Major depressive disorder (MDD) is a common psychiatric disorders (1), characterized by symptoms including depressed mood or anhedonia (2), affecting approximately 6% of the population worldwide (3). About 30% of this disease becomes treatment-resistant depression (TRD) among patients with major depression.

It is estimated that 2–3% of the population suffer from obsessive–compulsive disorder (OCD) (4), characterized by distressing intrusive thoughts (obsessions) and often time consuming repetitive behaviors (compulsions) (5). Apart from obsessions and compulsions, substantial comorbidity, anxiety disorders, mood disorders, impulse-control disorders, and substance use disorders are common in OCD, with patients feeling confused in relationships and social functioning domains (6). However, 10% of patients are not relieved with sufficient drugs and long duration (7).

In 1999, Deep brain stimulation (DBS) targeting the anterior limb of the internal capsule (ALIC) was found to be beneficial for intractable OCD (8), and many reports about other targets of DBS in patients with OCD have been published. Moreover, DBS had been proven to produce clinical benefits in patients with TRD since 2005 (9). With direct access to the subcortical activity, local field potentials (LFPs) has provided valuable insight into disease mechanisms and cognitive processes. It is known that aberrant activity in cortico-basal ganglia-thalamo-cortical loops may be involved in the symptoms, but the specific relationship between LFPs and OCD is unclear. Likewise, the relationship between LFPs and MDD is unknown.

This review aimed to summarize the changes of LFPs in patients with OCD and MDD. In addition, we expect to point out how electrophysiology research in patients with DBS will continuously offer new information with mechanism when combined with LFPs.

As an aspect of neural activity in the brain, LFPs are increasingly used as a reflection of the ongoing transmission through neural networks (10), which can be recorded using metal, glass electrodes, or silicon probes inserted in the deep brain areas during or after operation (11). In fact, LFPs have been neglected for a few decades because in vivo neurophysiological research focused mostly on isolating action potentials from individual neurons. They are of interest to researchers who study cortical function in recent years, because LFPs provide a unique window to integrative excitatory and inhibitory synaptic processes for neural population activity (12, 13).

According to the changes of frequency, LFP bands were defined as follows: delta (0–3 Hz), theta (4–7 Hz), alpha (8–12 Hz), beta (13–35 Hz), gamma (31–200 Hz), and high-frequency oscillation (>200 Hz) (14). Different frequency-band oscillations reflect diverse neural processing pathways and contributions of transmembrane current from cellular activities. Moreover, LFPs can provide stable signals for long-term chronic experiments or clinical applications.

Therefore, LFPs provide more chances to explore the electrophysiological mechanism of psychiatric disorders. It is known that the presence of abnormal beta bursts is significantly correlated with the degree of motor impairment and the severity of rigidity/bradykinesia in Parkinson’s disease (PD) (15). However, there are few reports on the electrophysiological characteristics of OCD and MDD.

Different neuronal oscillations have always been considered as brain organizers, coordinating brain areas into networks in the normal or disordered state (35–38). To our knowledge, this was the first review to investigate LFP oscillations in patients with OCD and MDD. In this review, low frequency bands, especially the theta band, appeared more positively involved in modulating psychiatric-condition-related networks in patients with OCD (Table 1). However, such involvement may also be observed in levodopa-induced dyskinesias, dystonia, Tourette’s syndrome, schizophrenia, and attention deficit hyperactivity disorder (27). Meanwhile, changes in other frequency bands have also been reported. Since frequency bands are interrelated and interact with each other in psychiatric-disorder-related networks, changes in one frequency band activity may also influence the activity of other frequency bands. Here, critical caution is needed when considering the theta band as a biomarker in patients with OCD.

In clinical studies, LFP findings in patients with MDD involved different frequency bands, unlike animal study observations. No clear conclusions could be drawn about the direction of the difference due to considerable inconsistencies between the study design and methodology. We concluded that the neuronal architecture of the mouse brain is different from that of the human. For example, the human hippocampus has a larger size and more neurons than rodents. Therefore, humans would have a longer period than animals to bring more cell assemblies to be linked synaptically in various brain processes such as memory and learning (39), showing lower frequency oscillations than rodents. Some researchers conjecture that the frequency of a neural oscillation has an inverse relationship with the size of a neuronal network (40–42).

The SCC LFP recordings in the bilateral hemispheres are asymmetric; left-sided stimulation resulted in widespread changes in the frequency bands, and the right-side stimulation effects were restricted to the beta and gamma bands (25). Similarly, the HB LFP recordings in the left hemisphere showed high-beta oscillation with depressive symptoms, while gamma oscillation was distinct in the right hemisphere (22). This lateralized alteration implied an asymmetric role of the bilateral hemispheres in modulating psychiatric-disorder-related networks, identical to the laterality of brain speech function.

However, current research has produced strikingly different results, as they were obtained through various targets and under different physiological states, which impeded arriving at definite conclusions. Furthermore, some studies did not use stimulation with the targets, and the stimulation parameters were not unified. As LFP is a partly ambiguous signal involving multiple neuronal processes, careful analytical modeling and empirical considerations are required to interpret the derived information.

Based on the findings, we conclude that synchronization in multiple bands, rather than just abnormalities in a single frequency band, may play an essential role in modulating OCD and MDD symptoms. In the future, we might consider modulating networks that support specific symptom patterns instead of focusing on a single, optimal gray matter target (43).

This mini-review was not comprehensive; however, it summarized the recent advances in LFPs of MDD and OCD. As mentioned above, LFPs provided more chances to understand the electrophysiological characteristics and explore the potential mechanisms. Low-frequency activity seemed closely related to OCD symptoms, whereas LFP findings in patients with MDD were complicated. While MDD in human had a close relationship with the alpha or beta frequency or gamma oscillations. All different frequency bands seemed to participate in MDD and OCD networks. In the future, other multiple measures with long-term LFP recordings in various physiological states (rest, sleep, and task states) could be applied in patients with OCD or MDD.

WZ, BX, and WW proposed the conception and design of the study. WZ and BX conducted the literature search, study screening, data extraction, and quality assessment. WZ, BX, YW, LX, and WW were involved in the analysis and interpretation of data. WZ and BX drafted the manuscript and all authors revised it critically under the guidance of WW. All authors contributed to the article and approved the submitted version.

This work was funded by the 135 Project of Outstanding Development of West China Hospital, Sichuan University (Grant number ZY2017307).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.