Hydrocortisone cuts mortality risk in ICU patients with severe pneumonia

In patients admitted to the intensive care unit (ICU) for severe community-acquired pneumonia (CAP), those who received hydrocortisone had a lower mortality risk compared with those on placebo, findings from the CAPE COD* trial suggest.

“CAP remains a major public health issue … In this large multicenter trial, early hydrocortisone therapy reduced the rate of death by day 28 among patients who had been admitted to the ICU for severe CAP,” said the researchers.

A total of 795 adults (median age 67 years, 69 percent male) were randomized 1:1 to receive IV hydrocortisone 200 mg QD for either 4 or 8 days depending upon clinical improvement, followed by tapering for 8 or 14 days, or placebo. All participants received standard therapy, including antibiotics and supportive care. [N Engl J Med 2023;388:1931-1941]

By day 28, only 25 patients in the hydrocortisone arm died as opposed to 47 in the placebo arm. This translated to 6.2 percent and 11.9 percent, respectively, yielding an absolute between-arm difference of −5.6 percentage points (p=0.006).

By day 90, mortality rate was still lower in the hydrocortisone vs the placebo arm (9.3 percent vs 14.7 percent; absolute difference, –5.4 percentage points).

 

Subgroup analysis

In patients who were not on any mechanical ventilation at baseline, endotracheal intubation frequency was lower in the hydrocortisone vs the placebo arm (18.0 percent vs 29.5 percent; HR, 0.59) as was noninvasive ventilation by day 28 (6.8 percent vs 10.9 percent; HR, 0.60).

Similarly, the cumulative incidence of invasive mechanical ventilation before day 28 was lower with hydrocortisone vs placebo among those who had not received endotracheal intubation at baseline (19.5 percent vs 27.7 percent; HR, 0.69).

A similar effect favouring hydrocortisone over placebo was observed in the subgroup of patients who had not received vasopressors at baseline (cumulative incidence of vasopressor initiation by day 28, 15.3 percent vs 25.0 percent; HR, 0.59).

 

Adverse events

Of the 169 serious adverse events (AEs) reported during the first 28 days after randomization, 99 occurred in the placebo arm. The hydrocortisone and placebo arms had similar rates of ICU-acquired infections (9.8 percent vs 11.1 percent; HR, 0.87; p=0.54) and gastrointestinal bleeding by day 28 (2.2 percent vs 3.3 percent; HR, 0.68; p=0.38).

There was, however, an increased incidence of hyperglycaemia in the hydrocortisone vs the placebo arm. By day 7, median daily insulin dose in patients on insulin therapy was higher in the treatment vs the placebo arm (35.5 vs 20.5 IU/day; p<0.001). Nonetheless, this aligns with the reported pharmacodynamic effects of glucocorticoids, the researchers noted. “Such increases are usually transient, [but] we did not verify this in the trial.”

 

Inconsistent results

However, the current findings diverge from evidence showing that methylprednisolone did not reduce mortality in ICU-admitted CAP patients relative ot placebo. [JAMA 2015;313:677-686, Intensive Care Med 2022;48:1009-1023] According to the researchers, the differences may have been driven by factors such as the different pharmacodynamic properties of glucocorticoids and exclusion of patients with septic shock in the study.

“[Moreover,] the very short median time between ICU admission and first administration of hydrocortisone or placebo in our trial (<15 hours) may have promoted an early effect,” they added.

Also, about a third of participants were women. This rate was higher than that seen in another trial showing that glucocorticoids did not alter mortality. “[This suggests] potential differences in response to glucocorticoids according to sex,” they continued.

Nonetheless, the findings suggest that in ICU-admitted CAP patients, early hydrocortisone treatment reduced 28-day mortality. “The results appeared to be consistent across important subgroups. Our data do not indicate any particular safety issues, including no between-group difference in the occurrence of hospital-acquired infections,” they concluded.