BACKGROUND: Cholecystokinin (CCK) and gastrin are related gastrointestinal hormones with documented cardiovascular effects of exogenous administration. It is unknown whether measurement of endogenous CCK or gastrin in plasma contains information regarding cardiovascular mortality.

METHODS: Mortality risk was evaluated using Cox proportional hazard regression and Kaplan-Meier analyses. Elderly patients in a primary care setting with symptoms of cardiac disease, i.e. shortness of breath, peripheral edema, and/or fatigue, were evaluated (n=470). Primary care patients were followed for 13years (from 1999); the 5-year all-cause and cardiovascular mortality was used as end point.

RESULTS: In univariate analysis, patients in the 4th CCK quartile had an increased risk of 5-year cardiovascular mortality (hazard ratio 3.9, 95% confidence interval: 2.1-7.0, p<0.0001). In multivariate analysis including established factors associated with cardiovascular mortality, CCK concentrations in the 4th quartile were still associated with increased 5-year cardiovascular mortality risk (HR 3.1, 95% C.I.: 1.7-5.7, p=0.0004), even when including 4th quartile NT-proBNP concentrations in the same model. We observed a marked difference between the genders, where CCK concentrations in the 4th quartile were associated with a higher 5-year cardiovascular mortality in female patients (HR 8.99, 95% C.I.: 3.49-102.82, p=0.0007) compared to men (1.47, 95% C.I.: 0.7-3.3, p=0.35). In contrast, no significant information was obtained from 4th quartile gastrin concentrations on 5-year cardiovascular mortality risk.

CONCLUSIONS: CCK in plasma is an independent marker of cardiovascular mortality in elderly female patients. The study thus introduces measurement of plasma CCK in gender-specific cardiovascular risk assessment.