Major depressive disorder is a mental health condition characterized by depressed mood and diminished interest, causing various mental and physical symptoms such as decreased appetite and concentration. These symptoms impair daily functions, including social and occupational functions. The World Health Organization reported that 280 million people worldwide suffer from depression, and that more than 700,000 individuals commit suicide each year as a result of depression [1]. Therefore, attempts have been made to predict and prevent suicide in patients with depression.

Serotonin concentration is associated with suicide in patients with major depressive disorder [2]. People who attempt suicide have been found to have lower levels of serotonin metabolites in their cerebrospinal fluid [3]. However, there are limitations to using results of cerebrospinal fluid tests to predict suicide in a clinical setting. Thus, some studies have been conducted to investigate the use of loudness dependence of auditory-evoked potentials (LDAEPs) as an alternative to measuring serotonin activity in the cerebrospinal fluid in clinical practice. LDAEP, a representative neurophysiological indicator, is related to serotonin activity [4] and is expected to be suitable for the classification of depression subtypes, evaluation of depression severity, exploration of psychopathology, and prediction of treatment responsiveness [5-8]. However, the association between LDAEPs and suicidality is unclear [9]. This is because there are limited studies on LDAEPs and suicidality. In addition, the number of participants included in the existing studies has not been large enough to yield significant evidence. The aim of this study was to investigate the relationship between LDAEPs and suicidal ideation or suicide attempts in a large group of patients with depression.

Study Design and Participants

This was a retrospective study of psychiatric outpatients aged 16−74 years who were diagnosed with major depressive disorder using the Diagnostic and Statistical Manual of Mental Disorders 5th edition criteria at Inje Paik Hospital from 2014 to 2021. The demographic and clinical data of the patients, including data on suicidal ideation and previous suicide attempts, were obtained from clinical interviews and medical records. The Beck Depression Inventory (BDI) [10] and Hamilton Depres-sion Rating Scale (HAMD) [11] were used to assess the severity of depression. The Barratt Impulsiveness Scale (BIS) [12], Hamilton Anxiety Rating Scale (HAMA) [13], and Beck Scale for Suicidal Ideation (BSS) [14] were used for the evaluation of impulsiveness, anxiety, and suicidal ideation, respectively. The BDI is a valid instrument used to determine the severity of self-reported depression. The HAMD is a scale commonly used for the evaluation of depression. The BIS is a frequently used self-report scale that measures several dimensions of impulse control. The HAMA is a scale used for rating the severity of anxiety, and the BSS is a self-report scale used to assess suicidal ideation. The protocol of this study was approved by the Ethics Committee of Inje University Ilsan Paik Hospital (ISPAIK 2022-09-043).

Electroencephalography Methods

Each patient was seated in a quiet room in a comfortable chair. The auditory stimulation consisted of 1,000 stimuli with interstimulus intervals ranging from 500 to 900 ms. Using headphones (MDR-D777; Sony), randomized tones of 1,000 Hz were presented at five intensities (55-, 65-, 75-, 85-, and 95-dB sound pressure levels) for 80 ms. The stimuli were created using the E-Prime software (Psychology Software Tools). The electroencephalo-graphy (EEG) data were collected from 64 scalp sites fitted with silver/silver-chloride electrodes using an Auditory Neuroscan NuAmp amplifier (Compumedics USA), in accordance with the international 10−20 system (imped-ance < 10 kO). Data were collected at 1,000 Hz sampling rate with a bandpass filter ranging from 0.5 to 100 Hz. In addition, both the horizontal and vertical electrooculograms were measured using four electrodes. The peak-to- peak N1/P2 amplitudes for the five stimulus intensities were calculated, and the LDAEP was calculated as the slope of the linear regression curve.

In addition to EEG, sLORETA was performed to estimate the current density [15]. We calculated the three values of current density for the left, right, and averaged data from both hemispheres over the primary auditory cortex. The slope of the linear regression curve of the current density of BA41 for the five stimulus intensities was used to calculate the source LDAEP.

Statistical Analyses

All statistical analyses were performed using Statistical Package for Social Sciences (version 25; IBM Co.). The demographic, psychopathological, and LDAEP data of the patients were compared using correlation analysis (Spearman’s correlation) and Student’s ttest. Multivariate linear regression analyses were conducted to identify the association between age, gender, suicidal ideation, previous suicide attempts, depression severity, and each LDAEP. Binary logistic regression analysis was performed to determine whether the LDAEP is associated with a history of suicide attempts.

A total of 221 patients were included in this study. The distributions of demographic characteristics; total BDI, HAMD, BIS, HAMA scores and scalp and standardized low- resolution brain electromagnetic tomography (sLORETA)- derived LDAEPs of the patients are outlined in Table 1 according to the presence or absence of a history of suicide attempts. There were significant differences in age, BDI score, BSS score, and HAMD score between patients with and without a history of suicide attempts; however, there were no significant differences in variables, including LDAEP. The average age of patients without a history of suicide attempts was higher than that of patients with a history of suicide attempts (t = 2.295, p = 0.023). The proportion of patients with a history of alcohol consumption was significantly higher in the group with a history of suicide attempts than in the group without a history of suicide attempts (χ² = 17.761, p < 0.001). The total BDI score of the group with a history of suicide attempts was higher than that of the group without a history of suicide attempts (t = −3.17, p = 0.002). In addition, the total BSS score of the group with a history of suicide attempts was higher than that of the group without a history of suicide attempts (t = −6.62, p < 0.0001). However, the total HAMD score of the group without a history of suicide attempts was higher than that of the group with a history of suicide attempts (t = 3.31, p = 0.001).

The results showed that total BSS score was negatively correlated with N1/P2 LDAEP (r = 0.221, p = 0.028) (Fig. 1). Linear regression analysis conducted using history of alcohol consumption and sex as covariates showed that an elevated N1/P2 LDAEP was associated with a low total BSS score (t = −2.015, p = 0.045), whereas a history of alcohol consumption was associated with a high total BSS score (t = 2.963, p = 0.003) (Table 2). This indicated that increased serotonergic activity was associated with an increased likelihood of suicidal ideation.

Linear regression analysis conducted using BDI, BSS, and HAMA scores; age; and sex as covariates showed that an elevated P2 sLORETA-LDAEP was associated with a low BDI score (t = −2.533, p = 0.012), a high number of previous suicide attempts (t = 2.037, p = 0.043), and female sex (t = 2.556, p = 0.011) (Table 3). This indicated that having a history of suicidal ideation is associated with decreased serotonergic activity.

Multiple binary logistic regression analysis of the presence or absence of a history of suicide attempts showed that a history of suicide attempts was associated with an elevated P2 left sLORETA-LDAEP (coefficient = 4.638, p = 0.038), a low total HAMA score (coefficient = −0.077, p = 0.031), and high total BSS and BDI scores (coefficient = 0.005, p = 0.040) (Table 4). This indicated that patients who had previously attempted suicide had lower serotonergic activity than those without a history of suicide attempts. In addition, individuals with severe depression and suicidal ideation had an increased probability of a history of suicide attempts.

In this study, we investigated the relationship between LDAEPs and suicidality and the severity of depression in a large group of patients with major depressive disorder. The initial results, without controlling for variables, showed that LDAEPs were not significantly associated with the presence or absence of a history of suicide attempts. However, linear regression and binary logistic regression analyses showed that high LDAEPs were associated with a history of suicide attempts. In addition, the results showed that the severity of suicidal ideation and depression increased as LDAEPs decreased. Patients with a history of suicide attempts, on the other hand, had higher LDAEPs than those without a history of suicide attempts. In the group with past suicide attempts, the BDI score was higher, but the HAMD score was significantly lower. Because previous suicide attempt in this study is a self-reported history, it is assumed that the suicide attempt group has a stronger correlation with BDI than with HAMD, the objective scale. When only BDI item 9 and HAMD item 3, each representing suicidal ideation, were independently compared by suicide attempt group, BDI item 9 was mea-sured significantly higher in the suicide attempt group than in the suicide attempt group, but HAMD item 3 showed no significant difference between groups.

Suicide attempters show low serotonin activity [16,17], which is consistent with the results of the present study. In a previous study, the investigators compared the LDAEPs of patients with and without a history of suicide attempts and observed that the LDAEP of patients with a history of suicide attempts was higher than that of patients without a history of suicide attempts [18]. Similarly, the investigators in another study observed that among patients with depression, patients with a history of suicide attempts had higher LDAEPs than those without a history of suicide attempts. Still, the number of participants included in the study was relatively small [19].

As discussed previously, the use of LDAEPs for the prediction of suicide is controversial. A small study (n = 16 participants) demonstrated that the BSS scores of the participant groups with low and high LDAEPs, which were classified using median LDAEP values, did not differ significantly [20]. In another study of 73 outpatients with depression, the LDAEPs of patients with suicidal ideation did not differ significantly from those of patients without suicidal ideation [21]. In a recent study, the LDAEPs of patients in the suicide attempt and suicidal ideation groups were lower than those of participants in the healthy control group. However, the difference was not significant after clinical variables were controlled [22]. In another study, the LDAEP of the group with a history of suicide attempts (n = 11) did not differ significantly from that of the group without a history of suicide attempts (n = 130) [23]. It should be noted that the number of participants in the group with a history of suicide attempts was smaller than that in the group without a history of suicide attempts.

The consideration of only suicidal ideation for the prediction of suicidality has yielded controversial results. A study on LDAEPs and suicidality in patients with different subtypes of major depressive disorder revealed a signifi-cant negative correlation between LDAEPs and suicidal ideation among patients in the atypical depression group (n = 53), but there was no significant correlation between LDAEPs and suicidal ideation among patients in the non-atypical depression group (n = 68) [24]. In a different study, most patients with acute suicidal ideation had a high LDAEP [25].

It is noteworthy that this tendency of LDAEPs was reversed between the suicide attempt group and the suicidal ideation group in the present study. It is possible that in patients with suicidal ideation only, partial compensation occurs, and serotonin activity is temporarily increased. Conversely, in patients with a history of suicide attempts, compensation is no longer possible and serotonin activity decreases to a level lower than the baseline value [26].

Suicidal ideation was not distinguished from suicide attempt in previous studies on LDAEP and suicidality [27, 28]. Except for one study, other previous studies on LDAEP and suicidality were focused only on suicidal ideation or suicide attempt [22]. Among the studies on suicide attempts and LDAEP, some studies were focused on acute suicide attempts, whereas others included all patients with a history of suicide attempts, regardless of the time when the suicide attempts were made [29]. To avoid this, suicidal ideation and suicide attempts should be properly categorized in future studies, and the timing and methods of previous suicide attempts should be considered as well.

A recent LDAEP study by Uhl et al. [30] was conducted using a more specific methodology. The investigators [30] measured LDAEP 2, 5, 9, and 16 days after a suicide attempt. The results showed that the LDAEP was the highest on day 9. Instability of central serotonin function might have affected the decreased serotonin activity 1 week after the suicide attempt. This finding suggests that the definition of suicidality should be specifically established and evaluated in future LDAEP studies. The definition of suicidality should be based on whether actual suicide attempts were made, the number of suicide attempts made, the method of the suicide attempts, and the timing of the suicide attempts.

LDAEP values are affected by variables such as sex, age, and smoking [31]. This was pointed out as a limitation in previous studies on the correlation between LDAEP and suicidality [27]. This study is meaningful, as it is a large- scale study of the correlation between LDAEPs and suicidality in which demographic factors, such as age and sex, were controlled for.

However, it has some limitations. First, this was a retrospective study. Second, the presence of suicide attempt was defined as a history of any suicide attempt and the timing of LDAEP measurements was not controlled; therefore, there are limitations in predicting LDAEP changes or trends immediately after suicide attempt. Third, although most patients in this study underwent EEG at the time of initial treatment, some patients underwent EEG several weeks after receiving treatment.

In conclusion, the results of this study suggest that suicidal ideation is associated with elevated serotonin activity, while suicide attempt is associated with decreased serotonin activity.

The authors would like to thank Jo BW, Son JY, and Kim BE for her assistance with data collection.

This study was supported by a grant from the National Research Foundation of Korea (NRF), funded by the Ministry of Science and ICT (2020R1F1A1073188).

No potential conflict of interest relevant to this article was reported.

Conceptualization: Young-Min Park. Formal analysis: Youngjun Cho, Miseon Shim. Methodology: Young-Min Park, Seung-Hwan Lee, Youngjun Cho. Software: Miseon Shim, Seung-Hwan Lee. Writing—original draft: Youngjun Cho. Writing—review and editing: Young-Min Park.