Exploring The Connection Between Infection And Atrial Fibrillation In Geriatrics

Objective:Our study aims to investigate the influence of infection on theoccurrence of atrial fibrillation in geriatric.Methods:Patients over65years old with sinus cardiac rhythm at outpatient oremergency were enrolled in our study, those who had suffered by hyperthyreosis,mitral stenosis, rheumatism, or anemia were not included. We observed the levels ofCRP(C-reaction protein), procalcitonin and NT-pro BNP(N terminal-B-typenatriuretic peptide)at first visit as well as those at peak during inpatient stay of nolonger than3months. Additionally, the number of people who encountered atrialfibrillation in the next3months and1year after the first visit correspondingly. Thestatistic difference was compared between patient diagnosed withinfection(pnuemonia) and those who were not. The correlation between the number ofpeople who encountered atrial fibrillation and CRP, procalcitonin or pro-BNP wascalculated. Besides, the predictive capability of atrial fibrillation of those threebiomarkers were also compared. The impact of infection on atrial fibrillation wasanalysed by COX proportional hazards regression model.Results:Total156male patients were enrolled in our study----120patients withinfection(infection group) while36patients without infection(non-infection group).compared with non-infection group. In infection group, the number of people whoencountered atrial fibrillation were obviously elevated, besides, that had positivecorrelation with the levels of CRP, procalcitonin and NT-pro BNP either at the firstvisit or at the peak during inpatient stay in3months which were prominently higherin infection group. As shown in the analysis of COX proportional hazards regressionmodel, infection is an independent risk factor in the occurrence of atrial fibrillation.PCT showed the highest positive predictive value of atrial fibrillation at both firstvistit or the peak. The highest negative predictive value of atrial fibrillation at firstvistit was occupied by PCT, while at peak by CRP. Conclusion:Infection(pneumonia) significantly increases the incidence of atrialfibrillation and the number of people who encounter atrial fibrillation, besides, is anindependent risk factor in in the occurrence of atrial fibrillation, and increases the riskof atrial fibrillation in geriatrics. CRP、NT-pro BNP、PCT are in positive correlationwith the number of people who encounter atrial fibrillation. PCT can be a good choiceto predict the chance of atrial fibrillation occurrence. The negative predictive value ofCRP level reduces the worrying about the occurrence of atrial fibrillation in clinics.However, the predictive values of these three biomarkers still need to be tested byfurther studies.