# Prognostic value of auditory evoked potentials in disorders of consciousness: a systematic literature review

## TL;DR

This paper reviews how auditory evoked potentials can help predict outcomes for patients with disorders of consciousness, especially when deciding on life support withdrawal.

## Contribution

The study provides a systematic review of recent evidence on the prognostic value of auditory evoked potentials in disorders of consciousness.

## Key findings

- Mismatch negativity reliably predicts both favorable and unfavorable outcomes in disorders of consciousness.
- Absence of brainstem or middle-latency auditory potentials is strongly associated with poor outcomes.
- P300 shows potential for predicting recovery, though results are inconsistent.

## Abstract

•Auditory evoked potentials can be used for prognosis in disorders of consciousness.•Mismatch negativity predicts outcomes in consciousness disorders, literature shows.•Absence of brainstem or middle-latency auditory potentials predicts poor outcomes.

Auditory evoked potentials can be used for prognosis in disorders of consciousness.

Mismatch negativity predicts outcomes in consciousness disorders, literature shows.

Absence of brainstem or middle-latency auditory potentials predicts poor outcomes.

Decisions regarding withdrawal of life support in disorders of consciousness (DoC) are challenged by prognostic uncertainty. Auditory evoked potentials (AEPs) are widely used, yet no recent systematic review has focused exclusively on their prognostic value. Given evidence that some withdrawal decisions are based on inaccurate prognoses, this study aimed to evaluate the prognostic utility of AEPs in DoC using recent evidence.

Comprehensive literature searches of PubMed, Scopus, and Web of Science Core Collection identified studies published between 2013 and 2023 assessing AEPs for prognostication in DoC. Three authors independently screened studies using predefined criteria. Eligible studies were original English-language research evaluating prognostic outcomes in adults (≥18 years) with DoC. Reviews and secondary literature were excluded. Risk of bias was assessed using the QUADAS-2 tool.

Of 2249 records screened, 24 studies met inclusion criteria, encompassing 1562 patients. Mismatch negativity (MMN), the most frequently studied AEP, reliably predicted both favourable and unfavourable outcomes. P300 demonstrated potential for predicting recovery, though findings were inconsistent. Although less studied, the absence of N100 predicted poor functional outcomes, including reduced autonomy. Absence of brainstem or middle-latency AEPs predicted negative outcomes with high specificity but low sensitivity.

MMN, P300, and N100 demonstrate meaningful prognostic value in DoC. Despite methodological heterogeneity, current evidence supports AEPs as useful clinical prognostic tools. Standardised, blinded protocols are needed to improve reliability.

Integrating AEPs into multimodal prognostic frameworks may enhance outcome prediction. Advances in electrode coverage, machine learning, and automated detection paradigms offer promising future directions.