The COVID-19 pandemic necessitated unprecedented restriction measures globally, impacting daily life and potentially exacerbating psychiatric symptoms. Previous outbreaks showed a correlation between quarantine and long-lasting psychological effects. The COVID-19 pandemic's extensive restrictions raised concerns about a surge in stress, anxiety, and depression, particularly among individuals with pre-existing mental health conditions like MDD. The diathesis-stress model and differential susceptibility framework suggest that individuals with mental disorders might be more vulnerable to external stressors. However, studies indicated less-than-expected changes in psychological symptoms during the pandemic for those with pre-existing conditions. Beyond psychometric measures, alterations in brain structure and function are key adaptational aspects affecting mental well-being in response to stressors. While literature on neurobiological mechanisms impacting coping and psychiatric symptoms due to restrictions is limited, studies show structural brain changes following traumatic life events, such as decreased grey matter volume linked to reduced BDNF levels. The COVID-19 pandemic presented a unique opportunity to study longitudinal changes in adaptational processes during a widespread stressful event. The researchers hypothesized that lockdowns would impact amygdala, hippocampus, anterior cingulate cortex, and prefrontal cortex volumes, serum BDNF levels, and perceived stress and depressive symptoms, more profoundly in MDD patients than healthy controls.

Existing literature highlighted the psychological impact of quarantine and the potential exacerbation of pre-existing mental health conditions during infectious disease outbreaks. Studies indicated associations between perceived stress during lockdowns and the development of depressive symptoms. The diathesis-stress model and differential susceptibility framework predicted heightened vulnerability in individuals with mental disorders. However, surprisingly, the pandemic’s impact on pre-existing conditions appeared less pronounced than anticipated. While some research showed brain structural changes in response to traumatic life events, data on neurobiological consequences of restriction measures during the COVID-19 pandemic were scarce. Studies observed decreased grey matter volume in various brain regions following prolonged social isolation, correlated with reduced BDNF levels. BDNF, crucial for neuroplasticity, learning, and physical activity, is frequently used as a peripheral marker in this context. This study aimed to fill the research gap by investigating multimodal changes during a real-world stressful event.

This prospective, quasi-experimental study utilized participants from previous studies, including those with recurrent MDD and healthy individuals. Exclusion criteria included SARS-CoV-2 infection, severe somatic/neurological disorders, MRI contraindications, and pregnancy/breastfeeding. Baseline assessments included the Structured Clinical Interview for DSM-IV (SCID-IV), physical examination, and blood tests. Participants underwent three measurement phases (September 2020 – July 2021) involving MRI scans (3 Tesla MAGNETOM PRISMA scanner), blood draws (sBDNF analysis via ELISA), and psychometric assessments (BDI-II and PSQ-20). FreeSurfer 7.1 software processed structural MRI data to reconstruct cortical surfaces and parcellate brain regions. Data were analyzed using linear mixed models to assess the effects of lockdowns on brain measures, sBDNF, and psychometric scores. Models included time point, group (MDD vs. healthy), hemisphere, ROI, age, sex, and medication as fixed effects, with subject as a random intercept. Interaction effects were analyzed, and non-parametric tests were used for non-normally distributed data. Pearson correlations assessed relationships between sBDNF, brain volumes, and psychometric scores.

The final sample included 28 healthy individuals and 18 patients with recurrent MDD. Linear mixed model analyses revealed no significant changes over time in cortical thickness, surface area, or subcortical volumes of a priori brain regions (amygdala, hippocampus, anterior cingulate cortex, prefrontal cortex) or in an exploratory analysis of all brain regions. There was also no interaction between time and group for these brain measures. A significant main effect of group was found for BDI-II scores (F=30.89, *p* < 0.001), indicating higher depression scores in the MDD group compared to the healthy group, but no change in these scores over time. The MDD group showed clinically relevant scores for depression and stress, as determined by PSQ-20 subscales. However, there were no significant main effects of time or interaction effects of time × group on any PSQ-20 subscales. Linear mixed models for sBDNF showed no main effect of time or interaction between time and group. No significant correlations were found between sBDNF levels and brain measures, or between sBDNF levels and psychometric scores.

This study found no significant changes in brain structure, neuroplasticity, or psychometric parameters in response to COVID-19 lockdowns in both MDD patients and healthy controls. This contrasts with initial hypotheses and some prior research suggesting negative mental health effects during the pandemic. The absence of exacerbation in depressive symptoms in the MDD group might suggest adaptational processes and resilience, potentially influenced by pre-existing treatment, structured routines, increased access to mental health services, and perhaps less severe impacts in Austria. The findings are consistent with some recent meta-analyses showing limited impact on mental health after the initial lockdown phase. However, the clinically relevant stress and depression scores in the MDD group highlight the existing burden of illness.

This study, the first to investigate the neurobiological effects of multiple COVID-19 lockdowns in Austria, found no significant changes in brain structure, neuroplasticity, or psychometric parameters in recurrent MDD patients or healthy controls. The absence of worsening mental health symptoms may indicate adaptational processes and resilience. These findings support the need for continued longitudinal studies and further research to understand the long-term impact of lockdowns and the factors contributing to resilience during pandemics. Future research should also focus on other potentially influential factors such as compliance with restrictions, lifestyle changes, and access to mental health services.

The study's naturalistic design, with the first measurement five months post-first lockdown, might have missed the immediate impacts of early adaptational processes. The absence of a non-lockdown control group limits the interpretation of findings as solely due to lockdowns. The relatively small sample size might have reduced the power to detect subtle changes in brain structure and neuroplasticity. A lack of data on compliance with restrictions, lifestyle changes, and digital phenotypes also impacts the interpretation. The inclusion criteria may have biased the sample towards individuals with access to regular care and treatment optimization, which might influence outcomes.