Customized Antibiotic Treatment Duration Among Hospitalized Patients With Moderately Severe Community-Acquired Pneumonia

Recent studies have suggested that community-acquired pneumonia (CAP) can be successfully treated with short-course antibiotic regimen when clinical improvement is rapidly obtained. Even if clinical response is obtained in 3 days in the majority of cases, it can vary widely among patients suggesting "one duration does not fit all". An individualised duration of therapy depending on the patient's response could help to ensure bacterial eradication while avoiding unnecessary antibiotic exposure and thus reduce antibiotic resistance. At present, this strategy has never been tested.

This is a phase III, pragmatic, non-inferiority randomised (1:1) national multicentre trial. Population of study participants will be patients admitted to hospital with suspected CAP and in need for antibiotics.

There is no need to establish a DSMB for this trial. The antibiotic treatments prescribed during this study are treatments used in current practice, with well-known adverse drug reactions. Furthermore, a 3-day treatment duration has already been studied in 2 previous randomized clinical trials, showing its safety for hospitalized CAP.

Statistical analysis:

Statistical inference for non-inferiority will be based on the confidence intervals (CI) of the difference in Day 15 cure proportions [proportion in reference arm - proportion in experimental arm] accounting for randomisation stratification factors (centre and delay from Day 0 to Day S (≤ 3 days or > 3 days).

The inferiority hypothesis will be rejected and non-inferiority will be claimed if the upper bound of the 95%CI of the difference is ≤ 10%.

Secondary efficacy endpoints evaluating the evolution of symptoms will be analysed using either a GMM or a GEE for categorical and continuous variables, respectively.

Other quantitative variables will be compared using the Student t-test (or a non-parametric test if the distribution remains skewed following transformation), while categorical variables will be analysed using either the Chi-squared or the Fisher-exact tests.

All statistical tests will be performed with a level of significance of 5%, except the primary endpoint which be analysed with a 2.5% level of significance.

Inclusion criteria

• Adult patient (≥18 years old)
• Admitted to hospital for suspected CAP defined by the presence of at least 2 of the following diagnostic clinical criteria:

• Fever (temperature > 38°C) or hypothermia (< 36°C)
• Dyspnea
• Cough
• Production of purulent sputum
• Crackles
• Radiological evidence of a new infiltrate (chest X-ray or CT scan)
• Negative viral respiratory testing
• Treated for a minimum of 48 hours with antibiotics (excluding azithromycin due to its prolonged half-life)Presenting with an early clinical response within the last 24 hours (up to 7 days after the start of antibiotic treatment, if planned treatment duration > 7 days), defined by the presence of all the following criteria:apyrexia (T°C ≤ 37.8)heart rate < 100/minrespiratory rate < 24/min, according to the patient's usual mode of oxygenation,arterial oxygen saturation ≥ 92%, according to the patient's usual mode of oxygenation,Patient currently under antibiotic treatment for his suspected CAP (i.e. the last dose of ATB has been administered to the patient less than 24 hours ago)
• Patient presenting a clinical response within the last 24 hours defined by the presence of all the following criteria:

• apyrexia (T°C ≤ 37.8)
• heart rate < 100/min
• respiratory rate < 24/min, according to the patient's usual mode of oxygenation,
• arterial oxygen saturation ≥ 92%, according to the patient's usual mode of oxygenation,
• systolic blood pressure ≥ 90mmHg, The last occurring of these criteria must have appeared within the last 24 hours.
• The antibiotic treatment for suspected CAP has started at least 48 hours ago and at most 6 days ago
• No other site of infection besides respiratory
• Affiliated to Health insurance
• Has given informed consent
• Patient understanding oral and written French, or presence of a relative who can explain and help him complete the study documents

Exclusion criteria

• Signs of severe CAP (abscess, massive pleural effusion, serious chronic respiratory insufficiency)
• Known immunosuppression (asplenia, neutropenia, agammaglobulinemia, transplant, myeloma, lymphoma, known HIV and CD4<200/mm3)
• Suspected or confirmed legionellosis
• Any other infection necessitating concomitant antibiotic treatment
• Confirmed or suspected aspiration pneumonia or healthcare-associated pneumonia
• Treatment of suspected CAP with azithromycin (due to its prolonged half-life)
• Concomitant steroid treatment (only for patients treated with fluoroquinolones antibiotics)
• Pre-existing aortic aneurysm or dissection, family history of aortic aneurysm or dissection, Marfan syndrome, Ehlers-Danlos syndrome, Takayasu arthritis, uncontrolled arterial hypertension, atherosclerosis (only for patients treated with fluoroquinolones antibiotics)
• Pregnant or breastfeeding woman
• Life expectancy < 1 month
• Patient under legal guardianship (French "tutelle" or "curatelle")
• Patient without fixed address
• Patient enrolled in another interventional clinical trial on CAP treatment