Lower CSF 5-HIAA linked to more lethal suicidal behavior, review finds

A narrative review of 23 studies links lower cerebrospinal fluid levels of 5-hydroxyindoleacetic acid (5-HIAA), a major serotonin metabolite, with more lethal or violent suicide attempts, while finding weaker and less consistent evidence for homovanillic acid (HVA), a dopamine metabolite. The strongest associations appeared in mood disorders and among people making high-lethality attempts, but the evidence does not establish that either metabolite can predict who will attempt suicide or die by suicide.

The review, published in Frontiers in Psychiatry, synthesized studies published from 2005 through 2026 that were identified in Medline/PubMed and Google Scholar. It focused particularly on cerebrospinal fluid (CSF) concentrations of 5-HIAA and HVA in relation to suicidal ideation, suicide attempts, attempt lethality and subsequent suicide-related outcomes. Studies in other biological matrices, along with clinical, genetic, inflammatory and metabolic research, were included only as context. The authors did not pool effect estimates, and their qualitative synthesis was complicated by small samples, differing diagnoses, inconsistent CSF collection procedures and multiple observational designs.

Across the evidence base, reduced CSF 5-HIAA was the more consistent neurochemical finding. In depressed patients with comorbid alcohol dependence, people whose attempts were classified as high lethality had about 23.5% lower 5-HIAA concentrations than those making low-lethality attempts. The difference remained after adjustment for age and sex, even though the groups did not differ in depression severity, suicidal ideation or impulsivity. HVA and 3-methoxy-4-hydroxyphenylglycol, a norepinephrine metabolite, did not differ between the two groups. This pattern suggests that low 5-HIAA may be more closely related to the medical severity of a suicidal act than to the presence of suicidal thoughts or depressive symptoms alone.

Prospective evidence adds an important qualification. In people with bipolar depression followed for up to two years, baseline CSF concentrations of 5-HIAA, HVA and MHPG did not distinguish patients who later attempted suicide from those who did not. Lower baseline concentrations of all three metabolites were instead associated with greater lethality of attempts occurring during follow-up. The authors interpret this as evidence that monoaminergic activity may relate more closely to the severity of suicidal behavior than to whether an attempt occurs. The distinction matters because suicidal ideation, attempts, non-suicidal self-injury and suicide death are not interchangeable outcomes, and attempts themselves vary greatly in intent, reversibility and medical severity.

Findings also depend on psychiatric diagnosis. Evidence for lower 5-HIAA was comparatively coherent in mood disorders, including depression with alcohol dependence and bipolar depression, but inconclusive in schizophrenia spectrum disorders. In a cohort of people with schizophrenia followed for approximately 25 years, absolute CSF 5-HIAA and HVA concentrations did not differ significantly between those who died by suicide and survivors. A lower HVA-to-5-HIAA ratio was associated with suicide, however, indicating a relative monoaminergic imbalance rather than a deficit in either metabolite alone. A separate long-term study found no significant relationship between CSF 5-HIAA and early mortality or suicidal history after adjustment for confounders. These findings argue against treating low 5-HIAA as a universal marker operating independently of psychiatric context.

The review also emphasizes that suicidal behavior probably reflects interacting systems rather than a single neurochemical defect. A principal-component analysis of 20 CSF biomarkers identified one profile combining higher 5-HIAA, HVA and interleukin-6 with lower orexin that was associated with violent attempts and greater suicide mortality during follow-up. This does not fit a simple model in which every suicidal state produces uniformly low monoamine concentrations. Other studies linked persistent kynurenine-pathway dysregulation, including higher quinolinic acid and lower kynurenic acid, to suicidal symptoms and depression severity, while elevated cortisol in attempters pointed to activity in the hypothalamic-pituitary-adrenal stress system. Genetic research in the contextual evidence also suggested that suicide-related traits share a highly polygenic architecture involving synaptic function, cellular stress and DNA repair, among other pathways.

Early-life adversity and sex may modify these neurochemical associations. In one study of people who had attempted suicide, greater childhood exposure to violence was associated with lower CSF 5-HIAA in women but not men; among trauma-exposed participants, lower 5-HIAA was also linked with greater adult expression of violent behavior. In another sample, lower CSF HVA was associated with more severe anxiety among female attempters, but HVA was not associated with depression, overall psychopathology or suicidal ideation. A long-term schizophrenia cohort found more childhood bullying among those who died by suicide, while the separate finding of a lower HVA-to-5-HIAA ratio connected this exposure with a relative neurochemical imbalance. These observational associations cannot establish whether early stress causes the measured changes or whether both reflect other vulnerabilities.

Postmortem findings further complicate a simple “low serotonin” interpretation. In brain tissue from suicide cases, serotonin and 5-HIAA concentrations were higher in the brainstem, but the 5-HIAA-to-serotonin ratio was lower, indicating reduced serotonergic turnover rather than uniformly low metabolite concentrations. No significant difference was detected in the prefrontal cortex. Such regional patterns are compatible with altered release, receptor regulation or compensatory changes, but they do not establish causality. Nor do they prove that increasing serotonergic transmission would prevent suicide. Clinical responses to selective serotonin reuptake inhibitors vary by diagnosis, age and treatment context, so the observational biomarker findings cannot be used to infer preventive or therapeutic effects.

Even if the associations were stronger, CSF testing is poorly suited to routine suicide-risk screening. Lumbar puncture is invasive, most commonly causes post-lumbar-puncture headache and requires trained personnel, specialized facilities, sample storage and chromatographic analysis. CSF metabolites also reflect cumulative neurotransmitter turnover rather than an acute state, limiting their value for identifying imminent risk. Peripheral alternatives have also performed inconsistently. A single plasma 5-HIAA measurement did not reliably infer an individual’s absolute CSF concentration, despite correlations in within-person changes over time. Salivary 5-HIAA showed no overall difference between drug-naive patients with first-episode major depression and controls, and platelet serotonin findings remain limited to particular populations. Broader biomarker panels may eventually prove more informative, but no threshold or assay has been validated for individual clinical decisions.

The review therefore treats CSF 5-HIAA and, to a lesser extent, HVA as research-level correlates of suicidal behavior rather than actionable biomarkers. Lower 5-HIAA has the strongest support as an association with high-lethality attempts in selected clinical groups, but prospective studies have not consistently shown that it forecasts future attempts or suicide. The evidence remains limited by small samples, heterogeneous diagnoses, residual confounding, differences in medication and substance use, and variation in CSF collection and assay methods. Larger longitudinal studies would need standardized protocols, diagnostic stratification, independent validation and assessment of false-positive and false-negative classifications before these measures could contribute to objective risk models. Even then, any future use would have to complement rather than replace clinical assessment of psychiatric history, previous attempts, intent, social circumstances and access to lethal means.

Subject of Research: Review of cerebrospinal fluid serotonin and dopamine metabolites as correlates of suicidal behavior and attempt lethality

Article Title: Neurobiological basis of suicidal behavior: a narrative review of cerebrospinal fluid monoamine metabolites

Article References: Stefani, L., Marsella, L. T., & Tittarelli, R. (2026). Neurobiological basis of suicidal behavior: a narrative review of cerebrospinal fluid monoamine metabolites. Frontiers in Psychiatry, 17, Article 1968256. https://doi.org/10.3389/fpsyt.2026.1968256

Image Credits: AI Generated

DOI: 10.3389/fpsyt.2026.1968256

Keywords: suicide risk, suicidal behavior, cerebrospinal fluid, 5-HIAA, serotonin, HVA, dopamine, attempt lethality, biomarkers, neurobiology, Neurobiological, basis