A Pilot Investigation of BDNF and GDNF in Chronic Insomnia: Absence of Group Differences and Correlation with Objective Sleep Parameters

1. Introduction

Insomnia, a sleep disorder marked by difficulty falling or staying asleep, or waking up too early, results in daytime impairment.¹ Chronic insomnia is defined as persistent sleep disturbances occurring at least three nights per week for a minimum of three months.² It is often associated with other medical or psychiatric conditions and can impair quality of life, increasing the risk of hypertension, diabetes, and mental health disorders.³

Diagnosis and assessment involve tools like the Insomnia Severity Index and the Pittsburgh Sleep Quality Index.² Its prevalence varies, with 30% of adults reporting symptoms, 9-15% experiencing daytime consequences, and 6% meeting diagnostic criteria.^4,5 Insomnia is more common in women and increases with age.^5,6 Risk factors for insomnia include comorbid medical and psychiatric disorders, shift work, and age-related changes.⁵ Long-term insomnia is linked to health consequences such as depression, anxiety, cognitive difficulties, and increased risk of cardiovascular disease and diabetes.⁷

Modifiable risk factors-such as difficulty initiating sleep, vulnerability to stress, maladaptive emotional coping, and pre-sleep arousal-are strongly associated with persistent insomnia and the development of anxiety and depressive symptoms.^8,9 In older adults, depression, anxiety, and chronic pain disorders further increase the risk.¹⁰ Additional predisposing and precipitating factors include female sex, loss of income, increased anxiety, depression, perceived stress, declining health, and increased pain.¹¹ Dysfunctional sleep-related beliefs also contribute to clinically significant anxiety and depressive symptoms.⁸ Addressing these modifiable factors is important for preventing insomnia onset and chronicity.

Genetic factors also play a significant role, with twin studies showing a mean heritability of 39%.¹² These genetic factors are linked to pathways such as cellular stress pathways, phototransduction, and muscle development.¹³ Insomnia is considered a complex polygenic, stress-related disorder resulting from interactions between genetic and environmental factors,¹⁴ and shares genetic architecture with traits like restless legs syndrome, aging, and various cardiometabolic, behavioral, psychiatric, and reproductive traits.¹⁵ Evidence indicates potential causal links between insomnia and coronary artery disease, depressive symptoms, and subjective well-being.^13,15 Epigenetic mechanisms may help explain how gene-environment interactions contribute to the long-term effects of insomnia.¹⁴

Recent research has explored the potential role of neurotrophic factors, particularly brain-derived neurotrophic factor (BDNF) and glial cell line-derived neurotrophic factor (GDNF), in the pathophysiology of insomnia. BDNF, a key protein supporting neuronal survival, synaptic plasticity, and cognitive function, has been implicated in sleep regulation.¹⁶ Some studies suggest that individuals with insomnia have lower BDNF levels than healthy controls, proposing a model where chronic stress and sleep disruption suppress BDNF expression, potentially creating a vicious cycle.^17,18 However, this finding is not universal, and the nature of the relationship remains unclear.

Similarly, GDNF, which is crucial for neuronal maintenance and protection, has been linked to sleep processes. Altered GDNF levels have been reported in chronic insomnia disorder, with some studies correlating these changes with poor sleep quality and cognitive impairment.^19,20 It is hypothesized that GDNF may be involved in the neuroinflammatory response and glial activation associated with chronic sleep loss.²¹

While both BDNF and GDNF are promising candidates, significant research gaps persist. Existing literature is limited by small sample sizes, methodological inconsistencies in biomarker assessment, and a lack of longitudinal data.^17,18

Despite these promising leads, the existing literature is characterized by inconsistent results, often attributed to small sample sizes, methodological variations in biomarker assessment, and a reliance on subjective sleep measures. The fundamental question of whether BDNF and GDNF levels are reliably altered in insomnia and directly associated with objective sleep deficits remains open.

To address this, the present study was designed to compare circulating BDNF and GDNF levels between well-characterized adults with chronic insomnia and normal sleepers, and to examine their associations with key objective sleep parameters derived from polysomnography. This approach aims to provide a clearer picture of the potential role of these neurotrophic factors in insomnia.

2. Materials and methods

3. Results

4. Discussion

This study examined the relationship between circulating neurotrophic factors and objective sleep parameters in adults with chronic insomnia and normal sleepers. Our findings present a clear contrast. We confirmed significant objective sleep disturbances in the insomnia group, who showed severely reduced total sleep time and sleep efficiency compared to controls. Contrary to our hypothesis, however, we found no differences in circulating BDNF or GDNF levels between groups, and initial correlation analyses showed no direct associations with sleep continuity. A more complex relationship was revealed through multivariate analysis: although the overall regression model was not significant (F(4,41) = 1.309, p = 0.283, R² = 0.113), sleep efficiency independently predicted BDNF levels (p = 0.034). This specific statistical association merits further investigation, although its biological significance remains uncertain.

We found no significant differences in BDNF (p = 0.610) or GDNF (p = 0.249) levels between the insomnia and control groups, nor were there direct bivariate correlations with sleep efficiency or total sleep time. However, a more complicared relationship emerged in the regression analysis: when controlling for diagnosis, age, and BMI, Sleep Efficiency was a significant individual predictor of BDNF levels (p = 0.034). These findings differ from prior studies reporting lower BDNF or GDNF in insomnia^19,20 or positive associations with sleep efficiency.^23,24 The discrepancies may reflect differences in sample size, assay methods, or participant characteristics. The low statistical power (9.7% for BDNF, 29.9% for GDNF) limits our ability to detect small effects, suggesting caution in interpreting these null results and the specific finding for BDNF and sleep efficiency.

Methodological differences in quantifying BDNF and GDNF likely contribute to the divergence from prior work. The absolute concentrations we measured differ from other reports, suggesting potential variations in assay specificity or the molecular forms detected (e.g., pro-BDNF vs. mature BDNF). This lack of standardization poses a significant challenge for cross-study comparison and highlights the need for consensus methodologies in the field. Furthermore, the strong positive correlation observed between BDNF and GDNF in our study is a notable finding. The strong BDNF-GDNF correlation warrants further investigation to understand their relationship in sleep disorders. The dissociation between bivariate and multivariate analyses implies that while BDNF and GDNF may not show simple linear relationships with sleep continuity, BDNF may have a more specific association with sleep efficiency when other factors are considered. This further suggests that their peripheral levels are not a direct reflection of the chronic inability to maintain sleep but may be involved in other aspects of sleep physiology.

Our multivariate regression analyses, corrected for multicollinearity, reveal a complex relationship. Despite the overall model’s lack of significance, the specific finding that Sleep Efficiency was a unique predictor of BDNF (p = 0.034) suggests that sleep continuity, specifically the ability to maintain sleep (efficiency), may be more relevant to BDNF regulation than total duration. This isolated finding should be interpreted as a preliminary signal that is conceptually aligned with models where sleep fragmentation disrupts neurotrophic support. Critically, other predictors, including diagnosis, did not make a unique statistically significant contribution. This collective evidence suggests that the utility of circulating BDNF and GDNF as standalone biomarkers for insomnia severity or sleep continuity is limited, though the specific link between BDNF and Sleep Efficiency merits further scrutiny. Our findings contribute to a growing body of evidence indicating that peripheral BDNF levels do not consistently serve as a reliable proxy for insomnia diagnosis or its core sleep continuity.¹⁷ While BDNF is unequivocally a key mediator of synaptic plasticity and neuronal health in the brain,²⁴ its expression in peripheral blood may be influenced by a multitude of factors not captured here, including genetic background, physical activity, diurnal rhythm, and unmeasured metabolic or inflammatory states.^16,25 Our evidence indicates the biological heterogeneity of insomnia and suggest that neurotrophic deficits may be a feature of a specific insomnia endophenotype, which was not differentiated in our present sample. The strong positive correlation observed between BDNF and GDNF, however, is an important finding that warrants further investigation to understand their coordinated regulation in the context of sleep disorders.

This proposed endophenotype, a distinct and measurable subtype within the broader insomnia population, might be characterized by specific physiological or genetic markers. For example, patients with high levels of cognitive and physiological hyperarousal, a core feature of insomnia, may represent a subgroup where neurotrophic factor dysregulation is most pronounced. Chronic stress and hyperarousal are known to suppress BDNF expression in limbic circuits^17,18; it is plausible that this central deficit is more likely to be reflected in the periphery within this hyperaroused subgroup. Also, genetic factors likely play a moderating role. Carriers of specific polymorphisms, such as the BDNF Val66Met variant (rs6265) which impacts activity-dependent BDNF secretion, may be more vulnerable to sleep-loss-induced alterations in neurotrophic signaling.^26–28 Our sample, while well-characterized for sleep, was not stratified by arousal metrics or genetic profile, which could have obscured a significant signal in a more homogenous subgroup. Another promising candidate for such an endophenotype is insomnia comorbid with objective cognitive deficits, particularly in memory consolidation, a process heavily dependent on sleep-related BDNF activity.

5. Conclusion

This study confirms objective sleep deficits in chronic insomnia but finds no simple bivariate correlations between sleep continuity metrics and neurotrophic factor levels. A multivariate analysis indicated that within a non-significant overall model, Sleep Efficiency was a significant predictor of BDNF, pointing to a potential specific relationship. However, the post-hoc power analysis revealed low statistical power to detect small effect sizes. Therefore, the null group differences and the isolated BDNF/Sleep Efficiency finding must be interpreted with caution and require confirmation in larger, sufficiently powered studies. The strong correlation between BDNF and GDNF further highlights a need to investigate their coordinated role in sleep disorders.

Ethics approval and consent to participate

All participants provided their signed, written informed consent. This study adheres to the Helsinki guidelines and was approved by the Ethics Committee of Kermanshah University of Medical Sciences (IR.KUMS.REC.1403.466).

Ethics statement

The study was approved by the Ethics Committee of Kermanshah University of Medical Sciences (IR.KUMS.REC.1403.466), which mandates protection of participant confidentiality. Controlled sharing of de-identified data for research purposes is permitted.

Access process

Researchers may request the de-identified dataset by emailing the corresponding author (Sepideh Nouraei, snouraei96@gmail.com) with a brief description of the intended use. Requests will be reviewed within 4 weeks. Approved requesters will receive the data under a CC-BY 4.0 license and may be asked to sign a simple data use agreement.