Dequin et al. N Engl J Med. 2023 Mar 21
Corticosteroids exert powerful anti-inflammatory and immunomodulatory properties that may attenuate the consequences of severe community-acquired pneumonia (CAP), and thereby, improve clinical outcomes. Although many randomized controlled trials (RCTs) have been carried out to evaluate the potential beneficial effect of corticosteroids in CAP, reduced mortality was observed in a lone RCT of 46 patients by Confalonieri et al. nearly two decades ago (1). The findings of meta-analyses have also been conflicting. Briel et al. observed reduced time to stabilization among corticosteroid-treated patients with no significant effect on mortality (2). In contrast, decreased mortality was observed by Stern et al. in a meta-analysis that included open-label trials with a high risk of bias (3). During the Covid-19 pandemic, corticosteroids were extensively used, with RCTs demonstrating favorable results (4).
Considering the equipoise with corticosteroid use, the CAPE-COD investigators evaluated the impact of hydrocortisone on the 28-day mortality in patients admitted to the ICU with severe CAP (5). The study was carried out before the outbreak of the Covid-19 pandemic.
Population and design
The study population included adult patients admitted with severe CAP to the ICU across 31 centers in France. The diagnosis of pneumonia was based on clinical and radiological criteria. The extent of diagnostic testing to identify the pathogen was left to the discretion of the clinician. Severe CAP was defined by the presence at least one of four criteria – 1. Mechanical ventilation (non-invasive or invasive) with a PEEP ≥5 cm H2O, 2. Support with high-flow nasal cannula with a PaO2:FiO2 of <300, 3. Supplemental oxygen through a non-rebreather mask with a PaO2:FiO2 of <300, or 4. A score of >130 on the Pulmonary Severity Index.
Patients were randomized to receive hydrocortisone or placebo in a 1:1 ratio. Randomization was stratified by trial center and the requirement for mechanical ventilation.
Patients with a do-not-intubate order, influenza pneumonia, and septic shock were excluded.
Hydrocortisone, 200 mg/day, was administered as a continuous infusion within 24 hours after fulfilling the severity criteria described above. After 4 days of treatment, the total duration of therapy was determined to be 8 or 14 days, based on the clinical status as predefined. The dose of hydrocortisone was weaned down and ceased according to a prespecified protocol.
Intravenous saline was administered as placebo in identical packages, ensuring blinding of the study.
Standard treatment, including antibiotics and supportive, were administered to both groups. The medical team chose the modality of respiratory support according to their best judgment.
The study was powered to detect a 25% relative reduction in the 28-day mortality in the hydrocortisone group (from 27% to 20.25%). The required sample size was 1146 patients and was rounded off to 1200 patients allowing for possible withdrawals.
The CAPE-COD study started enrolling patients in October 2015. By March 2020, the Covid-19 pandemic broke out in France. The study was suspended and continued as an embedded trial renamed “CAPE-COVID” to evaluate the effect of hydrocortisone in patients with acute respiratory failure due to Covid-19 pneumonia. The results of the CAPE-COVID trial have hence been published. The investigators had planned to continue the original trial once the pandemic had settled. However, enrollment was ceased based on the recommendation of the data and safety monitoring board after the second interim analysis. By this time, 800 patients had been enrolled from 5948 who were screened for eligibility. In the final analysis, there were 400 patients in the hydrocortisone arm and 395 patients in the control arm.
Mechanical ventilation was carried out in 44.5% of patients in the hydrocortisone arm; 23% received invasive and 21.5%, non-invasive ventilation. In the control arm, 44.3% of patients were mechanically ventilated – 21.5% received invasive and 22.8, non-invasive ventilation. The main clinical outcomes are summarised in Table 1.
Table 1. Main clinical outcomes
|28-d mortality||25/400 (6%)||47/395 (11.9%)||P=0.006|
|90-d mortality||36/388 (9.3%)||57/389 (14.7%)||-5.4, 95% CI: −9.9 to −0.8|
|Intubation by 28 d||40/222 (18%)||65/220 (29.5%)||HR, 0.59 (0.40 to 0.86)|
|Vasopressors by 28 d||55/359 (15.3%)||86/344 (25%)||HR, 0.59 (0.43 to 0.82)|
The 28-day mortality – the primary outcome
Mortality at 28 days was significantly lower among patients who received hydrocortisone compared with placebo (6.2% vs. 11.9%; absolute difference, –5.6 points, p = 0.006)
The 90-day mortality was also lower in the hydrocortisone compared to the control group (9.3% vs. 14.7%; difference, –5.4, 95% confidence interval, −9.9 to −0.8). Among patients who were not on any type of mechanical ventilation at baseline, the incidence of invasive ventilation was lower with hydrocortisone compared with placebo (18% vs. 29.5%; hazard ratio 0.59, 95% CI, 0.4–0.86). Among patients who were not on invasive mechanical ventilation at baseline, the incidence of endotracheal intubation was lower with hydrocortisone. By day 28, the requirement for vasopressors was also lower among hydrocortisone treated patient.
The study reported the incidence of possible adverse outcomes related to corticosteroid administration. Overall, the incidence of hospital-acquired infections, including ventilator-associated pneumonia and blood stream infection, was similar between the two groups. The incidence of gastrointestinal bleeding was no higher with hydrocortisone compared with placebo. However, hyperglycemia requiring insulin administration was significantly more common among hydrocortisone-treated patients.
- Although the study was terminated before reaching the target sample size, it included 800 patients from multiple centers in France, making it one of the most extensive studies on the topic.
- Initiation of treatment was expeditious; the time interval between ICU admission to the administration of the first dose of corticosteroid was less than 15 hours.
- Patients were well-matched at baseline.
- Although a large number of patients were excluded after initial screening, exclusions were largely based on predefined criteria.
- Follow-up was nearly complete, with the loss of only two patients from the study.
- The study excluded patients with septic shock, considering the difference in pathophysiology and the largely proven benefit of corticosteroids in this situation.
- The mortality in the control arm (11.9%) was much lower than predicted (27%), suggesting less severe illness in the cohort. However, nearly 44% of patients underwent mechanical ventilation, and the median PaO2/FiO2 ratio was around 160 mm Hg, implying a high-risk population.
- There was no standard protocol to identify the etiological agent; no pathogen could be isolated in 44.9% of cases. Hence, it remains doubtful whether the beneficial effects would apply to different types of pneumonia (bacterial vs. viral).
- The authors did not evaluate the incidence of neuromuscular weakness, a well-known adverse effect of corticosteroid treatment in critically ill patients that may impact outcomes.
- The administration of hydrocortisone as a continuous infusion does not have a strong physiological rationale, although it is recommended by the surviving sepsis guidelines considering the reduced propensity to hyperglycemia.
- There is uncertainty regarding the optimal corticosteroid preparation; dexamethasone was found to be beneficial in Covid-19 pneumonia in the RECOVERY trial and in patients with ARDS in the DEXA-ARDS trial.
The beneficial effect of corticosteroids in CAP has been debated vigorously, with no firm evidence of improved survival, except as the secondary outcome of an RCT of a small sample size. The CAPE-COD trial, although prematurely terminated due to the Covid-19 outbreak, included 800 patients with severe CAP. This trial demonstrated a beneficial effect of hydrocortisone on the 28-day mortality and a reduced requirement for mechanical ventilation and vasopressor support among patients with severe CAP. The incidence of adverse events, including nosocomial infections and gastrointestinal bleeding, was similar in both groups.
1. Confalonieri M, Urbino R, Potena A, Piattella M, Parigi P, Puccio G, et al. Hydrocortisone infusion for severe community-acquired pneumonia: a preliminary randomized study. Am J Respir Crit Care Med. 2005 Feb 1;171(3):242–8.
2. Briel M, Spoorenberg SMC, Snijders D, Torres A, Fernandez-Serrano S, Meduri GU, et al. Corticosteroids in Patients Hospitalized With Community-Acquired Pneumonia: Systematic Review and Individual Patient Data Metaanalysis. Clin Infect Dis Off Publ Infect Dis Soc Am. 2018 Jan 18;66(3):346–54.
3. Stern A, Skalsky K, Avni T, Carrara E, Leibovici L, Paul M. Corticosteroids for pneumonia. Cochrane Database Syst Rev. 2017 Dec 13;12(12):CD007720.
4. RECOVERY Collaborative Group, Horby P, Lim WS, Emberson JR, Mafham M, Bell JL, et al. Dexamethasone in Hospitalized Patients with Covid-19. N Engl J Med. 2021 Feb 25;384(8):693–704.
5. Dequin PF, Meziani F, Quenot JP, Kamel T, Ricard JD, Badie J, et al. Hydrocortisone in Severe Community-Acquired Pneumonia. N Engl J Med. 2023 Mar 21;