A pragmatic trial in Kenya confirmed that adjunctive glucocorticoids can reduce mortality in patients with community-acquired pneumonia (CAP) compared with usual care alone. The open-label randomized trial found that 10 days of low-dose glucocorticoids reduced 30-day mortality by 16%.
“In settings where mortality is already high, steroids are an inexpensive option that could help reduced mortality,” said senior author Anthony O. Etyang, MBChB, MMed, MSc, PhD, Head of Epidemiology and Demography and Interim Head of Clinical Trials for the Kenya Medical Research Institute–Wellcome Trust Research Programme, Kilifi, Kenya. “For high-income countries, there was already evidence that steroids are beneficial for severe pneumonia in ICU settings. Our results bolster the evidence that if steroids are useful even in low-income settings in a general ward setting, they’re probably also useful in high-income settings with severe pneumonia.”
CAP is a leading cause of morbidity and mortality globally, Dr. Etyang noted. The mortality rate in sub-Saharan Africa is up to five times higher than in high-income countries even though patients with CAP in sub-Saharan Africa tend to be younger. Lack of health care resources, limited access to care, and delayed hospitalization all contribute to more severe CAP in younger patients with elevated case mortality.
The Steroids for pneumonia in adults in Kenya (SONIA) trial grew out of clinical experience with COVID-19 in Kenya. Following publication of the Higher dose corticosteroids in patients admitted to hospital with COVID-19 who are hypoxic but not requiring ventilatory support (RECOVERY) trial showing survival benefits from dexamethasone, hospitals across Kenya began using steroids for all patients with pneumonia with or without proven COVID-19 infections.1–2
“The evidence was not so clear for people who did not have COVID-19, but the problem in Kenya was that limited testing was available,” Dr. Etyang said. “Clinicians reasoned that it might be COVID-19 or it might not, and we know steroids are useful in COVID-19. They may be useful in pneumonia, so let’s use them. That’s how we designed the trial.”
SONIA randomized 2,180 patients hospitalized for CAP across 18 hospitals in Kenya to either usual care (1,091 patients) or low-dose glucocorticoids (1,089 patients) for 10 days. Usual care included a beta lactam and a macrolide based on World Health Organization guidelines. Glucocorticoids included bioequivalent daily oral doses of dexamethasone, hydrocortisone, methylprednisolone, prednisolone, or prednisone, depending on local availability.
The median age of patients was 53 years, 46% were women, and 31.7% had oxygen saturations < 90% at admission. Patients were excluded if they had a clear or suspected indication for glucocorticoids such as asthma or COVID-19.
The primary outcome was all-cause mortality 30 days following enrollment. Secondary outcomes included death at 7, 14, and 21 days; death during hospitalization; and death following discharge up to 30 days after enrollment.
All-cause mortality was 26.0% in the usual care group vs 22.6% in the glucocorticoid group, for a hazard ratio of 0.84 (95% CI; 0.73-0.97, P = .02). Mortality risk at the secondary time points was consistent with the primary outcome.
The most common adverse events were pulmonary TB in both groups plus acute kidney injury (8.0%) in the usual care group and hyperglycemia (16.6%) in the glucocorticoid group.
“Many patients admitted to hospital get hyperglycemia even in the absence of steroids,” Dr. Etyang said. “This is something that needs to be monitored and managed in hospitals. But the risk of hyperglycemia also needs to be weighed against the much worse outcome of death from CAP.”
An accompanying editorial in The New England Journal of Medicine (NEJM) noted that SONIA is one of the rare trials to provide locally generated, context-relevant data for clinicians in Africa.3 The editorial advocates for researchers to take the lead in generating evidence to meet their own health care challenges.
Dr. Etyang added that 44 of the 46 SONIA authors are Kenyan, were trained in Kenya, and are practicing in Kenya, including both the lead and senior authors.
“SONIA shows that we are able to design studies that are relevant to our setting and conduct them to a high standard and influence practice around the world,” he said. “We hope publishing this study, and the NEJM editorial, will help convince funders.”
References
1. Lucinde RK, Gathuri H, Mwaniki P, et al. A pragmatic trial of glucocorticoids for community-acquired pneumonia. N Engl J Med. 2025;393(22):2187-2197. doi:10.1056/NEJMoa2507100
2. RECOVERY Collaborative Group, Horby P, Lim WS, et al. Dexamethasone in hospitalized patients with COVID-19. N Engl J Med. 2021;384(8):693-704. doi:10.1056/NEJMoa202143
3. Kwizera A, Dünser MW. Glucocorticoids for pneumonia in Africa – old therapy, new context. N Engl J Med. 2025;393(22):2263-2264. doi:10.1056/NEJMe2514533
