Regional differences in infective endocarditis epidemiology and outcomes in Spain. A population-based study

Abstract

Abstract Introduction Population-based studies of infective endocarditis (IE) provide nationwide representative information of the disease, with a lower risk of selection and referral bias. Purpose Our aim was to describe the contemporary epidemiological profile of IE in Spain, and to evaluate regional variations in the incidence, characteristics and outcomes of IE in the different Spanish regions (autonomous communities, AC). Methods Retrospective population-based study to analyse the incidence, epidemiological and clinical characteristics and outcomes of all patients discharged with a diagnosis of IE from hospitals included in the Spanish National Health System, from January 2016 to December 2019. Differences in the IE profile among the 17 Spanish AC were also evaluated. Results A total of 9008 episodes were identified during the study period. Mean age was 69.5 (14.6) years, and 66.1% of the patients were male. 52.6% of episodes were treated at referral centers. Direct standardized incidence of IE in the study period was 5.77 (95% CI: 5.12-6.41) cases per 100000 person-years. 26.8% of episodes were prosthetic valve carriers, 36.8% had some kind of valve heart disease, and 10.6% had a cardiac implantable electronic device. The most frequently isolated microorganisms were staphylococci (33.3%; 19.0% Staphylococcus aureus and 14.3% coagulase-negative staphylococci), followed by gram-negative (21.0%) and streptococci (20.8%). Culture-negative IE accounted for 20.7% of episodes. 38.6% of patients developed heart failure during hospitalization, and 27.5% had acute renal failure. Cardiac surgery was performed in 19.3% of episodes in the total cohort, and in 33.4% of episodes treated in high-volume referral centres. In-hospital mortality was 25.1%. The standardized incidence of IE in Spain in the period 2016-2019 was 5.77 (95% CI: 5.12-6.41) cases per 100000 person-years. Standardized incidence rates (cases per 1000000 person-years) and risk-adjusted in-hospital mortality rates of infective endocarditis among AC are shown at Figure 1, were significant differences were found. In a multilevel logistic regression analysis, cardiogenic shock (OR 5.95, CI 4.98-7.9) and septic shock (OR 4.54, CI 3.78-5.44) were associated with the highest odds ratio for in-hospital mortality, whereas Streptococci viridans was associated with the lowest (OR 0.44, CI 0.37-0.52). Irrespectively of the AC, patients treated at high-volume centres (> 22 episodes per year) had a slightly lower risk-adjusted mortality (22.7±4.1 vs 23±3.6) p<0.001) (Figure 2, blue dots represent low-volume hospitals, red dots represent high-volume hospitals). Conclusions We found relevant and statistically significant differences in standardized incidence rates of IE, and risk-adjusted in-hospital mortality rates among regions. Regardless of the AC, patients treated in high-volume centres (>22 episodes per year) had a slightly lower risk-adjusted mortality.Figure 1Figure 2