Acute coronary syndrome (ACS) and stroke are major circulatory diseases and leading causes of disability (Virani et al., 2020) with psychiatric sequelae including depression and anxiety (Smith et al., 2015; Chun et al., 2018). In turn, comorbid psychiatric problems have negative effects on course and prognosis in circulatory diseases (Kim et al., 2018a; Lee et al., 2021). Suicidal behaviours are also common and known to be associated with poor outcomes of these diseases (Kim et al., 2018b; Vyas et al., 2021). Suicidal ideation is a prerequisite for more severe suicidal behaviour including non-fatal and fatal suicidal attempt, and could be a treatment target if it was adequately detected (Kim et al., 2016). Identifying potential determinants of suicidal ideation would be an important first step.

The roles of immune and inflammatory dysfunction have long been investigated as markers of suicidal behaviour in various psychiatric disorders (Black and Miller, 2015) and high-sensitivity C-reactive protein (hsCRP), as a well-documented clinical marker of general inflammation, has been received particular attention. Most studies have been conducted in patients with depressive disorder, but the results have been controversial. Some studies have found that peripheral hsCRP levels are significantly higher in depressed patients with suicidal behaviours than in those without (Ventorp et al., 2015; Ekinci and Ekinci, 2017; Oh et al., 2019), although others have found no such differences (Karlović et al., 2012; Chang et al., 2017; Peng et al., 2018). In a meta-analysis, blood hsCRP levels were significantly increased in psychiatric patients with suicidality versus healthy controls but did not distinguish psychiatric patients with or without suicidality (Black and Miller, 2015). Inflammatory biomarkers, and hsCRP in particular, have also been associated with incident myocardial infarction and stroke (Ridker et al., 1997); however, inflammatory hypotheses for suicidal behaviours have rarely been investigated in these patients with cardio-/cerebrovascular diseases.

Social support reflects the perception that individuals are cared about and held in positive regard by those in their support networks. Previous studies have repeatedly reported that social support is associated with a reduced risk of suicidal behaviours, both directly associated with reduced incidence (Chioqueta and Stiles, 2007; Kleiman and Liu, 2013) and buffering the impact of risk factors including life stresses (Yang and Clum, 1994; Meadows et al., 2005). In cardio-/cerebrovascular diseases, associations of social support have been investigated with depression (Kim et al., 2017) but rarely with suicidal behaviours (Kishi et al., 1996).

Based on the existing evidence, we hypothesised that social support might modify the associations of higher hsCRP levels with increased risk of suicidal behaviours in patients with cardio-/cerebrovascular diseases, a question that had not been previously investigated to our knowledge. Using data from two prospective Korean cohorts with ACS and stroke, we investigated associations of serum hsCRP and social support levels measured at acute phase of diseases with suicidal ideation evaluated both at acute and chronic phases, then evaluated potential modifying effects of social support on the associations between serum hsCRP levels and suicidal ideation.