The prognostic significance of stress-phenotyping for stroke incidence: the Malmö Diet and Cancer Study.

Abstract

Background: Self-reported mental stress is not consistently recognized as a risk factor for stroke. This prompted development of a novel algorithm for stress-phenotype indices to quantify chronic stress prevalence in relation to a modified stroke risk score in a South African cohort. The algorithm is based on biomarkers adrenocorticotrophic hormone, high-density lipoprotein cholesterol, high-sensitive cardiac-troponin-T, and diastolic blood pressure which exemplifies the stress-ischemic-phenotype index. Further modification of the stroke risk score to accommodate alcohol misuse established the stress-diabetes-phenotype index. Whether positive stress-phenotype individuals will demonstrate a higher incidence of stroke in an independent Swedish cohort was unknown and investigated.

Methods: Stress-phenotyping was done at baseline for 50 participants with incident stroke and 100 age-, and sex matched controls (aged 76 ± 5 years) from 2,924 individuals in southern Sweden. The mean time from inclusion to first stroke event was 5 ± 3 years. Stress-phenotyping comparisons and stroke incidence risk were determined.

Results: A positive stress-ischemic-phenotype reflected higher incident stroke (72% vs. 28%, p = 0.019) and mortality rates (41% vs. 23%, p = 0.019). Whereas a positive stress-diabetes-phenotype reflected a higher incident stroke rate (80% vs. 20%, p = 0.008) but similar mortality rate (38% vs. 25%, p = 0.146). Both the positive stress-ischemic (OR: 2.9, 95% CI: 1.3-6.5, p = 0.011) and stress-diabetes-phenotypes (OR: 3.7, 95% CI: 1.5-8.9, p = 0.004) showed large effect size associations with incident stroke independent of cardiovascular risk confounders.

Conclusion: Positive stress-phenotype indices demonstrated a higher incidence of stroke. Ultimately the Malan stress-phenotype algorithms developed in South Africa could confirm incident stroke in an independent Swedish cohort. Stress-phenotyping could thus be useful in clinical routine practice in order to detect individuals at higher stroke risk.