Sleep deprivation refers to a state in which an individual gets less sleep than required to feel alert and awake [1]. It has been reported that 82.6% of retired Iranian elders suffer from poor sleep quality [2]. Moreover, from 2003 to 2012, sleep insufficiency was consistently increasing in children subjected to the National Survey of Children’s Health [3]. In support of the physiological importance of sleep, many studies have indicated that sleep deprivation is associated with several conditions, including psychiatric disorders, cardiovascular diseases, inflammation, metabolic disorders, immune dysfunction, and even premature death [4–9]. Furthermore, it is the leading cause of car accidents in Iran, accounting for the death of 20,000–80,000 individuals in Iran annually [10, 11]. Another recently proposed hypothesis for the adverse effects of sleep deprivation is the disturbance of pro-oxidant/antioxidant balance (PAB) in favor of oxidants, which leads to oxidative stress [12]. One study suggested that the superoxide dismutase1 (SOD1) enzyme, which is the first defense line against superoxide free radicals and also the most extensively studied antioxidant enzyme, is disrupted in individuals with sleep deprivation, which leads to oxidative stress (OS) [13]. Epidemiological studies have also suggested an association between sleep deprivation and affective disorders such as depression and anxiety in children and adolescents [14, 15]. Another study has suggested that early life sleep deprivation was a predictive risk factor for later onset of anxiety and depression [16]. According to the World Health Organization report in 2015, more than 322 and 264 million people are struggling with depression and anxiety disorders worldwide, making up 4.4% and 3.6% of the global population [17]. A systematic review showed that prevalence of depression and anxiety increased to 279.6 and 301.4 million people [18]. Moreover, the incidence of these two common psychiatric disorders is increasing, especially in countries with low or middle income [19]. Depression is the single largest, and anxiety is the sixth most significant cause of disability and health loss globally [19].

To date, there has been no consensus on the physiological functions of sleep. A controversial hypothesis for the physiological role of sleep is acting as an antioxidant mechanism for the brain, which was proposed by Reimund [20]. The Reactive Oxygen Species (ROS) and oxidative stress markers could be accumulated in the brain tissue during Sleep deprivation [21]. Some stud suggested, enough sleep can increase antioxidant activity which promotes a brain protection against free radicals via a decrease in oxidant production [22]. Oxidative stress can also result in a wide range of diseases such as neurodegenerative diseases, cancer, male infertility, diabetes, autoimmune diseases, and cardiovascular disorders [4, 23–27]. In the past few years, numerous studies have sought to evaluate the involvement of OS in psychiatric disorders. Predominantly, therapeutic methods for depression and anxiety have focused on gamma-aminobutyric acidergic and serotoninergic systems [28]. But, researchers have established a link between OS and depression/anxiety during the last few years, suggesting that other systems like oxidative metabolism can affect mental health, which might draw more attention to antioxidants [28–30]. Nonetheless, there remains a controversy surrounding the association of depression/anxiety with PAB. It is not yet well understood whether depression/anxiety are associated with increased or decreased PAB [31].

Understanding the physiological functions of sleep and its role in antioxidant mechanisms is necessary for identifying the impacts of sleep deprivation on mental health. Since most studies regarding the effect of sleep deprivation on PAB and anxiety/depression and the association between serum PAB and anxiety/depression are mostly conducted on animals or non-population-based with limited sample sizes, we cannot have a decisive conclusion. Consequently, the primary objective of this population-based cross-sectional study was to investigate the impact of sleep duration on PAB, SOD1, anxiety, and depression. The secondary objective was to evaluate the association of these oxidative stress determinants with anxiety and depression in subjects with very short nightly sleep (Figure 1).

In the current analysis, 9,184 subjects were eligible and enrolled, of whom 3,682 (40.09%) were male. As indicated in Table 1, a chi-square analysis found men were significantly more likely to experience short sleep duration than women. In addition, one-way ANOVA indicated a significant difference in age, depression and anxiety scores, and serum PAB level between three defined subgroups of nightly sleep duration. After performing post hoc analysis, we found that anxiety and depression scores are meaningfully higher in very short sleep subjects in comparison with the other groups. Furthermore, for PAB, it was considerably higher in the very short sleep subgroup than just normal subjects. No significant differences were found in all mentioned features between short and normal participants.

Based on results showed in Figure 2, in participants with very short nightly sleep, severe cases were found significantly more frequent depression and anxiety than the other groups. Moreover, as shown in Figure 3, the nightly sleep duration and serum PAB level were respectively, significantly lower and higher in severe depressive and anxiety cases than the other ones.

In crude analysis, it was found that in both depressive and anxiety subjects, serum PAB was significantly higher in very short sleep subjects in comparison with the others, respectively (Figure 4). After adjusting results for age and sex in a MANOVA regression analysis, we found that subjects with anxiety and depression and very short sleep had 12.09- and 10.69-unit PAB more than others, respectively (Table 2).

We have demonstrated that very short nightly sleep was significantly associated with higher depression and anxiety scores. There was a significant association between sleep duration and serum PAB level, but serum SOD1 activity was not associated with sleep duration. In individuals with Generalized Anxiety Disorder (GAD), sleep duration was conversely associated with PAB. In other words, the less sleep duration, the higher PAB.

In 1994, Reimund suggested that a key function of sleep might be acting as an antioxidant for the brain [20]. Oxidative phosphorylation -taking place in mitochondria-is the main ATP supply of the cell in all aerobic organisms that produce free radicals as a by-product [31]. At low concentrations, free radicals are necessary for normal cell functioning. They serve numerous physiological processes, including mitosis, gametogenesis, cell signaling, and cellular response to infection and injury [25, 42]. When the disturbance in pro-oxidant/antioxidant balance (PAB) occurs in favor of the oxidants, cells may be exposed to oxidative stress. OS results from either increased generation of reactive oxygen species (ROS) or impaired oxidative defense mechanisms. SOD is the most extensively studied antioxidant enzyme and serves as the first defense line against superoxide free radicals, which catalyzes the dismutation of the highly reactive superoxide radical, O²⁻, to O₂ and H₂O₂ [13, 31]. The brain is especially more susceptible to OS because it is mostly made up of lipid and consumes comparatively large amounts of oxygen and produces an excessive amount of free radicals with a relatively modest oxidative defense system [43]. In support of the Reimund hypothesis, Hill and associates have indicated that the upregulation of gene expression of antioxidant enzymes in wild-type flies shortens sleep duration, suggesting that decreasing PAB promotes wakefulness [44]. One previous study has reported that sleep deprivation may induce oxidative stress, suggesting that sleep clears the ROS accumulated in the brain during wakefulness [45].

We found that inadequate sleep is associated with a high serum PAB in subjects with anxiety and depression. To our knowledge, we are the first study to show this association. In contrast to our findings, in studies conducted on humans, OS was not observed in sleep-deprived individuals [46–48]. Furthermore, in a narrative review, Atrooz et al. have suggested that while chronic sleep deprivation diminishes the antioxidant response, acute sleep deprivation may augment antioxidant mechanisms [13]. There are inconsistent results regarding the impact of sleep deprivation on SOD1 activity. In line with our findings, Gopalakrishnan and coworkers found no significant change in SOD1 activity in rats with chronic sleep deprivation [46].

As previously discussed, the brain is especially vulnerable to oxidative stress (OS); and OS can affect synaptic integrity, neurotransmission, and the overall function of the brain, resulting in neurological and psychological disorders [49, 50]. Data on the relationship between OS and depression/anxiety are controversial. In line with our findings, some authors have revealed a strong positive relationship between OS and depression/anxiety [28, 51, 52]. Contrastingly, some authors have failed to demonstrate an association between OS and the presence of depression/anxiety [53–55]. When it comes to the activity of antioxidant enzymes in depression/anxiety, the data are discrepant. While we found no differences in serum SOD1 activity in different depression/anxiety severities, similarly, as observed in another study, both increased and decreased activity of SOD1 have been reported previously [31, 56, 57]. An increased expression of SOD1 may induce inflammation, which inflammation itself can contribute to depression/anxiety [29, 58, 59]. Far as we know, researches evaluating the association between different depression/anxiety severities and OS using PAB are scarce. In contrast with our results, in another study, no association was found between serum PAB level and severity of depression [55].

Our study has assessed the trinary relationship between subjects’ sleep duration, oxidative stress, and anxiety/depression scores of a large randomly-selected population in a city in the Middle East. In addition to our considerable efforts, some limitations existed in our procedures. First, our analysis was a cross-sectional one that could not accurately conclude cause and effect relationships between these three mentioned features. Moreover, data on the nocturnal sleep duration of our participants were collected using self-report questionnaires and we could not assess other oxidative stress parameters. Further cohort and case-controlled studies with more accurate methods for observation of sleep are required to better identify the risks of sleep deprivation.

In conclusion, our cross-sectional analysis revealed a logical trinary association between sleep duration, oxidative stress, and anxiety/depression and mainly indicated that sleep deprivation is associated with increased pro-oxidant/antioxidant balance and severe depression/anxiety. For future works we can use another categorization for response variables and also we can use machine learning algorithms for analyzing the data.