correlated with the CBF of the left insula, the bilateral hippocampus, or the left middle frontal gyrus.

Increased CBF in the right amygdala at baseline predicts the onset of depression in the FH + group

Table 2 shows the results of the logistic regression analysis in which the onset of depression was predicted by CBF in the right amygdala at baseline, gender, age at baseline, LESS score at baseline and follow-up, and HDRS score at baseline. The logistic regression demonstrated that neither baseline age (β = −0.041, P = 0.818, OR = 0.94) nor gender (β = −0.093, P = 0.862, OR = 0.91) uniquely predicted the first onset of depression. Similarly, neither the LESS score at baseline (β = −0.012, P = 0.630, OR = 0.99) nor follow-up (β = 0.047, P = 0.322, OR = 1.048) predicted the development of depression. In addition, the HDRS score at baseline did not significantly predict the development of depression (β = 0.089, P = 0.630, OR = 1.09). After adjusting for these covariates, the logistic regression analysis revealed that the baseline CBF in the right amygdala significantly predicted the first onset of depression in the FH + group (β = 0.198, P < 0.001, OR = 1.22). Furthermore, ROC analyses revealed the capacity of the CBF values in the right amygdala at baseline to predict the onset of depression in the FH + group (P < 0.001; area under the ROC curve [AUC], 0.86 [95% confidence interval {0.80; 0.92}]). According to the best Youden Index, the corresponding cutoff value in this sample was 40.50 mL/100 g/min, with a sensitivity of 87.5% and specificity of 73.8% (Figure 5(f)).

Table 2.

Predicting depression onset.

Variables	B	SE	Wald	Sig.	Exp(B)
Age	−0.041	0.177	0.053	0.818	0.960
Gender	−0.093	0.534	0.030	0.862	0.912
t1 HDRS	0.091	0.185	0.240	0.624	1.095
t1 LESS	−0.013	0.025	0.268	0.605	0.987
t2 LESS	0.046	0.049	0.904	0.342	1.047
t1 CBF	0.202	0.040	25.811	0.000	1.224

Note: t1, at baseline; t2, at follow-up.

Discussion

To the best of our knowledge, this study is the first to use quantitative CBF to predict the first onset of depression in never-depressed individuals with a family history of depression. In this study, we applied the ASL-MRI technique to investigate quantitative CBF before and after the diagnosis of depression. Following cross-sectional and longitudinal comparisons of CBF, a consistently abnormal brain area was identified: the right amygdala. Additionally, subsequent correlational analyses showed a significant relationship between altered CBF in the right amygdala and depression severity (HDRS scores) in FH + individuals. Taken together, these results reveal that altered CBF in the right amygdala is a neural biomarker of vulnerability to depression in participants with a family history of this condition. Afterward, we used a logistic regression model and ROC curves to demonstrate that healthy young adults with a family history of depression and increased CBF in the right amygdala are more susceptible to developing depression than young adults without increased CBF in the right amygdala.

In this study, we used the global mean CBF as a covariate in voxel-based analyses to eliminate individual differences. However, further analyses showed that with or without the global mean CBF included as a covariate, the group difference in the CBF in the abnormal brain areas was consistent, which suggests that the global mean CBF has no effect on the group differences in CBF, including the main finding of this study, i.e. increased CBF in the right amygdala in the DD/FH + group at baseline. Several studies, including the present one, have revealed amygdala abnormalities in FH + participants. The amygdala plays an important role in the perception, regulation and memory of emotional information.^37–39 Moreover, Xu et al.⁴⁰ found that FK506 binding protein 51 (FKBP5) and miR-124a-mediated GR function in the basolateral amygdala are involved in susceptibility to depressive disorder in rats subjected to stress during early adolescence. Additionally, some previous studies have found that the serotonin transporter gene (5-HTTLPR), in which a common polymorphism is an important mediator of individual differences in the brain responses associated with emotional behavior, mediates the modulation of amygdala activity.^41,42 Not surprisingly, given the relationship between the amygdala and depression as well as between the amygdala and genes, previous neuroimaging investigations of individuals with a family history of depression have focused on this structure.⁴³ Unlike these previous studies that primarily found abnormalities present in the volume and neural response of the bilateral, right, or left amygdala prior to the onset of depressive disorder in never-depressed individuals at familial risk for depression, our study showed altered CBF in the right amygdala, with or without amygdala volume included as a covariate.^16,43,44

Our study found that except for the HC/FH− group, the DD/FH + and HC/FH + groups showed reductions in right amygdala CBF at follow-up compared with baseline. Furthermore, the DD/FH + and HC/FH + groups showed reduced right amygdala CBF at follow-up compared with the HC/FH− group. It is also conceivable that the reduced CBF in the DD/FH + group reflects an interaction among depression risk (family history), depressive state, and stress in university life but that it reflects the interaction between depression risk (family history) and stress in university life in the HC/FH+. More importantly, we found increased CBF in the right amygdala prior to the onset of depression only in healthy young adults who were predisposed to depression and had a familial risk, and not in all individuals who had a family history of depression. Additionally, our studies applied a retrospective longitudinal approach to identify whether the increased CBF of the right amygdala at baseline in FH + individuals predicted the eventual onset of depressive disorder; no previous neuroimaging studies have conducted a similar approach. Interestingly, prior to the onset of depression in all FH + participants, decreased CBF in the left insula was observed when the volume of the left insula was not considered, but this difference was not observed when the volume was considered, which suggests that family risk influences the CBF of the left insula when the volume of the left insula is not considered. In addition, none of the HDRS scores were significantly correlated with the CBF of the left insula in our study. Thus, we believe that the left insula is not the site of the characteristic features of the FH + participants prone to depression in our study, although other studies have revealed insula abnormalities in FH + participants and suggested that neural processing abnormalities in the insula are an important feature of heritable vulnerability to the disease, given the crucial role that the insula plays in interoception and subjective emotional experience.^45,46

Another highlight of this study is that we used quantitative CBF as measured by ASL, a noninvasive perfusion MRI technique, to predict the onset of depression.⁴⁷ First, CBF (the rate of delivery of arterial blood to the capillary bed of a particular mass of tissue) is closely related to neural function and metabolism as well as directly and effectively reflects the altered neural activity of patients with depression. Second, compared with other neuroimaging techniques (e.g. fMRI-BOLD and structural MRI) used to predict depression onset, ASL can be used to collect CBF data rapidly, with clinical imaging times under 5 min; moreover, it can provide quantitative measurements in standard units of mL/100 g brain tissue/min to facilitate interobserver and intraobserver consistency.⁴⁸ Furthermore, compared with positron emission tomography (PET) and single-photon emission computed tomography (SPECT), both of which require the use of tracers, ASL uses an endogenous marker, the water molecule, which enables repeated use without radiation exposure.⁴⁸ Importantly, ASL has a diagnostic accuracy comparable with PET.^49–51

Recently, as mentioned in previous reviews, an increasing number of investigations have assessed ASL in patients with depression.^52,53 For instance, some studies have indicated that different subtypes and symptoms of depression are associated with specific patterns of CBF alterations, including increased white matter CBF in late-life depression and decreased CBF in the primary motor cortex in people with MDD featuring psychomotor retardation.^54,55 When volume was considered as a covariate in our study, individuals with depression (DD/FH + participants) showed reduced CBF in the left middle frontal gyrus and the right amygdala at the onset of depression compared with predepression episodes. At follow-up imaging consultations, individuals with depression showed a complex pattern of hypoperfusion in the left middle frontal gyrus but hyperperfusion within the left hippocampus compared with healthy participants (HC/FH + and HC/FH− groups). These findings partially matched the results of several previous imaging studies such as the study by Lui et al.⁵⁶ who found that participants with depressive disorder had increased CBF primarily at limbic-striatal sites (including the insula, amygdala, and striatal structures) as well as reduced CBF in the left middle frontal gyrus and many other areas compared with healthy participants. In a study of 25 medication-naive adolescents with MDD, Ho et al.⁵⁷ found hypoperfusion in the frontal gyrus and right amygdala as well as in many other areas. However, our finding might be the first to consider alterations in the CBF of untreated participants with first-episode depression and a family history of depression following a cross-sectional and longitudinal comparison. These abnormalities might be a specific characteristic of people with first-episode depression with a family history.

Undoubtedly, the most prominent highlight of this study is our finding that increased CBF in the right amygdala at baseline in individuals with a family history of depression helps predict the eventual onset of depressive disorder. Based on previous publications, preventive interventions targeting individuals at high risk for depression show promise not only in delaying the occurrence of depression but also in preventing depression.⁵⁸ At the same time, the discovery of biomarkers for predicting depression is a key part of the effort to prevent depression in high-risk groups. Furthermore, the biomarkers for predicting depression are indicators for monitoring the development of depression that are expected to enable timely, appropriate interventions to prevent depression and even help clinicians judge the effectiveness of interventions.

We acknowledge several limitations of our study. The DD/FH + group was smaller than either the HC/FH + or the HC/FH− group, which might have reduced the sensitivity to effects in the DD/FH + group and limited the generalizability of our findings. In addition, we did not provide a DD/FH− group for comparison because of the small sample size, which might also have reduced the sensitivity to the effects of the familial risk of depression. Fortunately, our study is continuing its follow-up assessments of the remaining individuals; thus, the DD/FH + and DD/FH− groups might have sufficient sample sizes to address this limitation and test for similar results in the future. Another limitation of the current study is the selection bias that arises from the recruitment of participants. First, although several previous studies have found that the most common period of depression onset is during late adolescence, the participants in our study were enrolled at 18 years old or older.⁵⁹ Second, college students have socioeconomic advantages and are more likely than the general population to be employed, which gives rise to the healthy worker effect; that is, the prevalence of physical and mental health problems is reduced among employed individuals.^60–63 We anticipate that future research will address these limitations and might find larger effect sizes than those reported in the present study. In the future, we might be able to explore the influence of familial risk on the severity of depression or other meaningful issues such as outcome, recurrence, and suicidal tendency. Furthermore, our study eliminated all participants with anxiety disorder for the sake of sample homogeneity during the follow-up period because comorbidity is common between depression and anxiety disorders; altered CBF in certain regions (e.g. amygdala) is common in people with anxiety; and the onset of anxiety disorders precedes the onset of depression in community and clinical populations of children, adolescents, and adults.^64–67 In the future, we might explore the influence of the familial risk of depression or anxiety on people with depression and comorbid anxiety.

Despite these limitations, the current study is important in that it is the first direct investigation of altered CBF levels prior to the onset of depressive disorder in healthy young adults with a family history of depression and susceptibility to depression. This study sought to identify whether altered CBF is useful for predicting the subsequent onset of depression. We conclude that increased ASL-CBF in the right amygdala predicts the first onset of depression in healthy young first-degree relatives of patients with major depression.

Supplemental Material

Supplemental material for Increased ASL-CBF in the right amygdala predicts the first onset of depression in healthy young first-degree relatives of patients with major depression

Supplemental Material for Increased ASL-CBF in the right amygdala predicts the first onset of depression in healthy young first-degree relatives of patients with major depression by Ningning Zhang, Jiasheng Qin, Jinchuan Yan, Yan Zhu, Yuhao Xu, Xiaolan Zhu, Shenghong Ju and Yuefeng Li in Journal of Cerebral Blood Flow & Metabolism

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by National Natural Science Foundation of China (grant numbers 81525014, 81871343); Jiangsu Provincial Key Research and Development Plan (grant number BE2017698); the Natural Science Foundation of Jiangsu Province (grant number BK20171311); project of Jiangsu Provincial Health and Family Planning Commission (grant number Q201605); project of young key teacher training (grant number 5524040001); project of Jiangsu Provincial young medical talents (grant number QNRC2016832, QNRC2016460); the six talent peaks project in Jiangsu Province(grant number 2016-WSW-125); “333 Project” research projects in Jiangsu Province (grant number BRA2016141), and project of Zhenjiang social development (grant number SH2017014).

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Authors' contributions

NZ, JQ and YL made a substantial contribution to the concept and design, analysis and interpretation of data; NZ and JQ drafted the article and processed MRI; YL, SJ and XZ revised the article critically for important intellectual content; JY, YZ, YX organized the study and supported the data analysis. All authors were critically involved in the theoretical discussion and performing of the experiments. All authors read and approved the final version of the manuscript.

Supplemental material

Supplemental material for this paper can be found at the journal website:

References

Buntrock

Ebert

Lehr

, et al. Evaluating the efficacy and cost-effectiveness of web-based indicated prevention of major depression: design of a randomised controlled trial. BMC Psychiatry 2014; 14: 25.

Lambert

Harter

Arnold

, et al. [Improving mental health literacy and mental illness stigma in the population of Hamburg]. Psychiatr Prax 2015(42 Suppl 1 S9–S13.

Whiteford

Degenhardt

Rehm

, et al. Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010. Lancet 2013; 382: 1575–1586.

Bembnowska

Jośko-Ochojska

. What causes depression in adults?. Pol J Public Health 2015; 125: 116–120.

Sullivan

Neale

Kendler

. Genetic epidemiology of major depression: review and meta-analysis. Am J Psychiatry 2000; 157: 1552–1562.

Weissman

Wickramaratne

Adams

, et al. The relationship between panic disorder and major depression. A new family study. Arch Gen Psychiatry 1993; 50: 767–780.

Weissman

Wickramaratne

Nomura

, et al. Families at high and low risk for depression: a 3-generation study. Arch Gen Psychiatry 2005; 62: 29–36.

Foland-Ross

Sacchet

Prasad

, et al. Cortical thickness predicts the first onset of major depression in adolescence. Int J Dev Neurosci 2015; 46: 125–131.

Pan

Sato

Salum

, et al. Ventral striatum functional connectivity as a predictor of adolescent depressive disorder in a longitudinal community-based sample. Am J Psychiatry 2017; 174: 1112–1119.

10.

LeMoult

Ordaz

Kircanski

, et al. Predicting first onset of depression in young girls: interaction of diurnal cortisol and negative life events. J Abnorm Psychol 2015; 124: 850–859.

11.

Amico

Meisenzahl

Koutsouleris

, et al. Structural MRI correlates for vulnerability and resilience to major depressive disorder. J Psychiatry Neurosci 2011; 36: 15–22.

12.

Baare

Vinberg

Knudsen

, et al. Hippocampal volume changes in healthy subjects at risk of unipolar depression. J Psychiatr Res 2010; 44: 655–662.

13.

Boccardi

Almici

Bresciani

, et al. Clinical and medial temporal features in a family with mood disorders. Neurosci Lett 2010; 468: 93–97.

14.

Hao

Talati

Shankman

, et al. Stability of cortical thinning in persons at increased familial risk for major depressive disorder across 8 years. Biol Psychiatry Cogn Neurosci Neuroimaging 2017; 2: 619–625.

15.

Peterson

Warner

Bansal

, et al. Cortical thinning in persons at increased familial risk for major depression. Proc Natl Acad Sci U S A 2009; 106: 6273–6278.

16.

Romanczuk-Seiferth

Pohland

Mohnke

, et al. Larger amygdala volume in first-degree relatives of patients with major depression. Neuroimage Clin 2014; 5: 62–68.

17.

Saleh

Carballedo

Lisiecka

, et al. Impact of family history and depression on amygdala volume. Psychiatry Res 2012; 203: 24–30.

18.

Gotlib

Hamilton

Cooney

, et al. Neural processing of reward and loss in girls at risk for major depression. Arch Gen Psychiatry 2010; 67: 380–387.

19.

Macoveanu

Knorr

Skimminge

, et al. Altered reward processing in the orbitofrontal cortex and hippocampus in healthy first-degree relatives of patients with depression. Psychol Med 2014; 44: 1183–1195.

20.

Lisiecka

Carballedo

Fagan

, et al. Altered inhibition of negative emotions in subjects at family risk of major depressive disorder. J Psychiatr Res 2012; 46: 181–188.

21.

Mannie

Taylor

Harmer

, et al. Frontolimbic responses to emotional faces in young people at familial risk of depression. J Affect Disord 2011; 130: 127–132.

22.

Monk

Klein

Telzer

, et al. Amygdala and nucleus accumbens activation to emotional facial expressions in children and adolescents at risk for major depression. Am J Psychiatry 2008; 165: 90–98.

23.

van der Veen

Evers

Deutz

, et al. Effects of acute tryptophan depletion on mood and facial emotion perception related brain activation and performance in healthy women with and without a family history of depression. Neuropsychopharmacology 2007; 32: 216–224.

24.

Buxton

Frank

Wong

, et al. A general kinetic model for quantitative perfusion imaging with arterial spin labeling. Magn Reson Med 1998; 40: 383–396.

25.

Obata

Liu

Miller

, et al. Discrepancies between BOLD and flow dynamics in primary and supplementary motor areas: application of the balloon model to the interpretation of BOLD transients. Neuroimage 2004; 21: 144–153.

26.

Chen

Zhao

, et al. Voxel-wise meta-analyses of brain blood flow and local synchrony abnormalities in medication-free patients with major depressive disorder. J Psychiatry Neurosci 2015; 40: 401–411.

27.

Kessler

Berglund

Demler

, et al. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry 2005; 62: 593–602.

28.

First MB, Spitzer RL, Gibbon M, Williams JBW. Structured clinical interview for DSM-IV-TR axis I disorders, research version, patient edition with psychotic screen. New York: Biometrics Research, New York State Psychiatric Institute, 2002.

29.

Galea

Ahern

Resnick

, et al. Psychological sequelae of the September 11 terrorist attacks in New York City. N Engl J Med 2002; 346: 982–987.

30.

Kendler

Karkowski

Prescott

. Causal relationship between stressful life events and the onset of major depression. Am J Psychiatry 1999; 156: 837–841.

31.

Swartz

Knodt

Radtke

, et al. A neural biomarker of psychological vulnerability to future life stress. Neuron 2015; 85: 505–511.

32.

Clements

Turpin

. The life events scale for students: validation for use with British samples. Pers Individ Diff 1996; 20: 747–751.

33.

Williams

. A structured interview guide for the Hamilton depression rating scale. Arch Gen Psychiatry 1988; 45: 742–747.

34.

Wang

. Improving cerebral blood flow quantification for arterial spin labeled perfusion MRI by removing residual motion artifacts and global signal fluctuations. Magn Reson Imaging 2012; 30: 1409–1415.

35.

Wang

Aguirre

Rao

, et al. Empirical optimization of ASL data analysis using an ASL data processing toolbox: ASLtbx. Magn Reson Imaging 2008; 26: 261–269.

36.

Grimm

Pohlack

Cacciaglia

, et al. Amygdalar and hippocampal volume: a comparison between manual segmentation, Freesurfer and VBM. J Neurosci Methods 2015; 253: 254–261.

37.

Adolphs

. Fear, faces, and the human amygdala. Curr Opin Neurobiol 2008; 18: 166–172.

38.

Kensinger

Corkin

. Two routes to emotional memory: distinct neural processes for valence and arousal. Proc Natl Acad Sci U S A 2004; 101: 3310–3315.

39.

Pessoa

Adolphs

. Emotion processing and the amygdala: from a ‘low road’ to ‘many roads’ of evaluating biological significance. Nat Rev Neurosci 2010; 11: 773–783.

40.

Wang

Liu

, et al. FKBP5 and specific microRNAs via glucocorticoid receptor in the basolateral amygdala involved in the susceptibility to depressive disorder in early adolescent stressed rats. J Psychiatr Res 2017; 95: 102–113.

41.

Hariri

Drabant

Munoz

, et al. A susceptibility gene for affective disorders and the response of the human amygdala. Arch Gen Psychiatry 2005; 62: 146–152.

42.

von dem Hagen

Passamonti

Nutland

, et al. The serotonin transporter gene polymorphism and the effect of baseline on amygdala response to emotional faces. Neuropsychologia 2011; 49: 674–680.

43.

Foland-Ross

Hardin

Gotlib

. Neurobiological markers of familial risk for depression. Curr Top Behav Neurosci 2013; 14: 181–206.

44.

Swartz

Williamson

Hariri

. Developmental change in amygdala reactivity during adolescence: effects of family history of depression and stressful life events. Am J Psychiatry 2015; 172: 276–283.

45.

Liu

, et al. Decreased regional homogeneity in insula and cerebellum: a resting-state fMRI study in patients with major depression and subjects at high risk for major depression. Psychiatry Res 2010; 182: 211–215.

46.

Opel

Zwanzger

Redlich

, et al. Differing brain structural correlates of familial and environmental risk for major depressive disorder revealed by a combined VBM/pattern recognition approach. Psychol Med 2016; 46: 277–290.

47.

Detre

Rao

Wang

, et al. Applications of arterial spin labeled MRI in the brain. J Magn Reson Imaging 2012; 35: 1026–1037.

48.

Watts

Whitlow

Maldjian

. Clinical applications of arterial spin labeling. NMR Biomed 2013; 26: 892–900.

49.

Chen

Wolk

Reddin

, et al. Voxel-level comparison of arterial spin-labeled perfusion MRI and FDG-PET in Alzheimer disease. Neurology 2011; 77: 1977–1985.

50.

Musiek

Chen

Korczykowski

, et al. Direct comparison of fluorodeoxyglucose positron emission tomography and arterial spin labeling magnetic resonance imaging in Alzheimer's disease. Alzheimers Dement 2012; 8: 51–59.

51.

Zhang

Herzog

Mauler

, et al. Comparison of cerebral blood flow acquired by simultaneous [15O]water positron emission tomography and arterial spin labeling magnetic resonance imaging. J Cereb Blood Flow Metab 2014; 34: 1373–1380.

52.

Haller

Zaharchuk

Thomas

, et al. Arterial spin labeling perfusion of the brain: emerging clinical applications. Radiology 2016; 281: 337–356.

53.

Théberge

. Perfusion magnetic resonance imaging in psychiatry. Top Magn Reson Imaging 2008; 19: 111–130.

54.

Colloby

Firbank

, et al. Regional cerebral blood flow in late-life depression: arterial spin labelling magnetic resonance study. Br J Psychiatry 2012; 200: 150–155.

55.

Yin

Wang

, et al. Decreased cerebral blood flow in the primary motor cortex in major depressive disorder with psychomotor retardation. Prog Neuro Psychopharmacol Biol Psych 2018; 81: 438–444.

56.

Lui

Parkes

Huang

, et al. Depressive disorders: focally altered cerebral perfusion measured with arterial spin-labeling MR imaging. Radiology 2009; 251: 476–484.

57.

Shin

, et al. Altered cerebral perfusion in executive, affective, and motor networks during adolescent depression. J Am Acad Child Adolesc Psychiatry 2013; 52: 1076–1091.e1072.

58.

Loechner

Starman

Galuschka

, et al. Preventing depression in the offspring of parents with depression: a systematic review and meta-analysis of randomized controlled trials. Clin Psychol Rev 2018; 60: 1–14.

59.

Hankin

Abramson

Moffitt

, et al. Development of depression from preadolescence to young adulthood: emerging gender differences in a 10-year longitudinal study. J Abnorm Psychol 1998; 107: 128–140.

60.

Kessler

Berglund

Demler

, et al. The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R). JAMA 2003; 289: 3095–3105.

61.

Peyrot

Lee

Milaneschi

, et al. The association between lower educational attainment and depression owing to shared genetic effects? Results in ∼25,000 subjects. Mol Psychiatry 2015; 20: 735–743.

62.

Sung

. A review of the healthy worker effect in occupational epidemiology. Occup Med 1999; 49: 225–229.

63.

Agerbo

. Effect of psychiatric illness and labour market status on suicide: a healthy worker effect?. J Epidemiol Community Health 2005; 59: 598–602.

64.

Lamers

van Oppen

Comijs

, et al. Comorbidity patterns of anxiety and depressive disorders in a large cohort study: the Netherlands Study of Depression and Anxiety (NESDA). J Clin Psychiatry 2011; 72: 341–348.

65.

Meier

Petersen

Mattheisen

, et al. Secondary depression in severe anxiety disorders: a population-based cohort study in Denmark. Lancet Psychiatry 2015; 2: 515–523.

66.

Fava

Rankin

Wright

, et al. Anxiety disorders in major depression. Compr Psychiatry 2000; 41: 97–102.

67.

Hasler

Fromm

Alvarez

, et al. Cerebral blood flow in immediate and sustained anxiety. J Neurosci 2007; 27: 6313–6319.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.13 MB