INTRODUCTION
Electroconvulsive therapy (ECT) is a method of treatment that was invented by Bini and Cerletti in 1938 and has been primarily used in the treatment of psychiatric illnesses [1]. This treatment modality involves inducing brain convulsions with the use of electrodes while a patient is under anesthesia. It is most commonly used to treat patients diagnosed with a severe episode of major depressive disorder (MDD), but those with bipolar disorder (BD), schizophrenia, catatonia, schizoaffective disorder or neuroleptic malignant syndrome may also benefit therapeutically [2]. A reduction in suicidal ideation and intentions has observed been observed in patients following ECT, which recommends it as a first-line method for rapid intervention in life-threatening conditions [3]. No absolute contraindications to ECT have been listed by the American Psychiatric Association; however, certain medical conditions have been identified as risks for complications following a session. For this reason, recommendations made to anesthetists include a thorough review of patient history and physical examination to note any signs of cardiopulmonary disease, history of stroke or intracranial tumors. These underlying conditions must be ruled out as each of them is associated with an increased risk related to ECT [4].
According to the guidelines of the National Institute for Health and Care Excellence, ECT is recommended only for the rapid and short-term improvement of severe symptoms, when other therapeutic options have failed and/or when the condition is considered potentially life-threatening. Ultimately, the decision to use ECT should be based on a documented assessment of the risks to the patient and potential benefits, which requires internal medicine, neurology and anesthesia consultations in order to make a decision about the best choice of anesthetic. Such an assessment must take into account, first and foremost, current comorbidities, possible side effects – including cognitive impairment – and the risks associated with offering no treatment [5]. Anesthetics used in ECT include propofol, ketamine, etomidate, thiopental and methohexital. With that said, ketofol and etomidate have been shown to be superior to propofol or thiopental in this context [6]. Methohexital is the anesthetic of choice in modern ECT practice, as it has the smallest effect on seizure properties, as compared to other barbiturates. Succinylcholine is used as an efficient muscle relaxant, lowering the chances of physical injury during a session; this is due to the fact that it does not need to be reversed, as well as its quick onset and quick-acting effects. Anticholinergic drugs are often used to counter symptoms like hypotension and bradycardia. An example is glycopyrrolate, which is now preferred over atropine, as the latter crosses the blood-brain barrier and causes adverse cognitive effects [7]. A typical course of ECT usually includes between six and twelve sessions, with pharmacological treatment given twice or three times weekly. To prevent relapse of mood disorders post-ECT, suggestions are to organize maintenance ECT weekly for 1-3 weeks, then taper off according to the patient’s requirements. In cases of depression, ECT may be continued after the first course if no adequate response to pharmacotherapy is observed, or if the results of ECT have been especially beneficial to the patient [8-11].
As mentioned earlier, ECT is the most effective treatment for MDD, but its use is hampered by limited availability and our lack of a complete understanding of the mechanism of action. One theory is the involvement of g-aminobutyric acid (GABA) with anticonvulsant effects, with an observed increase in seizure threshold and reduction in duration after repeated sessions of ECT [12]. One result of ECT is an increase in GABA concentrations, with an improvement in serotonergic function [13]. Moreover, studies have shown that ECT can enhance activity in the frontal cortex and the networks responsible for executive control. Evidence also indicates that it can modulate dysregulated neural circuits through repeated electrical fields or induced seizure activity, improving regulation at various biological levels, leading to more adaptive circuit functioning and the achievement of a state of euthymia [14]. Following ECT, an increase in gray matter volume has been repeatedly observed in various brain regions in patients with depression and schizophrenia. These changes may be related to the pathophysiological mechanisms characteristic for ECT [15].
Due to the incompletely understood mechanism of ECT, markers that can confirm efficacy and determine response to ECT are still being sought. Molecular markers under consideration are neuron-specific enolase (NSE), S-100 protein and Aβ peptides, the levels of which increase because of neurological disturbances, and it has been speculated that seizures occurring during ECT sessions can cause toxicity in and potentially the death of neuronal and glial cells [16, 17]. Levels of NSE, S-100 protein or Aβ peptides, however, did not rise significantly after ECT, with levels remaining stable or increasing transiently, suggesting that the cognitive side effects associated with ECT are not caused by neuronal cell damage [16-19]. In addition, predictors of response to ECT have been sought for polymorphisms within genes encoding the dopamine receptor 2 (DRD2) gene, dopamine receptor 3 (DRD3) gene, brain-derived neurotrophic factor (BDNF), catechol-O-methyltransferase (COMT), serotonin transporter (5-HTT), 5-hydroxytryptamine 2A receptor (5-HT2A) and norepinephrine transporter (NET), which appear to predict a good response to ECT [20].
In seeking to determine the neurobiological mechanisms of electroconvulsive effects and their efficacy, researchers have increasingly used functional brain imaging, enabling the identification of specific brain areas whose activity can be measured before and after ECT [21]. One of the proposed neuroimaging tests is functional near-infrared spectroscopy (fNIRS). This is a non-invasive optical imaging technique that uses near-infrared light to measure changes in oxygenated and deoxygenated hemoglobin concentrations in cortical tissue. Using the properties of light in terms of its absorption and scattering in biological tissues, this method detects changes in blood oxygenation associated with neuronal activity.
The present study was designed to determine the role of fNIRS as a potential predictor of response to treatment with ECT and its efficacy.
METHODS
To provide up-to-date information on the mechanism of ECT and fNIRS, their methods of application, purpose in the treatment of specific psychiatric conditions, and advantages and disadvantages, the following databases were searched: Web of Science PubMed, Google Scholar, and Scopus. In addition, existing research on the changes visible in fNIRS and ECT in patients with depression or schizophrenia was reviewed. Key words used were “fNIRS or functional near-infrared spectroscopy” and “ECT or electroconvulsive therapy” resulting in four relevant articles – two randomized control trials, one longitudinal study and one case report. All of these were articles that enabled us to address the aim of the study, published between 2014-2024.
NEUROIMAGING AS A POTENTIAL PREDICTOR OF RESPONSE TO ECT TREATMENT AND ITS EFFICACY
With the development of neuroimaging there are now a number of methods available to the study of ECT-induced brain changes. These neuroimaging techniques include positron emission tomography (PET), single photon emission computed tomography, structural magnetic resonance imaging, magnetic resonance spectroscopy, functional magnetic resonance imaging (fMRI), and fNIRS [22, 23]. The collective findings from the use of multiple neuroimaging techniques point to the neurobiological mechanism of ECT, which can regulate the functional activity of the brain and the structural plasticity of neurons, as well as balance the brain neurotransmitters. This feature is thought to be responsible for the therapeutic effect of the use of ECT [22].
In 1990, a method was developed to represent changes in the concentration of deoxyhaemoglobin caused by activity or the spontaneous modulation of the neuronal metabolism. This method used fMRI of blood oxygen level-dependent (BOLD) [24]. Studies using fMRI in ECT therapy have demonstrated long-lasting, interictal changes in neuronal activity in multiple brain areas are associated with functional neuroanatomical models of mood disorders [25]. In addition, a subtype of fMRI is functional resting-state magnetic resonance imaging (rs-fMRI/), based on spontaneous BOLD-dependent signal fluctuations that occur simultaneously in different brain areas without the subject being explicitly asked to perform a task [26]. Currently, rs-fMRI is widely used to investigate mechanisms of psychiatric disorders, including depression. Studies using rs-fMRI suggest that the right dorsal anterior cingulate cortex, bilateral superior medial frontal cortex and right precuneus play an overarching role in the response to ECT of depressed patients, given the dynamic local brain activity involved, indicating that the dynamic variability of low-frequency fluctuation amplitude (dALFF) may be useful in further understanding the mechanisms of the antidepressant effects of ECT [27].
However, each of the neuroimaging methods currently in use has limitations relating to resistance to motion artifacts, depth of penetration in the brain, spatial and temporal resolution, range of applications, range of patient groups, and the cost and transportability of devices [28].
The advantages of fMRI include its non-invasive, repeatability, high availability, very high spatial resolution and measurement of the whole brain. However, fMRI has low resistance to movement, requires a lying position during the examination, and is expensive and noisy. In addition, there are limitations regarding metal in the body of the person being examined, and some patients also fear the examination because, e.g., of claustrophobia [28].
MECHANISM OF fNIRS
Functional near-infrared spectroscopy is a brain monitoring technique which estimates the concentration of hemoglobin from changes in absorption of near infrared light [29].
In 1977, Frans Jöbsis showed that the transparency of brain tissue is relatively high in the near-infrared range, i.e. 700-900 nm of the optical spectrum, allowing us to record the average hemoglobin-oxyhemoglobin balance over the period, using non-invasive infrared transillumination spectroscopy [30]. Following in vivo animal studies, the technique was first used in 1985 to study brain oxygenation in sick neonates [31]. In 1984, the first quantitative measurement of multiple haemodynamic parameters of oxygenation in ill neonates was published, thanks to work started on several NIRS devices by David Delpy. The tests recorded the increase in oxygenated (HbO), deoxygenated (dHb), total hemoglobin and cerebral blood flow [32].
fNIRS systems consist of detectors and light sources placed on the scalp to quantify relative changes in deoxygenated and oxygenated haemoglobin in different areas of the cortex. The light source can be shared by multiple detectors and vice versa, creating multiple detection channels without having to add components. Control of imaging depth is possible by measuring the distance between the light source and the detector, which can range from 5 mm to 10 mm, which is referred to as a short channel, and from 20 mm to 60 mm, which is referred to as a long channel. The short channel is used to detect and correct for shallow physiological signals or movement artefacts not originating in the brain, while the long channel is used to detect haemodynamic activity in the brain [33].
APPLICATION OF fNIRS IN THE TREATMENT OF PSYCHIATRIC DISORDERS
Studies using fNIRS in patients have mainly focused on the task-induced activation of different brain areas, particularly the prefrontal cortex (PFC). The increase in HbO and the relative decrease in dHb in fNIRS after activation are indicative of an increase in cerebral blood volume in the respective brain area, which is due to local vasodilatation. As the PFC is responsible for various cognitive functions such as working memory, planned motor responses and decision-making, PFC dysfunction correlates with impairment in social life, memory loss and affective disorders in some neuropsychiatric diseases such as schizophrenia, depression, autism and Alzheimer’s disease [34].
Autistic children exhibit modified PFC activity, characterized by hyperactivity of the PFC and weakened connectivity in the right frontal area [35]. Results from both fNIRS and fMRI studies have shown that in attention deficit hyperactivity disorder (ADHD), during executive function testing tasks, there is reduced activity in the right PFC [36]. By monitoring progress in working memory tasks, fNIRS shows that brain activity is linked to task performance [37]. Brain function in autistic patients was analyzed pending task performance using fNIRS. The study showed that as the disease progressed, the concentration of HbO in a given brain area increased from mild cognitive impairment significantly up to moderate and even severe dementia [34, 38].
fNIRS in depression
In a study that assessed patients with BD or MDD, oxyhaemoglobin (oxy-Hb) levels were recorded prior to and post-ECT; near infrared spectroscopy (NRS) was used to compare the changes in oxy-Hb levels pending a verbal fluency task (VFT). In the right ventrolateral prefrontal cortex, oxy-Hb levels were associated with a lower severity of depressive symptoms [23]. fNIRS may serve as a trait marker, suggesting that it reveals lasting changes in brain function in individuals who have experienced depressive episodes, also after symptoms have subsided. This differs from a state marker, which is associated with the patient’s current clinical condition. The use of fNIRS shows persistent hypofrontality in patients with MDD who had not previously taken antidepressant medications during the treatment. Measurement of oxy-Hb levels during VFT indicated a similar level of activation before and after treatment, despite a marked reduction in depressive symptoms [39].
fNIRS studies have revealed a significant impact of PFC activity on various cognitive functions and its susceptibility to disorders associated with bipolar depression (BD-D). Reduced activity in the left dorsolateral PFC (dlPFC) related to flaws in executive functions, and significant differences were also observed in the right PFC. These results are in line with previous studies that demonstrated lower PFC activation in individuals at risk for psychosis. fNIRS signals from the right PFC during working memory tasks may prove helpful in diagnosing BD [40]. Hemodynamic changes in the bilateral prefrontal cortex were inversely related to the severity of depressive symptoms [41].
fNIRS can be useful in differentiating between anxious and non-anxious depression. The signal for hemodynamic activation was increased in the dorsolateral prefrontal cortex (DLPFC) of right side and the right frontopolar cortex (FPC) in the anxious depression group. There are major differences between patients with and without anxious depression and in the activation patterns of the right DLPFC and right FPC areas. Moreover, the right prefrontal cortex area seems promising as a brain region in which to assess the severity of anxious depression [42].
fNIRS in schizophrenia
Changes in the prefrontal brain area are associated with working memory and deficits in executive functions and in patients with schizophrenia. Most fNIRS studies related to schizophrenia focus on analyzing dysfunctional regions of the cerebral cortex that may contribute to deficits in executive functions. The most commonly observed finding is the hypoactivation of the PFC in patients performing cognitive tasks. Additionally, there is a suggestion that results obtained through fNIRS could serve as a biomarker for schizophrenia [43-45].
Other psychiatric disorders
Studies conducted using fNIRS have revealed hypoactivity in the frontal lobe regions of patients with schizophrenia and MDD during tasks related to verbal fluency, compared to control groups. Additionally, individuals with schizophrenia and MDD exhibited hyper- and hypo-connections between various brain areas during resting state. In patients with obsessive-compulsive disorder (OCD), a decrease in blood flow was observed in the symmetric regions of the lower prefrontal cortex when compared to their healthy peers. Dysfunction in the prefrontal cortex, characterized by both hypo- and hyper-connections during resting state and hypoactivity during various cognitive tests (such as the Stroop test and verbal fluency test), has been the subject of extensive research in groups of patients with various significant neuropsychiatric disorders, including schizophrenia, MDD, post-traumatic stress disorder (PTSD), BD, and ADHD. It can be concluded that the topographic distribution of functional abnormalities in these patients may reveal specific modules related to different disorders [46]. fNIRS systems offer the potential for precise quantification and parameterization of abnormalities in prefrontal lobe functioning, offering valuable support for the objective classification and diagnosis of principal psychiatric disorders that often share overlapping behavioral symptoms which make differentiation difficult [46, 47]. Research conducted using fNIRS suggests that the prefrontal cortex should be considered a crucial neuroanatomical area in which abnormalities are associated with significant psychiatric disorders such as schizophrenia, MDD, BD, generalized anxiety disorder, and borderline personality disorder. Furthermore, an underdevelopment of the right orbitofrontal cortex may contribute to increased susceptibility to psychosis and to the core symptoms of schizophrenia [48].
A summary of the use of fNIRS in the treatment of mental disorders is presented in Table 1.
APPLICATION OF FNIRS IN ECT
The use of fNIRS makes it possible to determine changes in cortical functional responses to cognitive tasks, and the use of fNIRS in the assessment of patients who have been treated with ECT appears to be extremely important given that the acute effects of the latter on brain function are still unclear and little is known about the extent and time course of ECT-induced functional brain changes observed during the conduct of cognitive tasks [23].
Table 1
Application of fNIRS in mental disorders
The few studies to date on the use of fNIRS in the assessment of patients’ cognitive function before and after ECT suggest that, prior to being treated with it, depressed patients show lower oxy-Hb values measured with fNIRS, and consequently have bilaterally lower frontal oxy-Hb responses to cognitive tasks compared with healthy subjects [23, 49, 50].
In contrast, there is a change in oxy-Hb values after ECT compared to the situation prior to treatment. However, the results obtained by different groups of researchers have not been consistent. Downey et al. [49] found that ECT further reduced the frontal HbO signal during the verbal fluency task (VFT), with no significant effect during the N-Back task, although it was numerically lower in the latter measurement. Also, Anderson et al. [50] found that electroshock further reduced prefrontal cortex hemodynamic responses on the VFT task in the absence of behavioral effects in the corresponding neuropsychological task. Quite different results were obtained by Hirano et al. [23], who found a significant increase in oxy-Hb values in bilateral frontal cortex during VFT after ECT in a group of patients. A reduction in the severity of depression was significantly correlated with an increase in oxy-Hb values in the right ventral-lateral prefrontal cortex after ECT. Lower oxy-Hb values at the start of the study were generally normalized at the endpoint. In a case report of a 30-year-old with schizophrenia who was imaged with fNIRS before and after ECT, a reduction in frontal cortex activity was observed before the start of the therapy. The evaluation of HbO and Hb fluctuations from one channel showed that frontal cortex hemodynamic activity was opposite to normal at the start of the task. In addition, abnormal patterns of functional asymmetry were evident, with the pattern of prefrontal cortex activity being opposite to the pattern at task onset. After a series of 15 treatments the pattern of activity was reversed, to a more normative direction [51].
The above discrepancies between these studies may have been due to the different group sizes in each study (n = 11 vs. n = 12 vs. n = 30 vs. n = 1), the number of ECTs performed or the interval after which cognitive tests measured with the fNIRS were performed. In the study by Downey et al. [49], fNIRS was used two days after the completion of 4 ECT series, as was that of Anderson et al. [50], while in the study by Hirano et al. [23]. fNIRS was used within 2 weeks of the completion of the series with a mean number of ECT sessions of –9.6 ± 2.3. It has long been known that cognitive function associated with ECT is limited to the first 3 days after treatment. By 15 days after ECT, processing speed, working memory, subsequent memory and some aspects of executive function improve above baseline [52]. In addition, it has been found that patients with levels higher educational and occupational attainment across the lifespan may experience less retrograde amnesia and faster recovery of orientation after ECT [53].
ADVANTAGES AND DISADVANTAGES OF fNIRS
Compared to other neuroimaging methods – PET, fMRI, single photon emission computed tomography (SPECT), magnetoencephalography (MEG) and electroencephalography (EEG) – fNIRS has several important advantages. One is that, at this point, the fNIRS system has a wear-and-go design, making it small in size and portable. The large equipment and high costs of fNIRS systems previously hindered the widespread adoption of them. Modern hardware can be easily configured in a way such that larger number of fNIRS channels are usable and allows easy collection of data during routine activities in naturalistic settings [33, 54]. In addition, fNIRS is characterized by good spatial and temporal resolutions (with maximum sampling rates approaching 100 Hz) and resistance against motion artefacts. Other advantages are its low cost, high degree of tolerance by patients due to the non-invasiveness of the method, and its compatibility with other therapeutic devices (e.g. EEG) [54]. However, current fNIRS systems cannot probe brain activity below the level of the cortex and their signal-to-noise ratio and spatial resolution are lower than those of fMRI [36]. Brain activity can also be measured in clinical settings thanks to fNIRS; it could be used as a potential biomarker of the safety and efficacy of ECT [55, 56].
APPLICATION OF FNIRS IN CLINICAL PRACTICE
fNIRS is not used directly to make a diagnosis, as it is rather an assistive device that reads functional brain activity and indirectly reports abnormalities in brain function. It can be used in monitoring the depth of anesthesia and levels of patient sedation in Alzheimer’s disease, schizophrenia, dyslexia, addiction, ADHD, epilepsy or depression. With fNIRS, functional haemodynamic activity is distinguished by certain behavioral tests, which constitutes its use as a predictive modality [29]. Compared to healthy individuals, people with psychiatric disorders show significantly reduced changes in HbO and significantly fewer words produced during the verbal fluency (VF) test. In most psychiatric disorders, changes in HbO are more sensitive than changes in deoxyhaemoglobin and VF performance in the detection of a psychopathology. Thus, the fNIRS-VFT criterion could aid in understanding, detecting and differentiating psychiatric disorders, and has the potential to develop accessible neuroimaging biomarkers for various psychiatric disorders [57].
SUMMARY AND CONCLUSIONS
Higher oxy-Hb values are achieved in healthy subjects than in ill subjects, so fNIRS may be used to estimate the therapeutic effect of ECT on patients. Therefore, further research into fNIRS used to assess cognitive function in patients after ECT is warranted to exploit its potential utility as a biomarker. It is crucial to conduct studies that could evaluate changes in the deoxy-Hb and oxy-Hb absorption spectra measured with fNIRS, including cognitive changes measure in relation to time passed post-ECT treatment, and according to the number of ECT treatments performed. It is worth investigating the use of fNIRS on a significant number of patients and in different disease entities such as schizophrenia, depression, BD and anxiety disorders, while maintaining homogeneous groups.
