Researchers have assessed the impact antibiotic therapy on short (hospital) and long-term (6 months) outcomes of ICU patients with severe community-acquired pneumonia.


Community-acquired pneumonia (CAP) remains one of the leading causes of hospital admission, with 5% to 15% of cases being admitted to an ICU, largely due to complications like shock and respiratory failure. Severe CAP has a mortality rate of about 30%. Yet, despite several guidelines and studies supporting the use of antibiotic therapy for the management of severe CAP, doubts remain regarding the optimal treatment approach for patients with this infection.



To evaluate the impact of different aspects of antibiotic therapy—timing, mono versus combination therapy, macrolide use, appropriateness, and duration—on the short- and long-term outcomes of critically ill patients with severe CAP, my colleagues and I conducted the Infection on Admission to the Intensive Care Unit (INFAUCI) study. Published in the Journal of Critical Care, the study focused on determinants of outcome and raising potential areas of process of care that may produce outcome improvements.

CAP was the most common infection (13.3%) among patients admitted to the hospital and to ICUs. Consequently, even small improvements in its management could have important health gains. We used a large prospective database of patients admitted with infection to ICUs and segregated patients with CAP. Although this was not a randomized trial, it had the advantage of evaluating real-life patients treated in clinical practice.

Overall, hospital mortality (32%) and length of stay (median 28 days) of the infected population were high. These outcomes were not only related to the acute disease severity, but more importantly, with host and process of care features. Overall length of stay was 14 days in the ICU and 25 days in the hospital. Mortality was 34% in the hospital and 35% at 6 months. As such, strategies are needed that may improve these results. In this study, combination antibiotic therapy was administered to 76% of patients but was not associated with better outcomes. However, the sub-group who received a β-lactam in combination with a macrolide did have lower risks of hospital (odds ratio [OR], 0.17) and 6-month (OR, 0.21) mortality. These findings were independent of the presence of septic shock on admission to the ICU. Moreover, we did not find any evidence that prolonging antibiotic therapy more than 7 days lead to better outcomes (Table). According to this data, we believe that all patients with CAP who need admission to an ICU should receive combination antibiotic therapy with a macrolide.


Unanswered Questions

Several questions still need to be addressed in the management of patients with severe CAP:

  1. How long should combination antibiotic therapy be given; only the first few days or during the entire course of therapy?
  2. If a long-acting macrolide is selected (eg, azythromycin), can the time of macrolide be reduced (assuming sensitivity to the β-lactam)?
  3. Is the benefit of combination therapy independent of microbiological findings or only related to Pneumococcal pneumonia?
  4. Are there specific patients who might benefit more from a combination therapy approach? Would precision antibiotic therapy be appropriate in this patient population, and if so, what markers should clinicians look for?
  5. Should combination antibiotic therapy be given only in the setting of empirical therapy, or is there a benefit also during microbiological-guided therapy?

For now, we recommend that severe CAP patients receive combination antibiotic therapy that includes a macrolide and that microbiological documentation should be looked for, as appropriate antibiotic therapy was found to improve short-term survival in our study. Routine courses of appropriate antibiotic therapy longer than 7 days were not found to be associated with a survival benefit and should therefore be avoided.


Pereira J, Goncalves-Pereira J, Ribeiro O, et al. Impact of antibiotic therapy in severe community-acquired pneumonia: Data from the Infauci study. J Crit Care. 2018;43:183-189. Available at: