Abstract:To know whether cirrhosis is a predisposing cause of IE and to determine the microbiology, prognosis and the role of cardiac surgery on mortality. Review of cases of infectious endocarditis in a university-affiliated hospital during 10 years. Thirty-one(9.8%) patients among 316 cases of infectious endocarditis had hepatic cirrhosis. Valve disorders were present in 62.2% of cirrhotic patients and infection occurred on aortic(48%) and mitral valves(45%).Endocarditis was hospital-acquired in 14(45%) and 11(17.7%) of cirrhotic and controls respectively(OR 3.82;95%CI:1.46-9.99;p 0.005).Staphylococcus aureus was the most common causative microorganism but βeta-hemolytic streptococci were most frequently isolated in cirrhotic patients (OR 8.75; 95% CI: 1.7-45.2;p 0.001).Renal failure was more frequent in patients with cirrhosis (OR 8.23; p 0.001).Cirrhotic patients had a higher mortality(51% vs 17.7%; OR 4.95;95%CI:1.89-12.91;p 0.001) associated with the severity of liver disease. Valve replacement was performed less frequently in cirrhotic patients(56.2% vs 92%) and the operative mortality was extremely high in patients with stages B and C. Hepatic cirrhosis is a frequent comorbid condition in patients with endocarditis. Due to the presence severe hepatic dysfunction, cardiac surgery is not undertaken even when indicated and mortality is high in stages B and C. Endocarditis is a serious hazard for hospitalized cirrhotic patients.Response to Reviewers: Dear Dr. Van Belkum:Thank you very much for the opportunity of revising the above mentioned manuscript. In doing the revision, I have taken into account all the suggestions of the reviewer.Firstly, let me tell you why we chose to perform a case-control study instead of a case-cohort comparison. A case-control study is a simple and very useful investigative tool to identify factors that may contribute to the understanding of many medical conditions. In this case, we compared patients with cirrhosis of the liver and endocarditis with subjects with endocarditis but without such underlying disease. In short, we looked back to see if there were characteristics of these patients that differ from those who did not have cirrhosis. The main reasons of our choice were: 1. Simplicity. 2. Accuracy. Because patients were seen over a relatively long period -10 years-we tried to bring nearer cases and controls to avoid biases associated with changes in diagnostic techniques (use of TEE, blood cultures), antibiotic usage or surgical team over the time. For this reason we selected two controls for each case: the previous and the consecutive to the index case.Reviewer:1. Our patients with cirrhosis of the liver were not on any type of chronic antimicrobial therapy with fluoroquinolones or other agents such as those that are administered for the prevention of primary bacterial peritonitis. We have included a short comment on this in the last paragraph of page 7. This is an interesting comment because such intervention may eventually favour enterococcal infections. Our cases, enter...