In this cross-sectional study, we have developed and validated a nomogram model that incorporates sociodemographic characteristics, anthropometric data, clinical laboratory parameters, and psychological assessment scale indicators. The purpose of this model is to predict the risk of SA in the acute phase among Chinese patients with FEDN MDD. By utilizing this nomogram, clinicians can effectively identify individuals who are at higher risk for SA. The components of the nomogram include HAMD, HAMA, TSH, Log (TPOAb), and SBP. Notably, the nomogram demonstrates excellent predictive performance and consistent calibration curve accuracy in both the training and validation groups. Furthermore, decision curve analysis reveals the clinical relevance of this nomogram. The objective and easily accessible properties of our nomogram contribute to its significance as a practical tool.
There exists a substantial body of evidence supporting the association between increased severity of depression and the likelihood of suicidal behavior, including SA. It has been found that depression strongly influences the occurrence of SA. In a 2-year follow-up study, it was observed that depression raised the risk of SA sevenfold, whereas each prior suicide attempt increased the risk by 30% [28]. Preclinical and clinical evidence suggests that dysregulation of central serotonergic (5-HT) neurotransmission is linked to depression, suicidal behavior, and aggressive acts towards oneself [29]. Most currently used antidepressant medications exert their therapeutic effects by acting on 5-HTergic neurons, thus highlighting the importance of the brain's 5-HT system as a target for antidepressant treatment [30]. Pharmacological evidence strongly supports the role of 5-HT1A receptors in the mechanisms underlying depression and depressive-like behaviors, as exemplified by the antidepressant effects of 5-HT1A receptor agonists that are comparable to those of traditional antidepressants from the selective serotonin reuptake inhibitor (SSRI) family [31–34]. Studies have reported higher 5-HT1A binding potentials for 5-HT1A autoreceptors in the raphe nuclei of depressed individuals who have attempted suicide, as well as in suicide victims [35, 36]. Furthermore, reduced activity of downstream effectors of cortical 5-HT1A receptors has been identified in individuals who have died by suicide [37], along with diminished quantities or affinities of somatic dendritic and postsynaptic 5-HT1A receptors [38]. Recent investigations have revealed disturbances in cortical 5-HT1A receptor activity among patients with suicidal MDD [39]. Importantly, significant similarities in 5-HT receptor patterns and regulation have been observed when comparing depressive symptoms with suicidal behavior, suggesting the influence of shared genetic and epigenetic pathways on both depression and suicidal conduct.
Secondly, our investigation has revealed a positive association between the risk of SA and the severity of anxiety symptoms. This finding is consistent with previous research that has identified depression as a significant risk factor for suicide and has also demonstrated the role of anxiety symptoms in elevating the risk of suicide [40, 41]. Prior studies have indicated that, both before and after accounting for the level of depression, individuals with higher levels of anxiety exhibit a higher frequency of SA compared to those without anxiety symptoms [42]. Furthermore, a notable 79% of patients who died by suicide either during their hospital stay or immediately after discharge displayed severe symptoms of anxiety or agitation [43]. A systematic review exploring suicide as an outcome has identified comorbid anxiety disorders as a risk factor for suicide among individuals with depression [44]. Additionally, several studies have observed a higher rate of SA among patients with MDD who also experience anxiety symptoms in comparison to those without anxiety symptoms [45, 46]. Based on a prospective study spanning five years, it is postulated that anxiety contributes to the risk of SA through its ability to exacerbate illness severity and subsequently increase the risk of suicide [47]. Serotonergic dysfunction represents a plausible biological mechanism linking anxiety symptoms to SA [48]. Moreover, patients with elevated levels of anxiety face an increased likelihood of engaging in substance abuse, which further amplifies the risk of SA [49]. These findings underscore the importance for clinicians to closely monitor patients with comorbid anxiety and depression, regularly assess their anxiety symptoms, and implement appropriate measures to mitigate and prevent suicidal behaviors.
Thirdly, our research findings have established a correlation between SA and TSH as well as TPOAb. Elevated levels of TSH, TgAb, and TPOAb have previously been linked to an increased risk of depression and suicide [50], indicating their potential utility as biomarkers for evaluating SA in individuals with depression. Early investigations have demonstrated that patients with MDD who have attempted suicide exhibit higher levels of TSH, which are significantly associated with SA [51]. For instance, a study conducted by Berlin reported a positive correlation between SA and serum TSH levels equal to or above the upper 25th percentile of the normal range among patients with MDD [52]. There is also evidence suggesting that TPOAb may contribute to an increased risk of suicide and exacerbate symptoms of anxiety and depression [53]. The mechanism underlying the association between TPOAb and suicide risk can be understood by considering the following factors: TPOAb serves as a critical enzyme involved in the production of thyroid hormones and serves as an indicator of autoimmune thyroid disorders, such as Graves' disease. It is well established that individuals with autoimmune thyroid disease have a higher prevalence of suicide. For example, a recent large-scale cohort study conducted in the Danish population revealed that even after adjusting for underlying medical and mental conditions, individuals with Graves' disease still had a heightened risk of suicide mortality [54]. Furthermore, somatostatin and serotonin, known modulators of the hypothalamic-pituitary-thyroid axis [55], establish a connection between thyroid function, depression, and suicide. While various thyroid hormones and antibodies have been associated with SA, our research suggests that TSH and TPOAb warrant particular attention as indicators of concern.
In light of the investigation into hypertension, numerous studies have delved into unraveling the relationship between high blood pressure and the peril of suicide [56, 57]. A recent study revealed that a staggering 19.6% of hypertensive Chinese individuals harbored suicidal ideations [57]. Our research findings demonstrate a noteworthy positive correlation between SA in patients suffering from FEDN MDD and SBP, irrespective of other influencing factors. To elucidate this connection further, Dr. Scott et al. meticulously examined data from the World Mental Health Surveys, encompassing a substantial sample size of 37,915 individuals aged 18 and above. Their study successfully established a significant and autonomous linear association between blood pressure (BP) and SA [58]. Moreover, the intricate interplay between SA and hypertension was found to be subject to the influence of psychological distress [59]. This entwined cycle perpetuates a reciprocal relationship wherein psychological stress and hypertension synergistically converge, potentially culminating in suicidal behavior. Notably, hypertensive patients face an escalated risk of SA, even after accounting for other physiological and psychological ailments [60]. Previous investigations have postulated that dysregulation of the stress response system may be implicated in the development of hypertension and subsequent elevation in blood pressure, which has been implicated in the pathogenesis of suicide [61, 62].
Despite the satisfactory performance of our nomogram, it is important to acknowledge certain limitations. Firstly, our prediction model lacks verification in a heterogeneous population and relies solely on validation within the same training group. As a result, the obtained results may exhibit an excessive level of optimism. Secondly, this cross-sectional investigation exclusively included Han Chinese participants and was conducted solely within an outpatient clinic located in Shanxi Province, China. Therefore, additional external validation studies are crucial to ascertaining the applicability of our findings to diverse populations and other ethnic groups. Thirdly, the study exhibited an overrepresentation of female patients, which raises concerns about potential selection bias influencing the results to some extent. Fourthly, it should be noted that the current assessment outcomes may not fully meet practical expectations. The incorporation of novel biochemical markers or indicators, particularly those related to genetic factors, holds the potential to enhance the predictive capacity of the model in future research endeavors. Finally, it is imperative to recognize the existence of potential limitations when studying patients with FEDN in this investigation. Particularly due to the overlapping presentation of depressive symptoms in patients with unipolar depression and bipolar disorder, we cannot entirely exclude the inclusion of patients whose diagnosis may transition to bipolar disorder, although we made a second diagnosis during the 3- to 6-month follow-up period and only included patients who were diagnosed with MDD at both time points.