**Background:** Suicide is a major public health issue, with over 700,000 deaths annually and many more individuals exhibiting suicidal behaviors. Current assessments for suicidal risk are largely subjective and prone to false negatives or non-disclosure. Objective biomarkers are needed to improve screening and prediction. Electroretinography (ERG) is a non-invasive electrophysiological tool that records retinal electrical responses and has shown promise in detecting changes associated with psychiatric conditions, including suicidal behaviors. This systematic review aimed to synthesize available evidence on ERG correlates of suicidal behaviors, given that no prior review had been conducted on this topic.
**Methods:** The review followed PRISMA guidelines and included all article types without language restrictions. Databases searched included Ovid (Embase, APA PsycInfo, Ovid Emcare, Ovid Medline, Epub Ahead of Print), PubMed, Web of Science, and CINAHL. Search terms combined electroretinography and suicidal behaviors. After duplicate removal, 33 articles were screened by title/abstract, 19 were eligible for full-text review, and 14 were included. Two full texts could not be retrieved. Data extraction was performed independently by pairs of authors, covering study design, population, medication, ERG testing details, and findings. Quality was assessed using JBI critical appraisal tools.
**Key Results:** Among the 14 included studies, 12 were case reports, one was a case series, and one was a cross-sectional study. Publication dates ranged from 1980 to 2022. Participants' ages ranged from 19 to 59 years, with most being female. A total of 19 individuals had attempted suicide. Common substances used in suicide attempts included quinine (six studies), methanol, ergotamine, carbon monoxide, sodium valproate, lorazepam, flurazepam, and sildenafil. ERG testing was performed at various time points, from the first day post-attempt to several months or years later.
ERG findings were summarized across wave components:
- **a-wave:** Most studies reported normal amplitude and implicit time. However, a-wave latency was decreased in individuals with suicidal behaviors compared to controls (Fountoulakis et al., 2004). Some studies noted increased amplitude in the first two days that later normalized.
- **b-wave:** Amplitude was reduced in most studies (e.g., Zahn 1981, Verdon 2008, Treichel 2004, Brinton 1980, Meshi 2015, Kohli 2021). Implicit time was increased or faster than normal, and latency was increased (Fountoulakis et al., 2004). In one case, the b-wave was absent (Traill 2007).
- **b-a amplitude ratio:** Decreased over time (Zahn 1981, Kohli 2021).
- **Oscillatory potentials (OPs):** Decreased in several studies (Brinton 1980, Zahn 1981, Canning & Hague 1988, Kohli 2021) and completely extinguished in some (Brinton 1980, Zahn 1981).
- **Overall ERG patterns:** Reduced in some studies (Miura 2022, Canning & Hague 1988), but normal in one case series (Simmons & Good 1998).
The cross-sectional study by Fountoulakis et al. (2004) included 52 patients with major depressive disorder divided into three groups: no thoughts of death (n=17), nonspecific thoughts of death (n=23), and suicidal ideations (n=10). No significant differences were found between groups, but a-wave latency was lowest and b-wave latency was highest among those with suicidal ideations.
**Clinical Implications:** The consistent pattern of reduced b-wave amplitude, increased b-wave latency, and decreased b-a ratio and OPs suggests dysfunction in the inner retinal layers (bipolar, Müller, and amacrine cells), possibly reflecting alterations in neurotransmitters such as dopamine, serotonin, and glutamate that are also implicated in suicidal behaviors. However, the evidence is predominantly from case reports and small studies, limiting generalizability and causal inference. Confounding factors include the direct toxic effects of substances used in suicide attempts (e.g., quinine, methanol, carbon monoxide) on retinal function, as well as potential hypoxia-related neuronal damage. The review highlights the need for adequately powered, prospective studies with standardized ERG protocols (e.g., ISCEV standards), control for medication and comorbidities, and longitudinal follow-up proximal to the suicide attempt. Future research should also explore the link between specific retinal neurotransmitters and brain regions involved in suicidality to advance the potential of ERG as a non-invasive biomarker for suicidal risk.