OBJECTIVES: Increased C-reactive protein (CRP) and reduced heart rate variability (HRV) both indicate poor prognosis. An inverse association between HRV and CRP has been reported, suggesting an interaction between inflammatory and autonomic systems. However, the prognostic impact of this interaction has not been studied. We thus investigated the prognostic impact of CRP, HRV and their combinations. DESIGN: Population-based study. SUBJECTS: A total of 638 middle-aged and elderly subjects with no apparent heart disease from community. METHODS: All were studied by clinical and laboratory examinations, and 24-h Holter monitoring. Four time domain measures of HRV were studied. All were prospectively followed for up to 5 years. RESULTS: Mean age was 64 years (55-75). During the follow-up, 46 total deaths and 11 cases of definite acute myocardial infarction were observed. Both CRP and three of four HRV measures were significantly associated with increased rate of death or myocardial infarction. In a Cox model with CRP >or=2.5 microg mL(-1), standard deviation for the mean value of the time between normal complexes <or=100 ms, and their combination, hazard ratio and 95% CI for subjects with both abnormalities was 3.20 (1.55-6.56), P = 0.0016, and for subjects with either abnormality 1.63(0.83-3.20), P = 0.15, after adjustment for conventional risk factors. The combination of CRP and other measures of HRV gave similar results. This indicates an interaction between CRP and HRV with a synergistic effect. CONCLUSIONS: The combination of CRP and HRV or heart rate (HR) predicts death and myocardial infarction with synergism, indicating interaction between inflammatory and autonomic systems with a prognostic significance.