Brain scans reveal how recurrent depression leaves a lasting mark on the amygdala

Brain scans reveal that elevated activity in the emotion-processing center is tied to a person’s history of recurrent depression, rather than their current mood. The large study, published in Psychological Medicine, suggests that each major depressive episode may leave a lasting biological mark that increases future vulnerability to the disease.

Major depressive disorder is a common psychiatric condition that affects millions of people worldwide. Currently, psychiatrists diagnose the condition based on clinical interviews and patient history. There are no biological markers, like a blood test or a brain scan, to guide treatment choices.

Functional magnetic resonance imaging, or fMRI, allows researchers to observe the brain in action by tracking blood-oxygen levels. When a brain region becomes active, it requires more oxygen, leading to localized changes in blood flow. The scanner detects these magnetic differences to map out neural activity.

Scientists often use this technology to study the amygdala, an almond-shaped structure deep inside the brain that processes fear and negative emotions. Early brain imaging research suggested that people with clinical depression have hyperactive amygdalae when looking at negative images. But a recent analysis of a massive dataset called the UK Biobank found no association between amygdala activity and current depression symptoms.

Jerke J. van den Berg, a biomedical researcher at the University of Amsterdam, and his colleagues designed a new study to better understand this discrepancy. They suspected that previous research might have missed the broader picture by only looking at a patient’s current mood. The researchers focused on a concept called the kindling theory.

The kindling theory proposes that an initial depressive episode makes the brain more sensitive to stress. After the brain has been sensitized by that first experience, it takes progressively less trauma to trigger a relapse. To test if the amygdala reflects this effect, the research team decided to look at a person’s lifetime history of depression, known as a trait, rather than their current symptoms, known as a state.

The researchers utilized data from the UK Biobank, a long-term population health study. They analyzed functional MRI scans from a subset of participants, totaling more than 11,000 individuals. While inside the scanner, participants completed a visual exercise called the Hariri task.

During the task, participants were shown a target image of an angry or fearful face at the top of a screen. They were then asked to select the matching face from two options at the bottom. This specific visual matching exercise is known to reliably stimulate the amygdala.

Brain activity can vary widely from person to person based on age, gender, and head movement during a scan. To account for this natural variation, the research team used a statistical technique called normative modeling. They analyzed scans from over 6,400 healthy participants to establish a baseline of expected amygdala activity. This works much like a pediatric growth chart, which maps out normal ranges for a child’s height and weight.

Next, the researchers evaluated how much the brain activity of nearly 5,000 other participants deviated from this baseline model. They categorized these individuals based on their self-reported mental health histories. The groups included healthy controls, people who had experienced a single depressive episode, those with moderate recurrence involving two to five episodes, and those with a high recurrence of six or more episodes.

For this initial cross-sectional analysis, the team focused strictly on participants who were currently in remission from their depression. The initial results were not statistically significant when the team analyzed the unaltered brain scans. However, once they applied the normative modeling technique to account for age and gender variations, a distinct pattern emerged.

The analysis revealed a measurable association between an individual’s history of depression and their amygdala response. Participants with a high recurrence of depressive episodes showed a heightened amygdala reaction to negative faces compared to healthy controls.

When the researchers looked at individuals actively experiencing a depressive episode, they found no distinct increase in brain activity compared to controls. This suggested that amygdala reactivity represents a long-term biological trait, rather than a temporary state reflecting current mood.

The researchers also wanted to know how medication might influence these brain signals. They noticed that a higher percentage of people in the severe recurrence group were taking antidepressants compared to those with a single past episode. They repeated their cross-sectional analysis, this time removing any participants who were actively taking antidepressant medications.

Excluding medicated individuals strengthened the observed differences between the healthy controls and the recurrent depression groups. The findings indicated that antidepressants might dampen the hyperactive amygdala signal associated with a history of recurrent depression. Because the medication reduced amygdala reactivity, including these participants in the initial data pool slightly masked the true extent of the brain changes.

To see how the brain changes over time, the team conducted a longitudinal analysis. They focused on a smaller group of participants who returned for a second brain scan roughly two and a half years after their initial visit. The researchers categorized these individuals based on whether they had suffered new depressive episodes between the two scans.

For this longitudinal evaluation, the team specifically analyzed people who were in remission during both of their imaging sessions. Participants who began the study with a history of just one depressive episode, but then experienced multiple new episodes before their second scan, exhibited an increase in amygdala reactivity over time.

This brain change supported the kindling theory. It suggests that new depressive episodes incrementally alter how the brain processes negative emotional information, leaving a biological mark even after symptoms fade.

The study relied on a large dataset, but the researchers noted that the effect sizes were relatively small. These findings do not mean that a functional MRI scan can be used to diagnose depression in a clinical setting right now. The results are not robust enough to predict an individual’s exact risk of a relapse based on a single brain scan.

The data collection methods also presented certain limitations. The study depended on participants accurately recalling their own mental health histories, which can introduce memory biases. People might misremember exactly how many distinct depressive episodes they experienced over the course of their lives.

The mental health questionnaires also combined treatments for nerves, anxiety, and depression into a single metric. Because of this, the researchers could not strictly isolate the effects of anxiety disorders from the effects of clinical depression. Future studies will need to track larger groups of symptomatic individuals over extended periods of time to untangle these variables.

Scientists hope that advancing neuroimaging techniques will eventually reduce the normal fluctuations seen in brain scans. Over time, mapping the biology of recurrent depression could help psychiatrists tailor treatments to a patient’s individual history, moving away from the current trial-and-error approach to prescribing medication.

The study, “Normative amygdala fMRI response during emotional processing as a trait of depressive symptoms in the UK Biobank,” was authored by Jerke J. van den Berg, Henricus G. Ruhé, Henk A. Marquering, Liesbeth Reneman, and Matthan W. A. Caan.