Association Between The Loudness Dependence of The Auditory Evoked Potential and Age in Patients With Schizophrenia and Depression

The loudness dependence of the auditory evoked potential (LDAEP) reects serotonin neurotransmission. Abnormality in serotonergic activity is dominant in patients with schizophrenia (SCZ) and major depressive disorder (MDD). Patients with SCZ show weak LDAEPs, reecting high serotonergic activity. Some patients with MDD show high serotonergic activity. Although the changes in serotonin neurotransmission in the aging brain of SCZ and MDD have been observed, the relationship between central serotonergic activity and age remains unclear. The present study compared LDAEP between patients with SCZ and MDD, and healthy controls (HCs). We further examined whether age correlated with LDAEP and clinical symptoms, controlling usage of serotonin-related drugs. and MDD. We hypothesized that patients with SCZ would have smaller LDAEPs than patients with MDD and HCs, indicating increased serotonin neurotransmission in patients with SCZ. We further hypothesized that LDAEP would correlate with age in patients with SCZ and MDD. The current study intends to explain the paradoxical relationship between lower LDAEP in patients with MDD and SCZ compared to HCs and its increase with age in SCZ and MDD. In conclusion, the increased LDAEP as patients get older might be explained to protect the brain against the serotonergic hyperactivity with the disease progression. LDAEP could predict treatment responses in effect of age in patients with MDD and SCZ. This study included relatively small sample size and lacked the information such as illness of duration and drug dosage. Future longitudinal studies with larger sample size and full details in pharmacological history would expand our ndings. LDAEP Loudness dependence of the auditory evoked potential, SCZ:Schizophrenia, MDD:Major depressive disorder, HCs:Healthy controls, EEG:Electroencephalography, SSRIs:Selective serotonin reuptake inhibitors, PANSS:Positive and negative syndrome scale, HAMD:Hamilton depression rating scales, HAMA:Hamilton anxiety rating scales, BDI:Beck depression Inventory, MANCOVA:Multivariate analysis of covariance.

involved in the present study. The Hamilton Depression and Anxiety rating scales (HAMD and HAMA) [39,40] were evaluated in patients with MDD by a trained psychiatrist. In addition, the Beck Depression Inventory (BDI), which is a self-rating scale, was measured in patients with MDD and HCs [41]. The present study was approved by the Institutional Review Board of Seoul St. Mary's Hospital College of Medicine, The Catholic University of Korea (approval number: KC09FZZZ0211). Written informed consent was provided, and all experimental procedures followed the relevant institutional guidelines and regulations for all participants.

LDAEP paradigm and analysis
The auditory stimulation protocol comprised 500 stimuli with xed interstimulus intervals of 2000 ms. Tones of 1000 Hz with a duration of 100-ms (rise and fall time: 10 ms) were delivered at ve intensities (60, 70, 80, 90, and 100 dB SPL) through MDR-D777 headphones (Sony, Tokyo, Japan). A total of 500 stimuli comprised of each intensity of 100 stimuli were triggered via the STIM2 system (Compumedics USA, El Paso, TX, USA) to ensure accurate synchronization between the stimuli and EEG recordings. A xation cross was displayed in the middle of the monitor screen. A trained person with no information about the origin of the data removed gross artifacts through visual inspection. Artifacts related to eye blinks were removed using an established mathematical procedure [42]. Based on vertical EOG, positive and negative components exceeding 300 µV from before and after onset stimulus (-100 ms to 300 ms) were removed. Data were epoched in the range of -100 ms and 700 ms. Pre-stimulus baseline correction and linear detrend were applied to all electrodes. Artifacts exceeding ± 100 µV were rejected at all electrode sites. Off-line bandpass ltering between 1 Hz to 30 Hz was applied. The trials of each intensity were averaged at midline electrodes, respectively (Fz, FCz, Cz, Pz, and Oz). N100-P200 peak detection was performed using MATLAB 2019 software (Mathworks Inc., Natick, MA, USA) and Scan 4.5 software. For each intensity, the most negative peak amplitude of the N100 component was de ned between 80 ms and 160 ms after the stimulus onset, while the most positive peak amplitude of the P200 component was de ned between 130 ms and 280 ms. Finally, LDAEP was calculated based on the linear regression slope, whereby the slope variations of the sound intensities were ascertained by subtracting N100 from P200. The mean LDAEP was also calculated by averaging the LDAEPs at all ve midline electrodes (Fz, FCz, Cz, Pz, and Oz).

Statistical analyses
Demographic data including age, sex, and symptom scores were analyzed using the chi-squared test, multivariate analysis of variance, or t-test as appropriate.
To test the interaction effect between LDAEP and group, we analyzed the group differences in LDAEP using repeated measures analysis of covariance. The between-subject factor was group, and the within-subject factors were the LDAEPs at the ve electrode sites. Age and sex were controlled as covariates. In addition, for each single electrode, LDAEP was compared among the groups based on the multivariate analysis of covariance (MANCOVA), controlling for age and sex as covariates. LDAEPs were also compared between patients with SCZ and MDD with covariates, age, sex, and drug usage. In MANCOVA analyses, the signi cance level was set at p < 0.008 (two-tailed), considering the multiple comparison issue based on the Bonferroni correction [43]. Furthermore, partial correlation was performed among age, LDAEP, and symptom severity in each group, controlling for drug and sex. Binary classi cation of medication was designed for the presence or absence of drug usage that could modulate LDAEP (Table 4). Forty-seven patients with SCZ and twelve patients with MDD administered serotonin-related drugs in the current study. In terms of correlation analyses, p-values were adjusted via the Bonferroni correction with a signi cance level of p < 0.003. All statistical procedures were performed using IBM SPSS for Windows, version 21.0 (IBM Corp., Armonk, N.Y., USA).

Discussion
The current study focused on the disharmonic phenomena between the decrease of LDAEP and the age-related increased LDAEP in patients with SCZ and MDD. We observed the following signi cant results. First, patients with SCZ and MDD showed a lower mean LDAEP than HCs. Second, patients with SCZ showed a lower LDAEP compared to patients with MDD at the Oz electrode. Third, age was positively correlated with LDAEP in patients with SCZ and MDD.
Low levels of LDAEP entail high levels of serotonergic activity in patients with SCZ [20], corroborating the serotonin hypothesis of SCZ [37]. In this regard, we found smaller mean LDAEP at midline electrodes in patients with SCZ than in HCs. The present study suggests that patients with SCZ have serotonergic dysregulation, potentially caused by alterations in serotonergic projections to the dorsal hippocampus from the median raphe nucleus of the brainstem where corresponds with midline electrodes [44]. The selected EEG channels are suitable sites qualifying signal for referencing both mastoid. The longer distancing from the reference electrodes acquires the clear and larger amplitudes. However, LDAEP was not involved with symptom severity that would be caused from multi-dimensional relationships among LDAEP, serotonin activity, and clinical symptomatology. This argument is associated to which sub-type receptor has speci city on LDAEP and clinical symptoms. Recently, 5HT1A was positively correlated with LDAEP and 5HTT binding was negatively correlated with LDAEP [9]. Regarding the differences in each electrode among the groups, patients with SCZ and MDD exhibited smaller LDAEPs than HCs. Patients with MDD showed a higher LDAEP than patients with SCZ at the Oz electrode. In terms of MDD, they showed failures of responding to SSRI administration when they had low LDAEP and patient with high LDAEP were favorable responder [29,45]. SSRIs help to maintain presynaptic serotonergic neurons by inhibiting serotonin transporter binding to its receptor in MDD [46]. Further studies are warranted to elucidate the relationship between aging and several clinical domains, such as cognitive function, motor behaviors, and sub-clinical symptoms after controlling for medications. When observed ndings could be related with age, the impact of drug should be considered carefully on interpretating results. This fact, in general, has supposed to take for granted because the physiological complexity underlying mechanism of drug.
In this study, therefore, two critical impacts on LDAEP have claimed that rst, a natural decline of serotonergic activity as patients get older could be addressed. Second, both natural effect of aging and intentional use of drug might be appeared in complex to be decreased serotonin activation. With aging, patients with SCZ might show decreased psychotic symptoms, improved cognitive function, and reduced use of antipsychotics [47]. However, the neurophysiological mechanisms of the age-related improvements in SCZ have not been clearly understood. Down regulation of serotonin neurotransmission could induce a reduction of clinical symptoms and a relief of cognitive impairment in older SCZ patients [48,49]. This is partially in line with the results from the current cross-sectional study observing correlations between age and LDAEP. However, further longitudinal studies including information on duration of illness and pharmacological history would be needed to expand our ndings.
The current study intends to explain the paradoxical relationship between lower LDAEP in patients with MDD and SCZ compared to HCs and its increase with age in SCZ and MDD. In conclusion, the increased LDAEP as patients get older might be explained to protect the brain against the serotonergic hyperactivity Availability of data and materials The data presented in the current study are available from the rst author (K-I J) or corresponding author (J-H C) upon reasonable request.

competing interest
All authors declare no con icts of interest.

Funding Sources
This study was supported by a grant from the Korea Health Technology R&D project through the Korea Health Industry Development Institute (KHIDI) (HI17C2272), as well as by the KBRI basic research program through the Korea Brain Research Institute, which is funded by the Ministry of Science and ICT (21-BR-01-13).
Author contributions K-IJ and J-HC contributed to the conception and design of the study. K-IJ and CL contributed to the acquisition and analysis of data. K-IJ contributed to drafting the article. K-IJ, SK, CL, and J-HC contributed to the review of the article. J-HC and CL contributed to the supervision of the study. All authors approved the nal version of the article.

Figure 2
N100 and P200 amplitudes in selected electrodes between groups.

Figure 3
Partial correlations between age and mean LDAEP. *Signi cance level was set at an adjusted p-value (p < 0.003).