Brainstem auditory evoked potential (BAEP) is a widely used and objective electrophysiological method for assessing hearing in neonates. Its recording involves capturing a series of rapid, sinusoidal waves that occur within 10–15 ms of a click stimulus delivered via earphones. This set is generally composed of seven stable waves, namely I–VII, among which wave V has the highest amplitude and is the last to disappear as the stimulus intensity decreases. Consequently, wave V has the highest discrimination rate (1). Auditory evoked potential (AEP) wave VI is the first reproducible sine wave detected after wave V (2). The precise neural origin of wave VI remains uncertain. Some studies suggest it may arise from the internal capsule (3), while others propose the medial geniculate body of the thalamus as its likely generator (4). Balogh et al. observed AEP wave VI in 53 of 54 healthy subjects, demonstrating its prevalence in normal physiology, while its absence correlated with coma severity (5).

Chloral hydrate is a short-acting sedative–hypnotic frequently used in neonates undergoing BAEP testing. It acts via activation of the pH-dependent human proton-activated chloride channel, producing benzodiazepine-like effects (6). The Ramsay Sedation Scale is a common evaluation method for the sedative effect of chloral hydrate, classifying it as mild (Ramsay score of 1–2), moderate (3–4), or deep (5–6) (7). For non-invasive testing procedures, a Ramsay score of at least 4 is typically required to ensure minimal movement and cooperation (7, 8).

In our clinical practice, we observed the absence of wave VI in AEP recordings from neonates under partial chloral hydrate sedation. This observation led us to hypothesize that wave VI disappearance may be associated with sedation depth. Therefore, we conducted a prospective controlled trial to investigate whether AEP wave VI latency and disappearance correlate with varying levels of sedation and to evaluate their potential as objective neurophysiological indicators of sedation depth in neonates.

A total of 120 neonates were recruited and assessed in this study, with 20 excluded (11 failed to meet the inclusion criteria and 9 refused to participate); thus, 100 neonates were enrolled in the final analysis. The study flowchart is shown in Figure 1. There was no significant difference between the 80 cases in the treatment group and the 20 cases in the control group in terms of phototherapy, gender, gestational age, birth mode, birth weight, and proportion of small for gestational age (SGA) (P > 0.05). In the treatment group, there were 30 cases of wave VI disappearance, and the difference was statistically significant compared with the control group (P < 0.01) (Table 1).

The statistical analysis of the wave VI results from the neonates in the treatment group revealed that of the 35 neonates with a Ramsay score of 6, 14 cases had unilateral disappearance of AEP wave VI, and 10 cases had bilateral wave VI disappearance. The proportion of positive cases was 68.6%. Among the 23 neonates with a Ramsay score of 5, AEP wave VI disappeared unilaterally in four cases and bilaterally in two cases, accounting for 26.0% of the positive cases. The frequency of bilateral or unilateral wave VI disappearance increased significantly with an increase in the Ramsay score (P < 0.05) (Table 2). In the control group, no disappearance of wave VI was found (Table 3). Representative AEP tracings are presented in Figure 2 (wave VI under normal physiological conditions) and Figure 3 (absence of wave VI despite maximal stimulation intensity of 80 dBnHL under deep sedation). The AEP thresholds (wave V detection thresholds) did not differ significantly between the treatment and control groups: for the left ear, the treatment group averaged 22.1 ± 4.9 dBnHL vs. 23.0 ± 6.2 dBnHL in controls (p = 0.473); for the right ear, the values were 23.9 ± 5.6 dBnHL vs. 25.5 ± 5.6 dBnHL, respectively (p = 0.250) (Table 4). The Cochran–Armitage analysis revealed a significant relationship between Ramsay score and wave VI outcomes (P < 0.01) (Figure 4).

The ROC analysis demonstrated that the latency of wave VI is a significant predictor of its disappearance under deep sedation (Ramsay score > 5), with an AUC of 0.861 [95% confidence interval (CI): 0.746–0.975, P < 0.001]. At a Youden index of 0.589, the optimal cut-off latency was 8.465 ms, resulting in a sensitivity of 72.7%, a specificity of 86.2%, a negative predictive value of 89.2%, and a positive predictive value of 76.2% (Figure 5).

Chloral hydrate is the most commonly used sedative for routine neonatal examinations due to its efficacy and safety (9). Its mechanism of action parallels that of benzodiazepines (6). It provides rapid onset, deep sedation, prolonged duration, and rapid recovery, making it particularly suitable for neonatal AEP testing (10). With dosing protocols guided by experienced anesthesiologists over decades of neonatal clinical practice, its use has become well-established (11). However, individual variability persists, primarily due to factors including body weight, gestational age, and postnatal age. Despite extensive clinical experience, precise control of sedation depth remains challenging, as excessive sedation may induce life-threatening complications, including arrhythmia, respiratory depression, and airway obstruction (12). Therefore, an objective and effective indicator is urgently needed to accurately assess sedation depth.

The Ramsay scoring method is a simple and rapid tool for assessing sedation level and is supported by extensive clinical experience (13). Scoring requires an experienced clinician to evaluate behavioral responses at 10-minute intervals. This approach offers moderate accuracy but lacks objectivity. Consequently, research efforts have focused on identifying physiological indicators with greater sensitivity and specificity for sedation monitoring.

The clinical application of BAEP began in the early 1970s (14). Clinical research has demonstrated that AEP abnormalities in comatose patients with traumatic brain injuries manifest as the disappearance of wave VI (5). This phenomenon is thought to result from inhibition of the medial geniculate body (MGB) in the thalamus (15). The relay nucleus of the auditory radiation—composed of fibers from the primary auditory nucleus—transmits auditory signals through the posterior limb of the internal capsule to the auditory cortex, serving as a secondary processing center in the auditory pathway. Under deep sedation, inhibition of the thalamic MGB may disrupt signal transmission along the auditory pathway, reducing the information processing load on the cerebral cortex and thereby decreasing overall neural activity (16). Because the MGB within the internal capsule receives but does not relay feedback signals under inhibition, the corresponding AEP wave VI fails to appear (17). However, the precise neural mechanism underlying wave VI disappearance requires further investigation.

Owing to its non-invasive nature, objectivity, and high sensitivity, BAEP measurements remain stable across varied external conditions. Consequently, it has been widely applied to assess hearing and central nervous system dysfunction in neonatology, such as perinatal asphyxia, cerebellar atrophy, intrauterine growth restriction, encephalitis, bilirubin encephalopathy, and intracranial hemorrhage (18). AEP waveforms comprise seven peaks (waves I–VII), each originating from distinct neural generators. Wave I reflects auditory nerve action potentials, wave II arises from the cochlear nucleus, wave III from the pontine olivary complex, and waves IV and V correspond to activity in the lateral lemniscus and inferior colliculus, respectively. Waves VI and VII originate from the medial geniculate body and thalamic auditory radiations, respectively. The latencies of these seven waves index conduction times along successive segments of the auditory pathway (19). Prolonged latency of any wave indicates abnormalities in its corresponding conduction pathway, thus serving as an indicator of dysfunction (20). Therefore, clinical investigations of wave VI abnormalities should include latency analysis. Prolonged wave VI latency, wave VI disappearance, and reduced wave VI amplitude constitute positive indicators of subclinical auditory pathway suppression. As inhibitory neuronal activity increases, the wave VI signal weakens progressively, with latency increasing until complete disappearance. This progression is observed in brainstem evoked potentials during the gradual onset of chloral hydrate effects after neonatal oral administration.

Using the Ramsay Sedation Scale as the reference standard, we generated ROC curves of wave VI latency to predict its disappearance under deep sedation. The ROC analysis yielded an AUC of 0.861, with an optimal latency cutoff at 8.465 ms, corresponding to 72.7% sensitivity, 86.2% specificity, 89.2% negative predictive value, and 76.2% positive predictive value. These results indicate that prolonged wave VI latency may correlate with its eventual disappearance during chloral hydrate-induced sedation. Notably, recent research by Claesdotter-Knutsso et al. investigating the influence of methylphenidate on auditory evoked responses in patients with attention deficit hyperactivity disorder has also reported medication-induced alterations in wave VI (21). While their study involved a different patient population and medication, it offers additional evidence that pharmacological agents can alter BAEP waveforms. However, because our study focused solely on the effects of chloral hydrate, the applicability of these findings is limited to sedation induced by this agent. Incorporating a positive control group using alternative sedatives would strengthen generalizability. Future research should therefore investigate whether wave VI presence and latency reliably reflect varying depths of consciousness or sedation across different pharmacological agents and clinical settings.

Our study had certain limitations. First, subjects were stratified by Ramsay score to form three groups with progressively increasing sedation levels. Based on the abnormal wave VI incidence per group, we found a significant increase in wave VI loss frequency with higher Ramsay scores (P < 0.05). This suggests a positive correlation between sedation depth and wave VI alterations; confirming whether this relationship is linear requires validation in future studies with larger sample sizes. Second, none of the 20 control group cases exhibited wave VI disappearance during wakefulness or light sleep. This implies that drug-induced sedation may be a prerequisite for using wave VI disappearance as a sedation depth indicator, which requires further verification. Third, we found that wave VI disappearance incidence increased with sedation depth (Ramsay score of ≥5), reaching approximately 66.7% under deep sedation. The single-ear incidence of wave VI disappearance was slightly higher than that from two ears, but whether there is a relationship between the two is still unclear based on the current data. Future studies with larger samples should investigate whether there are statistical differences in left/right ear wave VI abnormalities.

In summary, AEP wave VI disappearance and latency serve as objective indicators of sedation depth during chloral hydrate-induced anesthesia in neonates, exhibiting high specificity and sensitivity. This holds significant importance for accurate sedation assessment and precision anesthesia practice. With further research, wave VI may emerge as a reliable indicator for sedation assessment, while its latency could serve as a potentially valuable metric for anesthesia evaluation.