Article Text

Download PDFPDF

Oral switch versus standard intravenous antibiotic therapy in left-sided endocarditis due to susceptible staphylococci, streptococci or enterococci (RODEO): a protocol for two open-label randomised controlled trials
  1. Adrien Lemaignen1,2,
  2. Louis Bernard1,
  3. Pierre Tattevin3,
  4. Jean-Pierre Bru4,
  5. Xavier Duval5,6,
  6. Bruno Hoen7,
  7. Solène Brunet-Houdard8,
  8. Jean-Luc Mainardi9,
  9. Agnes Caille8,10
  10. On behalf of the RODEO (Relais Oral Dans le traitement des Endocardites à staphylocoques ou streptOcoques) and AEPEI (Association pour l’Etude et la Prévention de l’Endocardite Infectieuse) study groups
  1. 1Service de Médecine Interne et Maladies Infectieuses, Centre Hospitalier Régional Universitaire de Tours, Tours, France
  2. 2Université de Tours, Faculté de Médecine, PRES Centre-Val de Loire Université, Tours, France
  3. 3Service de Maladies Infectieuses et de Réanimation Médicale, Centre Hospitalier Universitaire de Rennes, Rennes, Bretagne, France
  4. 4Service d’infectiologie et de médecine interne, Centre Hospitalier Annecy-Genevois, Epagny Metz-Tessy, Rhône-Alpes, France
  5. 5INSERM Clinical Investigation Center 1425, IAME 1138, Universite Paris Diderot, Sorbonne Paris-Cité, Paris, Île-de-France, France
  6. 6Service de Maladies Infectieuses et Tropicales, Assistance Publique-Hôpitaux de Paris, Hôpital Bichat, Paris, France
  7. 7Service de Maladies Infectieuses et Tropicales, CHRU de Nancy, Vandoeuvre-les-Nancy, France
  8. 8Unité d’Evaluation Médico-Economique, EA7505, Education Ethique, Santé, Centre Hospitalier Régional Universitaire de Tours, Université de Tours, Tours, Centre, France
  9. 9Service de Microbiologie, APHP, Hôpital Européen Georges Pompidou, Paris, France
  10. 10INSERM CIC1415, CHRU de Tours, Tours, France
  1. Correspondence to Dr Louis Bernard; louis.bernard{at}


Introduction Left-sided infective endocarditis (IE) is a serious infection with a heavy burden for patients and healthcare system. Oral switch after initial intravenous antibiotic therapy may reduce costs and improve patients’ discomfort without increasing unfavourable outcomes. We describe the methodology of two simultaneously conducted open-label randomised trials aiming to assess non-inferiority of oral switch as compared with entirely intravenous antibiotic therapy for the treatment of left-sided IE.

Methods and analysis Two simultaneous multicentre open-label prospective randomised trials assessing non-inferiority of oral switch during antibiotic treatment as compared with entirely intravenous therapy in patients with left-sided IE are ongoing. One trial is dedicated to left-sided IE caused by multisusceptible staphylococci (Relais Oral Dans le traitement des Endocardites à staphylocoques ou streptOcoques (RODEO)-1) and the other is dedicated to left-sided IE caused by susceptible streptococci or enterococci (RODEO-2). It is planned to randomise 324 patients in each trial after an initial course of at least 10 days of intravenous antibiotic therapy either to continue intravenous antibiotic therapy or to switch to oral antibiotic therapy. The primary outcome is treatment failure within 3 months after the end of antibiotic treatment, a composite outcome defined by all-cause death and/or symptomatic embolic events and/or unplanned valvular surgery and/or microbiological relapse (with the primary pathogen). Secondary outcomes include patient quality of life, echocardiographic outcome, costs and efficiency associated with IE care. Statistical analysis will be performed with a non-inferiority margin of 10% and a one-sided 2.5% type I error.

Ethics and dissemination Written informed consent will be obtained from all participants. This study was approved by Tours Research ethics committee (CPP TOURS-Region Centre-Ouest 1, 2015-R26, 23 February 2016). Study findings will be published in peer-reviewed journals and disseminated through presentation at relevant national and international conferences.

Trial registration number EudraCT Number: 2015-002371-16 and NCT02701608; NCT02701595.

  • infective endocarditis
  • oral administration
  • anti-bacterial agents
  • randomised controlled trial
  • adult

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

View Full Text

Statistics from

Strengths and limitations of this study

  • Relais Oral Dans le traitement des Endocardites à staphylocoques ou streptOcoques 1–2 trials are multicentre randomised controlled trials appropriately designed and powered to assess non-inferiority of oral switch of antibiotic therapy as compared with entirely intravenous antibiotic therapy in patients with left-sided infective endocarditis.

  • An economic evaluation, including a cost analysis, a cost–utility analysis and a budget impact analysis, will be conducted alongside the trials.

  • In these trials, few regimens are proposed for oral switch, in line with local epidemiology with a relative paucity of resistant bacteria, thus allowing a better homogeneity in the analysis.

  • Limitation due to the open-label design of those randomised trials will be limited by the use of a blinded committee to adjudicate the primary outcome. Primary outcome will be measured at 3 months in order to decrease the risk of lost to follow-up, the outcome at 6 months being measured as a secondary objective.


Infective endocarditis (IE) is a serious infectious disease with a heavy burden for patients and healthcare system.1 In France, median length of hospital stay for patients with IE is 43 days2 partly linked to the prolonged intravenous antibiotic therapy recommended by international guidelines (between 4 and 6 weeks in most situations).3 Current guidelines for IE management are mostly based on expert opinion, in vitro studies, animal experiments or clinical studies performed before the 1990s, as very few randomised studies have been conducted.4 5 The only exception to the golden rule of ‘intravenous treatment for all IE’ is right-sided IE due to meticillin-susceptible Staphylococcus aureus, in which the efficacy of an oral combination of ciprofloxacin and rifampicin has been validated in one randomised trial which only included 44 patients.6 Most experts acknowledge that the pharmacodynamic and pharmacokinetic characteristics of antibiotics such as amoxicillin, fluoroquinolones and rifampicin allow a high level of efficacy in the treatment of severe infections due to S. aureus, including IE, when orally administrated after an initial phase with adequate intravenous antibiotic therapy.7–12 A recent systematic review of oral therapy for the treatment of right-sided or left-sided IE found only one observational study reporting 80% cure rate with oral amoxicillin in 15 cases of streptococcal left-sided IE.13 Two recent studies regarding the management of IE in France showed that a switch from intravenous to oral antibiotics is feasible when patients with left-sided Staphylococcus or Streptococcus IE are stable after an initial course of intravenous antibiotic treatment, with or without valvular surgery.14 15 These practices have not been associated with unfavourable outcome, while significantly reducing the duration and cost of hospitalisation, the risk of nosocomial infection and patients’ discomfort. A first randomised trial recently found non-inferiority of partial oral treatment as compared with continued intravenous antibiotic treatment in IE due to Gram-positive cocci whatever its species.16 Other well-designed randomised controlled trials are however needed to confirm the clinical non-inferiority of this strategy in IE due to most common bacteria (multisusceptible staphylococci, susceptible streptococci or enterococci), specifying for each group of species. Addressing these bacteria in two simultaneously performed trials would ensure an optimal recruitment, reduce cost of research and argue for/against oral switch in the majority of patients with IE. The Relais Oral Dans le traitement des Endocardites à staphylocoques ou streptOcoques (RODEO) project corresponds to two pragmatic open-label randomised trials assessing non-inferiority of oral switch during antibiotic treatment as compared with entirely intravenous standard therapy in patients with left-sided IE. One trial is dedicated to left-sided IE caused by multisusceptible Staphylococcus and the other dedicated to left-sided IE caused by multisusceptible Streptococcus including Enterococcus.

Methods and analysis

Study hypothesis

We hypothesise that oral switch for antibiotic therapy is non-inferior to entirely intravenous antibiotic therapy in the treatment of left-sided IE as assessed by the proportion of patients with treatment failure within 3 months after the end of antibiotic treatment.

Study design

RODEO project comprises two simultaneously performed nationwide, multicentre, open-label non-inferiority randomised controlled trials comparing oral switch with entirely intravenous antibiotic therapy in patients with left-sided IE and an initial course of at least 10 days of effective intravenous antibiotic therapy. One trial is dedicated to left-sided IE caused by multisusceptible Staphylococcus (RODEO-1 trial) and the other dedicated to left-sided IE caused by multisusceptible Streptococcus including Enterococcus (RODEO-2 trial). Both trials are based on the same protocol provided below. Nevertheless, they are considered as two distinct trials, and sample sizes were calculated separately so that each trial has 80% power to show non-inferiority of oral switch as compared with standard intravenous antibiotic therapy.


Trials are ongoing at the time of publication in 28 university hospitals, 14 non-university hospitals, 3 private hospitals and 1 military hospital, all in France. The planned duration of the project is 67 months: 60 months for recruitment, and 7 months for maximal follow-up. The first patient was enrolled on 29 February 2016. End of recruitment is planned on 28 February 2021. At the time of submission, 97 patients have been included in the RODEO 1 trial (staphylococci) and 205 in the RODEO 2 trial (streptococci/enterococci). During the COVID-19 crisis, the maintenance of new inclusions was left to the discretion of the Research Department of the participating centres from 17 March to 11 May 2020. However, the follow-up visits for the patients already included were maintained as planned, in teleconsultation if necessary.


Eligibility criteria

Patients will be considered for inclusion in a trial if they have a left-sided IE and are in a stable condition after an initial course of at least 10 days of intravenous antibiotic therapy. Full eligibility criteria for both trials are listed in box 1. Most inclusion or non-inclusion criteria are common to both trials, apart from microbiological diagnosis. Microbiological analyses are not centralised but all participating microbiological wards are certified ISO15-189 and follow the current CASFM (Comité de l'antibiogramme-Société Française de Microbiologie)/EUCAST (European Committee on Antimicrobial Susceptibility Testing) guidelines.17 Drug-susceptibility testing follows the EUCAST disk diffusion method18 and minimal inhibitory concentration (MIC) is determined by broth microdilution or calibrated diffusion strips.

Box 1

Eligibility criteria for Relais Oral Dans le traitement des Endocardites à staphylocoques ou streptOcoques (RODEO) 1 and RODEO 2 trials

Inclusion criteria

For both trials

Diagnosis of definite left-sided infective endocarditis (IE) according to Duke criteria (3) on native or prosthetic valve

Age≥18 year old

Appropriate parenteral antibiotic treatment received for at least 10 days.

In case of valvular surgery, appropriate parenteral antibiotic treatment received for at least 10 days after surgery.

Planned duration of antibiotics of at least 14 days at the time of randomisation (ensuring to have at least 14 days of oral therapy remaining in the experimental group).

Absence of fever (temperature <38°C) at each time point during the last 48 hours (at least two measures/day) at the time of randomisation.

Negative blood cultures for at least 5 days at the time of randomisation.

Informed, written consent obtained from patient.

Patient covered by or having the rights to French social security.

For RODEO 1: left-sided Staphylococcus IE

Left-sided IE due to Staphylococcus sp (S. aureus or coagulase negative staphylococci) susceptible to levofloxacin and rifampicin

For RODEO 2: left-sided Streptococcus IE

Left-sided IE due to Streptococcus/Enterococcus sp susceptible to amoxicillin (minimal inhibitory concentrations ≤0.5 mg/L)

Non-inclusion criteria

For both trials

Body mass index <15 kg/m² or >40 kg/m²

Inability or unwillingness to take oral treatment for any reason (digestive intolerance, significant malabsorption) at the time of randomisation.

Absence of an entourage to support and watch for him/her at discharge

Expected difficulties regarding compliance with oral antibiotic treatment or follow-up (eg, severe cognitive impairment, severe psychiatric disease…)

Valvular surgery planned within the next 6 months

Presence of cardiac devices (pacemaker, implantable cardiac defibrillator) with suspected device-related IE without removal of the device

Breast feeding or pregnancy, or women on childbearing age without effective contraception

Expected duration of follow-up <7 months at the time of randomisation (eg, expected life expectancy <7 months, patient living abroad…)

Medical history of IE in the last 3 months

Other infection requiring parenteral antibiotic therapy after the randomisation

Inclusion in another interventional clinical trial

For RODEO 1: left-sided Staphylococcus IE

Glomerular filtration rate <50 mL/min/1.73 m2 for patients with Staphylococcus sp (S. aureus or coagulase negative staphylococci) infection

Contraindication to oral antibiotics administered in the experimental arm (ie, fluoroquinolones or rifampin)—including anticipated non-manageable drug interactions with rifampicin, and allergy or severe intolerance

Taking of an estrogen–progesterone treatment interacting with rifampicin

For RODEO 2: left-sided Streptococcus IE

Glomerular filtration rate <30 mL/min/1.73 m2 for patients with Streptococcus/Enterococcus sp infection

Contraindication to oral antibiotics administered in the experimental arm (ie, amoxicillin)—including allergy or severe intolerance

Study recruitment

To better coordinate inclusions, only one department is open in each recruiting centre. All but one are in the Infectious Diseases Unit. Potential participants are identified at the time they are hospitalised and receive intravenous antibiotic therapy for left-sided IE in one of the participating centres. Patients who meet selection criteria receive a brief study presentation and full participant information sheet by a clinician. After selection criteria confirmation and answering to patient questions about the trial, written informed consent is obtained.

Baseline data are collected following consent.


Randomisation takes place between day 10 and day 28 after initiation of the intravenous antibiotic therapy (and at least 10 days of intravenous conventional antibiotic treatment after valvular surgery, if performed), once the patient fulfils the inclusion criteria without having non-inclusion criteria and at least 15 days of remaining antibiotic therapy. In each trial, participants are randomly assigned in a 1:1 ratio to experimental group (switch to oral antibiotic treatment) or standard treatment (continuation of intravenous antibiotic treatment). Randomisation is carried out with stratification on whether or not the patient underwent valvular surgery for the control of the current IE episode. There is one random computer-generated sequence for each trial. Centralised randomisation is performed using a secure web-based randomisation system.


Patients and care providers are not blinded for pragmatic reasons (oral vs intravenous treatment).

Nevertheless, this potential bias is counterbalanced by the objectivity of primary outcome assessment (described below) and the presence of an independent blinded Endpoint Committee (EC). The EC is composed of one specialist in infectious diseases, one cardiologist and one microbiologist with expertise in IE management, research methodology and experience with clinical trials. The EC will review each suspected case in order to classify the primary outcome. Adjudication occurs after patients have completed their follow-up. Any disagreements among the EC members will be resolved during conference calls. All decisions made by the committee are final.

Study interventions

All patients initially receive an intravenous antibiotic therapy during 10–28 days before being randomised if they fulfil the eligibility criteria. The choice of which intravenous antibiotic agents are used and the expected total duration of antibiotic therapy, from 4 to 6 weeks, should be consistent with the 2015 European Society of Cardiology guidelines,3 and is under the responsibility of the physician in charge of the patient. Only patients who still require at least 14 days of treatment for their IE will be randomised.

Experimental group

Patients switch from initial intravenous antibiotic therapy to oral antibiotic therapy for the remaining duration of the treatment.

For left-sided IE due to multisusceptible Staphylococcus sp, patients ≤70 kg receive levofloxacin 500 mg one time a day in combination with rifampin 600 mg one time a day; patients >70 kg receive levofloxacin 750 mg one time a day in combination with rifampin 900 mg one time a day, as proposed for prosthetic joint infections.11

For left-sided IE due to multisusceptible Streptococcus sp or Enterococcus sp (ie, susceptible to amoxicillin with an MIC ≤0.5 mg/L), patients ≤70 kg receive amoxicillin 1500 mg three times a day and patients >70 kg receive amoxicillin 2000 mg three times a day.

If an adverse event (AE) leads to discontinuation of one antibiotic, the physician in charge of the patient will choose another oral antibiotic agent according to susceptibility testing. The patient will be classified as non-compliant with the strategy if a switch back to intravenous treatment is needed faced with the impossibility of finishing the remaining oral treatment period.

Control group

Patients continue intravenous antibiotic therapy for the remaining duration of treatment.

Study outcomes

Primary outcome

The primary efficacy outcome measure is the occurrence of treatment failure within 3 months after the end of the antibiotic treatment. Treatment failure is a composite outcome and is reached once a patient meets at least one of: (1) death from any cause; (2) symptomatic embolic events defined as secondary osteoarticular, splenic, brain or other symptomatic localisation after randomisation. Silent embolic events will not be included; (3) unplanned valvular surgery defined as cardiac surgery not planned before randomisation. Surgery due to sterile pericardial effusion or haemorrhage is, however, not included in this end point; (4) microbiological relapse (with the primary pathogen) defined as any blood culture positive yielding the same Staphylococcus sp isolate or the same Streptococcus sp or Enterococcus sp isolate, as the one responsible for the initial episode of endocarditis (ie, same species, same antibiotic susceptibility profile, the realisation of genotypic testing is not mandatory and left to the discretion of investigator).

Failures will be confirmed at the end of the follow-up by an independent endpoint adjudication committee, blinded from group allocation.

We also defined a primary safety outcome of all-cause mortality at day 30 after randomisation which will be analysed after recruitment of one-third and two-thirds of patients within each trial.

Secondary outcomes

The following variables will be compared between allocation groups as secondary outcomes:

  1. As advised for composite outcomes, each component of the primary outcome will also be considered independently.

  2. Treatment failure within 6 months after the end of the antibiotic treatment.

  3. New infection defined as the recurrence of positive blood cultures with a different pathogen to initial isolate sample within 3 and 6 months after the end of antibiotic therapy.

  4. Outcome assessed by echocardiography.

    Ultrasound examinations will measure: left ventricular ejection fraction, apparition, increase or decrease of the following items: vegetation, abscess, perforation, fistula, dehiscence of a prosthetic valve. A control echocardiography will be performed at the end of antibiotic treatment, at 3 months and 6 months after the end of antibiotic treatment.

  5. Catheter-related AEs and healthcare acquired infections as defined:

    • Catheter-related AE: infectious (eg, catheter-related bacteraemia) or non-infectious catheter-related complications (eg, extravasation, thrombophlebitis).

    • Other healthcare-acquired infections, including urinary tract infections, pneumonia, surgical site infection, Clostridium difficile infections.

  6. Quality of life

    We will assess patient’s quality of life at the end of antibiotic treatment, at 3 months and 6 months after the end of antibiotic treatment, using the EuroQol Five Dimensions (EQ5D3L).

  7. Antibiotic modification

    All change regarding antibiotic treatment administered will be recorded (drug, dose or duration). We will assess whether there is a need for a return to intravenous antibiotic in the experimental (oral switch) group.

  8. Compliance with oral antibiotic treatment

    The assessment of compliance with oral antibiotic treatment will be carried out at each visit during the treatment period by two combined methods: through a ‘patient leaflet’ which will permit to note take/omissions of treatment, filled by the clinician during hospitalisation, and by the patient or his caregivers after returning home; and through the return of the treatments’ boxes to the pharmacy of the investigational site, thus allowing a pill count.

  9. Economic outcomes

    The difference in costs (and length of hospital stays) will be computed from the healthcare system viewpoint between each new strategy of left-sided IE management (depending on the bacteria involved) and the real-life situation. The budget impact of the diffusion of each new strategy will be computed on a three-year timeframe. Incremental cost–utility ratios will be computed to assess the clinical and economic non-inferiority of the two new strategies.

Study procedures

All patients will be followed for a 6-month period following the end of antibiotic treatment. Follow-up is planned as follows: a visit at baseline or day 1 for randomisation (which is performed between day 10 and day 28 following the start of intravenous antibiotic therapy), one visit per week during the remaining antibiotic treatment duration, one visit at the end of antibiotic treatment and one visit at 14 days, 6 weeks, 3 and 6 months following the end of antibiotic treatment (figures 1 and 2).

Figure 1

Study design. IE, infective endocarditis; IV, intravenous.

Figure 2

Study schedule. *Clinical examination will collect the following information: body temperature, blood pressure, heart murmur (new or modified), any infectious site, list and tolerance of any drug, with a special focus on digestive symptoms, rash, neuropsychiatric problems. **Residual concentration of antibiotic treatment is realised only for patients randomised in ‘oral therapy’ group. ATB, antibiotic treatment; EQ5D3L, EuroQol Five Dimensions; IE, infective endocarditis; IV, intravenous.

Once a subject will be randomised in the study, every reasonable effort will be made to follow the subject for the complete study period even if there is a deviation from the intervention protocols, an early discontinuation of study treatment or if a participant misses one follow-up visit. If a subject is withdrawn from treatment due to an AE, the subject will be followed and treated by the investigator until the abnormal parameter or symptom has resolved or stabilised. All subjects who discontinue study treatment will be encouraged to complete all remaining scheduled visits and procedures.

Data management

Data are recorded on study-specific electronic case report forms (eCRFs) via an electronic data capture system (eCRF model is available on request to the principal investigator). To maintain participants’ anonymity, CRFs are identified only by a patient number and initials. All records that contain patient names or other identifying information will be stored separately from the study records in each centre and can be identified only by the patient number and initials.

Sample size

Sample size calculations are based on a null hypothesis of H0: π2–π1≥delta (ie, inferior); where π1 is the proportion of patients expected to experience failure in the intravenous group, π2 is the proportion in the oral switch group, and the non-inferiority margin delta is 10%. The alternative hypothesis is π2–π1<delta (ie, non-inferior). We considered each pathogen separately to ensure that we will have sufficient statistical power to explore non-inferiority of oral switch for staphylococci as well as for streptococci/enterococci. Thus, for each pathogen, Staphylococcus sp and Streptococcus/Enterococcus sp, we assumed an expected failure proportion of 10%,3 19 20 taking into account the fact that we will only enrol patients who have a favourable outcome after the first 10 days of IE treatment, a non-inferiority margin of 10%, a one-sided type I error of 2.5% and a power of 80%. The number of subjects required is estimated at 145 evaluable subjects per group, thus a total of 290 randomised patients. It is expected that approximately 10% of patients will not be available for the per-protocol (PP) outcome assessment, leading to a total of 324 patients to be enrolled, to be sufficiently powered for the PP analysis. The total required sample size is thus 648 patients: 324 patients for the Staphylococcus sp IE (RODEO 1 trial), and 324 further patients with Streptococcus/Enterococcus sp IE (RODEO 2 trial).

Statistical analyses

Statistical analyses will be conducted in both intention-to-treat (ITT) and PP methodology as recommended for non-inferiority trials. The PP population will exclude patients for whom there is a clear major protocol violation as defined during a blind review prior to any statistical analysis. Analyses will be conducted using two-sided significance tests at the 5% significance level. A participant flow diagram will be reported. Group characteristics at baseline will be studied with descriptive statistics. No statistical tests will be performed on baseline characteristics. For each trial, the risk of the primary outcome will be estimated within each intervention group. Difference of failure proportions between oral switch (p2) and entire parenteral treatment (p1) for the end of antibiotic treatment will be estimated. We will report point estimate for the between-group difference in failure risks (p2–p1) with its one-sided 97.5% CI calculated using the Wilson score method without continuity correction.21 We will declare oral switch to be non-inferior to parenteral treatment if the upper bound of the one-sided 97.5% CI is less than 10%. This analysis will be performed in both the ITT and PP populations. In the ITT analysis, missing primary outcome data will be handled by assuming that patients with missing data have treatment failure whatever the randomised group (worst case single imputation, assuming data are missing not at random). A sensitivity analysis will be performed excluding patients with missing primary outcome (complete-case analysis, assuming that data are missing completely at random). Another sensitivity analysis with adjustment on the stratification variable (initial valvular surgery for the control of the current IE episode) will be performed using a linear model (identity link function). Subgroup analyses will be performed considering the two strata defined by requirement of valvular surgery before randomisation or not. To assess the impact of a potential centre effect, a sensitivity analysis of the primary outcome will be performed with a random-centre-effect model. Potential post-hoc sensitivity analyses will be performed.

Statistical analysis will be first performed separately for each trial, that is, for staphylococci IE and streptococci–enterococci IE. Then, according to the results, we will consider a pooled analysis.

Concerning secondary objectives, the statistical analysis will be the same as for the primary outcome for the components of the primary outcome. Proportions of abnormalities will be compared using χ2 tests for echocardiographic outcomes.

Healthcare-acquired infection proportions and catheter-related non-infectious AE proportions will be estimated per group and compared using χ2 tests or Fisher exact tests.

Change in health-related quality of life will be analysed considering a linear mixed-effect regression model taking into account repeated measures for a given patient.

No imputation of missing data will be performed for the secondary outcomes. Descriptive statistics of compliance with oral therapy will be provided in the experimental group. Analysis will be performed in SAS V.9.4 (SAS institute) and R V.3.322 softwares (or latest versions).

Economic evaluation

From the data of three recruiting centres, cost analysis will evaluate, from the healthcare system viewpoint, which strategy between the oral switch (after an intravenous period of induction) or the intravenous antibiotic treatment (reference strategy) is less costly.

On this basis, the budget impact on the healthcare system of the diffusion of the oral switch strategy will be computed using a budget impact analysis on a 3 years’ timeframe.

Direct medical costs will be assessed from the healthcare system perspective in both groups and during the whole induction and follow-up period, that is, 6 months after the end of treatment. For each patient, we will collect the healthcare resources used both in the hospital setting and primary care services. This covers the initial hospital stay, subsequent hospital stays due to complications/infections, rehabilitation stay and antibiotics delivered in primary care. Data will be collected from the local hospital discharge databases of three centres (for hospitalisations) and from the CRF of all patients (rehabilitation care and antibiotics).

Using data from all recruiting centres, a cost–utility analysis will be performed to compute an incremental cost–utility ratio ‘cost per QALY gained’. QALY (Quality-adjusted life year) will be computed from the survival data and utility scores obtained from the responses to the questionnaire EQ5D-3L.

Data monitoring

Clinical research associates will ensure that patient inclusion, data collection, registry and rapport are in accordance with the standard operating procedures of the sponsor and the French Good Clinical Practices. They will verify during the quality control visits (at least once a year per centre), in collaboration with investigators: the presence of written consent, compliance with the research protocol, the quality of prespecified data collected in the case report form and its consistency with the ‘source’ documents and the management of treatments used.

Moreover, a Data Safety Monitoring Committee (DSMC) comprising two independent clinicians and one independent statistician meets approximately every 6 months to discuss any issues related to patient safety. All serious AEs will be reviewed by the DSMC as well as interim analysis of the primary safety outcome. Interim analyses of all-cause mortality at 30 days following randomisation will be performed after recruitment of one-third and two-thirds of patients within each trial. Early stopping rule will be to stop the trial for safety concerns if a p value <0.01 is observed. The role and responsibilities of the DSMC are set out in a written charter. The DSMC provides written recommendations to the trial steering committee following each meeting.

Ethics and dissemination

This protocol was approved by local ethics research committee (CPP TOURS—Region Centre—Ouest 1, 2015-R26, 23 February 2016). An agreement from the French national drug safety agency (ANSM) has also been obtained.

In conformity with the Declaration of Helsinki, all participants sign a written informed consent form that describes this study and provides sufficient information for patients to make an informed decision about their participation. Consent is obtained from patients before they undergo any study procedure. Participants may withdraw from the study at any time during the clinical trial without any impact on their care. In that event, data collected prior to participant withdrawal will be used in the trial analysis. Sponsor of the study may audit trial conduct as deemed appropriate. A formal amendment to the local research ethics committee will be required for any amendments to the study protocol which may impact the conduct of the study, or the potential safety of, or benefits to patients. If needed, an amendment will also be required from the National regulatory Agency for Security of Medicines and healthcare products (ANSM). Any protocol amendments will be communicated to investigators and oversight authority but also to trial participants and registries, if deemed necessary. The eighth amendment was the most recently approved, on 17 December 2018.

Reports will follow international guidelines: Consolidated Standards of Reporting Trials (CONSORT) Statement and Extension of the CONSORT Statement for reporting of non-inferiority and equivalence trials. Research findings will be submitted for publication in peer-reviewed journals regardless of whether or not they are statistically significant. Authors will be individuals who have made key contributions to study design and conduct. Trial findings will also be submitted for presentation at scientific meetings. The study findings will also be presented at relevant national and international conferences.

Patient and public involvement

Patients and public were not involved in the study design, recruitment or conduction of the study. The burden of intervention was assessed by representatives of patient associations participating in the ethical committee. Participants may obtain access to the final results of the study through the local principal investigator.


Several recent reviews point out the necessity of high-quality clinical studies in order to improve the level of evidence for the IE management.3–5 The RODEO trials aim to respond to this demand.

Iversen et al in the POET (Partial Oral versus Intravenous Antibiotic Treatment of Endocarditis) study have recently documented, in a first randomised open-label controlled trial, that a partial oral antibiotic treatment in left-sided IE was non-inferior to continued intravenous treatment and was not associated with unfavourable outcome.16 However, this study had some limitations which could be addressed in the RODEO study. First, strict inclusion criteria resulted in a large number of exclusions among screened patients (1554 out of 1954). We expect that the broader inclusion criteria of the RODEO project will lead to better external validity of the results. Second, unlike the POET study, the oral treatment regimen in our study will be more homogeneous, and closely controlled as the investigational products will be provided and controlled by the trial sponsor. Another limitation in the POET study was the potential bias of merging staphylococci, streptococci, and enterococci for analysis. Indeed, S. aureus is regularly isolated as a risk factor for poor outcome in IE,2 19 23 while IE due to streptococci with low MICs for amoxicillin could be treated with a short course of intravenous antibiotic treatment.24

The RODEO trials will be the biggest multicentre randomised controlled trials to assess non-inferiority of oral switch of antibiotic therapy as compared with entirely intravenous antibiotic therapy in adult patients with left-sided IE due to Gram-positive cocci (staphylococci for RODEO 1, streptococci and enterococci for RODEO 2).

If the non-inferiority is confirmed, this strategy could be a way to improve patients’ quality of life and reduce IE-associated healthcare costs. In order to evaluate this point, a medicoeconomic evaluation will be conducted alongside the trial.

The pragmatic design of these studies with wide eligibility criteria will permit to evaluate properly the medicoeconomic analysis, close to the real-life situation.

One of the limitations in the RODEO trials is that oral regimens are simplified in the experimental arm, contrarily to the recent POET trial which proposed many different oral combinations.16 Several reasons explain that choice. First, the homogeneity of treatment will be easier to interpret in each of the experimental arm. The combination therapy with rifampicin and quinolones has already been approved in other deep infections due to staphylococci.25 26 For streptococcal IE, oral amoxicillin has been recommended with reassuring results.8 13 Then, this is adapted to the French epidemiology of IE with a relative paucity of resistant bacteria.2 23 Therefore, precaution will have to be taken in the extrapolation of the results, notably for IE due to staphylococci resistant to quinolones and enterococci, as MICs are frequently over 0.5 mg/L.27 Then, we choose an evaluation of the primary outcome at 3 months after the end of the treatment as previous studies suggest that most of poor outcomes (mainly death related to IE) occur in the first 3 months after diagnosis,23 and a shorter duration for the evaluation of the primary outcome is supposed to decrease the risk of lost to follow-up. The evaluation of a composite score of poor outcome at the end of follow-up is scheduled as a secondary objective. Finally, risk of bias linked with the absence of blinding for the primary outcome measure is attenuated by the use of an independent blinded EC.

The expected non-inferiority of the experimental arm should help to modify the actual recommendations for IE management. Some retrospective studies had already pointed out the interest of oral switch of antibiotic treatment in IE,6 13 15 and a first randomised assay found the same results.16 The RODEO trials will possibly confirm these conclusions and try to demonstrate a potential medicoeconomic benefit to this strategy.

Their design will also permit to give robust conclusions for both streptococci and staphylococci IE with appropriate power.


View Abstract


  • Collaborators RODEO (Relais Oral Dans le traitement des Endocardites à staphylocoques oustreptOcoques multi-sensibles) Study group: Elisabeth Rouveix-Nordon; Lelia Escault; Raphael Lepeule; Caroline Charlier; Aurélien Dinh; Benjamin Davido; David Lebeaux; Laure Surgers; Virginie Vitrat; Cécile Janssen; Thomas Guimard; Xavier Lemaire; Olivier Patey; Hikombo Hitoto; Gabriela Illes; Emmanuel Forestier; Olivier Rogeaux; Simon Sunder; Gilles Dumondin; Valérie Gaborieau; Aurélia Eden; Lydie Khatchatourian; Mathieu Dupont; Eric Senneville; Laurent Hocqueloux; Jean-Philippe Lanoix; Vincent Dubée; Catherine Chirouze; Fabrice Camou; Renaud Verdon; Magali Vidal; Lionel Piroth; Olivier Epaulard; Karine Faure; François Goehringer; Audrey Therby; David Boutoille; Elisa Demonchy; Catherine Lechiche; Guillaume Beraud; Firouze Bani-Sadr; Matthieu Revest; Claire Chapuzet; Lydie Porte; Claudia Carvalho-Schneider; Frédéric Bastides; Marie-Charlotte Hallouin-Bernard; Bertrand Issartel; Franck Sibellas; François Delahaye; André Boibieux; Clémence Richaud; Christelle Diakov. The AEPEI (Association pour l’Etude et la Prévention de l’Endocardite Infectieuse) Study Group: François Alla; Francois Delahaye; Francois Goehringer; Christine Selton-Suty; Christophe Strady; Bernard Iung.

  • Contributors LB conceived and designed the final trial protocol. AC is responsible for the methodological design of the study and designed the protocol for statistical analysis. SB-H is responsible for the economic evaluation. AL, LB and AC wrote the first draft of the manuscript. LB, PT, J-PB, XD, BH and J-LM are members of the scientific committee. AL, LB, PT, J-PB, XD, BH, J-LM and members of the RODEO study group will be investigators and will recruit patients and conduct the trial. All authors read, reviewed and approved the final manuscript. No author has reported any competing interest.

  • Funding This research is funded by a national clinical research programme (Programme Hospitalier de Recherche Clinique: PHRC 2014) from the French Ministry of Health (EudraCT number: 2015-002371-16).

  • Disclaimer The funding source has no role in the design of this trial and will not have any role during its execution, analyses, interpretation of the data, or decision to submit results.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Patient consent for publication Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.