Major depressive disorder (MDD) and anxiety disorders, which are highly comorbid psychiatric disorders, may be influenced by the alterations in brain region interactions within different networks. It is unknown whether or not treatment with antidepressants may revert these changes back to normal. Authors of a study published in Psychiatry Research examined the dynamic reconfigurations in individuals with depression, anxiety, or both, with and without antidepressant use.
Baseline data were gathered from participants aged 18 to 57 years who met all DSM-IV criteria for half-year diagnosis for a MDD with or without an anxiety disorder, such as panic disorder, social anxiety disorder, or generalized anxiety disorder. Participants who were undergoing pharmacological treatment were only included if their use of either serotonin reuptake inhibitors (SSRIs) or serotonin-norepinephrine reuptake inhibitors (SNRIs) was considered stable (i.e. daily use for over half of the days in the past month). In addition, these participants had to comply with therapy with or without infrequent use of benzodiazepines that did not exceed 2 10-mg doses twice per week, without use within 48 hours prior to screening.
The study authors enrolled 207 participants, of which 150 met the DSM-IV criteria of current MDD (n = 52), 1 or more anxiety disorders (ANX; n = 46), or both (comorbid group or COM; n = 52). The remaining 57 had no current or prior DSM-IV diagnosis and were considered to be controls. Using magnetic resonance imaging (MRI), the investigators assessed the differences between the controls and those in the MDD, ANX, and COM groups, and the differences between individuals with a depressive or anxiety disorder who are using antidepressants with those who are not and controls.
If there were significant differences between groups, additional validation analyses were performed and antidepressant use was added as a covariate. In addition, the model was reran using only individuals that have not used antidepressants within the past 3 years then associations with symptom severity scores were calculated.
Of the 150 participants with a depressive and/or anxiety disorder, 49 were using antidepressants and the remaining 101 were non-users. Most antidepressant non-users were younger than the controls, and the majority used selective serotonin reuptake inhibitors (SSRIs; n = 38; paroxetine: n = 20, citalopram: n = 10, fluoxetine: n = 3, fluvoxamine: n = 3, sertraline: n = 1, and escitalopram: n = 1) while the others used venlafaxine, a serotonin and noepinephrine reuptake inhibitor (SNRI; m = 11).
According to the results, there was no difference between healthy controls, MDD, ANX, and COM in network reconfigurations across the whole brain for both global promiscuity and global flexibility. In addition, results did not change when controlling for antidepressant use and no differences were observed when specifically looking at individuals who had not regularly taken antidepressants within the past 3 years. There was no relationship observed between global network configuration and the severity of anxiety symptoms; however, a positive relationship between depressive symptoms and promiscuity was observed but lost when controlling for antidepressant use.
Key Takeaways
- Dynamic Brain Network Changes: Major depressive disorder (MDD) and anxiety disorders, often occurring together, may be linked to altered interactions among brain regions within various networks; however, these alterations do not seem to differ significantly between individuals with MDD, anxiety, or both, compared to healthy controls.
- Effect of Antidepressants: Antidepressant use appears to influence global brain network characteristics, with users showing higher levels of network promiscuity and flexibility compared to non-users and the control group. These differences do not seem to correlate with the type or dosage of antidepressants used.
- Specific Network Effects: Antidepressant users showed increased cohesion and disjointedness in brain networks compared to non-users and controls. The sensorimotor network in antidepressant users also displayed elevated promiscuity and flexibility, suggesting a specific effect of antidepressants on certain brain regions; however, the clinical implications of these findings require further investigation.
Furthermore, global promiscuity and flexibility were significantly different between antidepressant users, non-users, and controls, with both measures elevated in antidepressant users compared to non-users (promiscuity: β = 0.042, 95 % CI = [0.016, 0.068], p = 0.002; flexibility: β = 0.024, 95 % CI = [0.009, 0.039], p = 0.002) and controls (promiscuity: β = 0.060, 95 % CI=[0.029, 0.092], p < 0.001; flexibility: β = 0.032, 95 % CI = [0.014, 0.050], p < 0.001); however, there was no difference observed between non-users and controls. Despite SNRIs showing the largest correlation coefficients, the defined daily dose in antidepressant users was not correlated with either global promiscuity or flexibility.
Global cohesion and disjointedness were both different between antidepressant groups, with both levels increased in antidepressant users versus non-users (cohesion: β = 0.020, 95 % CI = [0.006, 0.034], p = 0.006; disjointedness: β = 0.004, 95 % CI = [0.001, 0.007], p = 0.006), and for cohesion compared to the control group (cohesion: β = 0.029, 95 % CI = [0.012, 0.046], p = 0.001; disjointedness: β = 0.003, 95 % CI = [0.000, 0.006], p = 0.063). In addition, there weren’t any differences observed between non-users and controls, like promiscuity and flexibility.
Further, global and sensorimotor network (SMN) were elevated in antidepressant users than non-users (promiscuity: β = 0.084, 95 % CI = [0.016, 0.153], p = 0.016; flexibility: β = 0.042, 95 % CI=[0.003, 0.081], p = 0.033) and controls (promiscuity: β = 0.171, 95 % CI = [0.089, 0.253], p < 0.001; flexibility: β = 0.104, 95 % CI = [0.057, 0.151], p < 0.001); however, SMN flexibility and promiscuity were both higher in non-users compared to the control group. The investigators note that the defined daily dose was not correlated with SMN promiscuity or flexibility for both SNRIs and SSRIs.
A limitation of this study is the cross-sectional design, which resulted in the inability to properly investigate naïve participants both before and after first antidepressant treatment. The authors also note that patients being scanned in different centers, however, the distribution of centers was comparable between groups and differences were controlled for the analyses. Further, the investigators acknowledge that taking ongoing cognitive processes into consideration is not practical using resting state fMRI.
Reference
Broeders TAA, Linsen F, Louter TS, et al. Dynamic reconfigurations of brain networks in depressive and anxiety disorders: The influence of antidepressants. Psychiatry Res. 2024;334:115774. doi:10.1016/j.psychres.2024.115774
