Impact of the duration of antibiotics on clinical events in patients with Pseudomonas aeruginosa ventilator-associated pneumonia: study protocol for a randomized controlled study

Patients aged ≥18 years, diagnosed with ventilator-associated pneumonia documented with Pseudomonas aeruginosa (PA-VAP), affiliated to French social security, and providing a written informed consent signed by themselves or their next of kin are eligible for inclusion. The diagnosis of PA-VAP will include a clinical suspicion (≥ two criteria including fever > 38.5 °C, leukocytosis > 10⁹/L or leukopenia < 4.10⁸/L, purulent tracheobronchial secretions, and a new or persistent infiltrate on chest radiography) and confirmation by a Pseudomonas aeruginosa positive quantitative culture of a respiratory sample: bronchoalveolar lavage fluid (significant threshold ≥10⁴ colony-forming units (CFU)/mL) or plugged telescopic catheter (significant threshold ≥10³ CFU/mL) or quantitative endotracheal aspirate distal pulmonary secretion samples (significant threshold ≥10⁶ CFU/mL) [20].

Patients are not eligible in case of the following conditions: pregnancy, immunosuppression (HIV, immunosuppressive therapy, corticosteroids >0.5 mg/kg per day for more than a month), current antibiotic therapy active on PA for extrapulmonary infection, procedure of withdrawing life-sustaining treatment, chronic pulmonary colonization with Pseudomonas aeruginosa (chronic obstructive pulmonary disease (COPD) or bronchiectasis, with a positive respiratory sample below the threshold rate for Pseudomonas aeruginosa (i.e., <10³ CFU/mL for protected specimen brush or <10⁶ CFU/mL for tracheal aspirate), obtained in the absence of pneumonia or exacerbation during the 6 months before the ICU admission).

Antibiotic treatment should be started just after realization of bacteriological sampling, without waiting for the result. The choice of initial antibiotic therapy will be left to the discretion of the physician according to usual care based on the clinical context, previous antibiotic therapy, the presence or absence of risk factors for multidrug-resistant (MDR) pathogen antibiotics or hospitalization in the previous 90 days (current hospitalization ≥5 days, mechanical ventilation ≥5 days, support in a dialysis center or residency in a nursing home), local epidemiological data, and finally knowledge that the patient is already known as being colonized by a MDR pathogen. In these situations, a broad-spectrum antibiotic will be recommended immediately, with the association of a β-lactam/β-lactamase inhibitor or an antipseudomonal cephalosporin, and an aminoglycoside or an antipseudomonal fluoroquinolone for 3 to 5 days. An initial antibiotic with a narrow spectrum will be accepted in the face of early onset pneumonia (mechanical ventilation <5 days) and in the absence of risk factors for MDR pathogens and in accordance to the marketing authorization (AMM), contraindications, dosing recommendations, and drug interactions (http://​base-donnees-publique.​medicaments.​gouv.​fr/​). Investigators are strongly encouraged to convert this initial regimen into a narrow-spectrum therapy, based on culture results. All antibiotics would be withdrawn, either at the end of day 8 or day 15, according to the randomization assignment, except those prescribed for a documented pulmonary infection recurrence or a new extrapulmonary infection before that day. The patient will be randomized just after antibiotic susceptibility test results and assigned to the SD arm or the LD arm. Screening of MDR pathogens will be realized with a surveillance culture of swab samples from the rectum for extended-spectrum β-lactam-producing Enterobacteriaceae (ESBL) and of swab samples from the anterior nares for methicillin-resistant Staphylococcus aureus (MRSA) at the patient’s admission in the ICU, and then once weekly until the patient’s discharge from the ICU (see Figs. 1 and 2).

Dr. Adrien Bouglé, coordinating investigator, Hôpital Universitaire La Pitié-Salpêtrière (France), is responsible for the study. Pr. Julien Amour is the scientific director, and the methodological support will be provided by the Clinical Research Unit Est (Pr. Tabassome Simon, Hôpital Saint Antoine, APHP, Paris, France). A steering committee includes the coordinating investigator, the scientific director, nine French ICU experts, and the head of the Clinical Research Unit (see Table 1 for the composition and affiliations of the members of the steering committee). This committee will control the proper conduct of the trial in terms of methodological, ethical, and logistical aspects.

The study will be performed in 39 French adult ICUs (Table 1). This work is supported by institutional grants from the French 2014 Programme Hospitalier de Recherche Clinique National (PHRC P140923 – AOM14515). The study is sponsored by the Clinical Research and Development Department of Assistance Publique – Hôpitaux de Paris. The sponsor had no role in the study design, collection, management, analysis, or interpretation of data, writing of the report, or the decision to submit the report for publication. The total duration of the study will be 27 months, including the inclusion period (24 months) and the duration of participation per patient (90 days).

The secondary objectives will be to compare between short and long durations of antibiotics regarding:

The primary assessment criterion will be a composite endpoint combining day 90 mortality and PA-VAP recurrence rate during hospitalization in the ICU (within 90 days). Recurrence will be defined with a post hoc analysis by two independent experts with predefined criteria: clinical suspicion of VAP (≥ two criteria including: fever >38.5 °C, leukocytosis >10⁹/L or leukopenia <4.10⁸/L, purulent tracheobronchial secretions, and a new or persistent infiltrate on chest radiography) associated with a positive quantitative culture of a respiratory sample (bronchoalveolar lavage fluid (significant threshold ≥10⁴ CFU/mL), plugged telescopic catheter (significant threshold ≥10³ CFU/mL), or quantitative endotracheal aspirate distal pulmonary secretion samples (significant threshold ≥10⁶ CFU/mL)). In cases of disagreement between the two experts, a third expert will reach a consensus. Thereby, the percentage of patients with PA-VAP recurrence and/or death will be assessed.

Secondary assessment criteria will be 30 day and 90 day mortality rate, PA-VAP recurrence rate, mechanical ventilation (MV)-free days, ICU-free days, antibiotic-free days during the hospitalization in the ICU, number and types of extrapulmonary infections during the hospitalization in the ICU, and acquisition of MDR pathogens during the hospitalization in the ICU (from swab sample of rectum and anterior nares).

All efficacy parameters will be collected during hospitalization of the patient in the ICU. In case of hospital discharge, phone calls will be made by the investigators or the research team with at least three attempts to contact the patient, his/her physician, or the patient’s emergency contact in a period of 10 days at different times of the day. If there is no answer, vital statistics will be collected via contact of the town council of the patient’s birthplace.

Data will be collected by the investigators and their teams in an electronic case report form (e-CRF). Data will be monitored by a clinical research associate (CRA) and checked for consistency and missing values by a data manager.

Centralized blocked-balanced randomization, in a 1:1 ratio, will be computer-generated by the clinical research unit (URC-EST). Randomization will be stratified on center. Investigators are blinded to the block size.

The study is designed to demonstrate the non-inferiority on the composite endpoint of mortality and recurrence at 90 days of the 8-day strategy versus the 15-day strategy for PA-VAP. Considering the mortality and/or recurrence of PA-VAP rate of 35.7% [7], 284 patients will be required in each group to achieve a power of 80% to exclude a 10% difference between the two groups with an α risk of 5%. To account for a possible 5% patient loss to follow-up, we planned to enroll 600 patients.

Baseline characteristics of patients will be described per group. Qualitative data will be described with frequencies and percentages; quantitative data will be described with mean and standard error or with median, interquartile interval, and range. Since this is a non-inferiority study, analysis of principal criteria will be performed on a per protocol population. The per protocol population will be defined as randomized patients without major protocol deviations such as non-respect of all selection criteria, clear non-respect of the antibiotic duration allocated (+/– 24 hours), missing data for the primary endpoint, and major protocol deviation(s) identified during a blinded data review before data base freezing. Non-inferiority will be tested by a Dunnett and Gent chi-squared procedure [21]. Secondary analysis in an intent-to-treat (ITT) population will also be performed.

The study will be conducted in accordance with the Helsinki Declaration and French laws and regulations. The study has received its approval from the French ethics committee (Comité de Protection des Personnes Ile de France VI) as well from the French Drug Safety Agency (Agence Nationale de Sécurité du Médicament et des Produits de Santé) (EudraCT 2015-003102-17).

All records and subjects’ identities will remain confidential in accordance with the following French regulations: the French National Committee of Informatics and Liberties and the French Consultative Committee for Data Processing in Health Research rules. The trial protocol was written following the Recommendations for Interventional trials (SPIRIT) Checklist (see Additional file 1).