Short summary

Infectious endocarditis (IE) is a severe disease with 700 new cases per year in Denmark. Left-sided IE and cardiac device-related right-sided IE are well described compared to non-device related right-sided IE. This may be explained by it mostly occurring in intravenous drug users, for whom compliance can be suboptimal. We aim to describe the epidemiology, clinical characteristics, treatment and outcome of non-device right-sided IE in the Region of Southern Denmark. It is a retrospective descriptive registry study. Data will be retrieved January 2007 to May 2017. We hope the study results will contribute with important new epidemiological knowledge, and may serve better management for the patient group in the future. 

Rationale

Background

Infectious endocarditis (IE) is one of the lethal infectious diseases in the western world. Without treatment, the mortality is almost 100 %. When treated the short term mortality is 5-30%, depending on localization and microorganism. In Denmark, there are approximately 700 new cases yearly with IE. Of these, right-sided IE comprises 5-10%.

Left-sided IE is well-described in relation to the patient characteristics, treatment and outcome. In contrary, rightsided IE is not well-described, which can be explained by the low prevalence compared to left-sided IE.

Right-sided IE occurs in intravenous drug users (IVDUs), patients with pacemaker, Implantable Cardioverter Defibrillator, central venous catheter or congenital heart diseases. Based on these characteristics, rightsided IE can be dived into device-related IE and non-device related IE. Device-related IE is defined as an infection extending to the electrode lead, cardiac valve leaflets or endocardial surface. The frequency of patients with permanent pacemaker range from 0,38% to 1%. There is a standard treatment guideline for device-related IE as removal of device and intravenous antibiotic therapy.

The management of device-related IE is standardized compared to non-device IE, which can be explained by the increased number of patients who get implanted devices. Another explanation is the difference in the patient population. Non-device-related right-sided IE are mostly occurring in IVDUs and the patients compliance can be poor regarding to diagnosing, treatment and follow-up.

The treatment of non-device-related IE depends on severity, patient population and the underlying

microorganism. For right-sided IE, Staphylococcus aureus and Coagulase-negative Staphylococci are the most common microorganisms. The tricuspid valve is frequently affected. Studies suggest, that rightsided IE more frequently affects younger people compared to left-sided IE.

The recommended treatment for non-device-related IE is four weeks of antibiotics. In uncomplicated cases (absence of lung abscess), the treatment can be shortened to 14 days of intravenous followed by 2 weeks tablet antibiotic treatment. Right-sided IE is often benign and therefore rarely requires surgical therapy.

Compared to left-sided IE and device-related IE, the data is lacking regarding epidemiology, demography and management of non-device-related right-sided IE. That is despite there are more people with non-device rightsided IE. Furthermore, it is unknown the extend of surgical treatment, re-operation and outcome as heart failure.

Aim

The aim of this study is to describe prevalence, patient characteristics, isolated microorganisms in blood, treatment regime, if the treatment is carried out as planned and outcome in the form of e.g. the need of operation, reoperation, heart failure and death in patients with right-sided non-device-related IE admitted to Odense University Hospital. Indicators that correlate with best and worst outcome will be identified.

Perspective

We hope, this study results will contribute with an important new epidemiological knowledge in the field of non-device rightsided IE, and may serve better management for this patient group in the future.

Description of the cohort

Design: Retrospective descriptive registry study.

Study population: All patients admitted to Odense University Hospital with right-sided IE in the period January 2007 to May 2017. Initially, we will retrieve data from the Danish National Hospital Registry(NPR) for all patients with IE with the following ICD-10 codes:

DI33 Acute and subacute endocarditis

DI330 Acute or subacute infectious endocarditis

DI339 Acute endocarditis, unspecified

DI38 Endocarditis without specification of valve affection

DI389 Endocarditis, unspecified

DI39 Endocarditis and heart valve disorders in diseases classified elsewhere

DI398 Endocarditis, valve unspecified, in diseases classified elsewhere

It is not possible to differentiate between left-sided and right-sided IE based on ICD-10 codes, and therefore a systematic survey of hospital records will be made. Following, right-sided IE will be divided into device and nondevice IE. For patients with non-device-related right-sided IE a standardized review will be made with registration of age, profession, IVDU, known risk factors for right-sided IE, co-morbidities, isolated microorganisms in blood including antibiotics resistance, duration of planned treatment, is treatment completed (independent variable), and outcome as the need of operation, reoperation, heart failure and death (dependent variable). Through the study period it is estimated that 5.000 patients have been treated for IE in Denmark. Right-sided IE

represents 5-10% of these, corresponding to 250-500 patients in Denmark in the study period. We estimate, that approximately 20% of these ware treated in the Region of Southern Demark. We expect to include 50-100 patients in this study.

Treatment of IE is centralized at the Department of Cardiology and the Department of Infectious Diseases at Odense University Hospital in the Region of Southern Denmark.

Statistics

This is a descriptive register-based study based on ICD-10 codes and hospital records. Data will be described in contingency tables with continuous variables as medians with interquartile ranges (IQR) and categorical variables as proportions (percentages). Logistic regressions analysis will be used to identify univariate and multivariate correlations between the independent and dependent variable as described above. A p-value less than 0,05 will be considered as statistic significant. Data will be processed in REDCap. The statistical analysis will be performed using STATA program (version 14).

Data and biological material

The study is conducted in accordance with Helsinki II declaration. The Danish Data Protection Agency and Danish Patient Safety Authority granted permission. Data is ordered from NPR, the Danish Civil Registration System.

Data will be handled pseudo anonymous and results will be reported anonymous. The study is a register study and there will not have any patient contact.

Collaborating researchers and departments

Department of Infectious Diseases, Odense University Hospital

• Main supervisor professor Isik Somuncu Johansen  
• Primary investigator stud.med. Helena Lassen

Medical Departmant, Kolding Hospital

• Co-supervisor Stig Lønberg Nielsen, MD, PhD

Department of Cardiology, Odense University Hospital

• Collaborator senior consultant Sabine Ute Alice Gill