SONIA — Pragmatic Trial of Glucocorticoids for CAP (Kenya)

1. Publication

• Title: A Pragmatic Trial of Glucocorticoids for Community-Acquired Pneumonia (SONIA)
• Acronym: SONIA
• Year & Journal: New England Journal of Medicine, epublished October 29, 2025 (print: 2025;393(22):2187-2197)
• Citation: 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. Context & Rationale

Background: Adjunctive glucocorticoids may reduce mortality in severe CAP in well-resourced settings (ICU-level care with advanced diagnostics/monitoring), but whether this benefit extends to low-resource settings with limited diagnostic and treatment facilities — where CAP mortality is often even higher — was unknown.

Research Question/Hypothesis: In adults with CAP admitted to Kenyan general medical wards (not ICU-level care), without a clear separate indication for glucocorticoids, does adding oral low-dose glucocorticoids to standard care reduce mortality compared with standard care alone?

Why This Matters: Nearly all prior adjunctive-corticosteroid-for-pneumonia evidence came from high-income, ICU-capable settings; this is described as the first pragmatic trial testing this inexpensive, widely available intervention specifically in a resource-limited general-ward context where most of the world's CAP burden actually occurs.

3. Design & Methods

• Study Type: Pragmatic, randomized, controlled, open-label trial
• Setting & Centers: 18 public hospitals, Kenya (general medical wards, not ICU)
• Population:
• Inclusion: Adult patients diagnosed with CAP, without a clear separate indication for glucocorticoids, enrolled within 48 hours of admission
• Exclusions: Clear indication for glucocorticoids (e.g., another condition requiring steroids)
• Intervention: Standard CAP care plus oral low-dose glucocorticoids for 10 days
• Comparator: Standard CAP care alone
• Blinding: Open-label
• Statistical Power & Follow-Up: Primary outcome: 30-day mortality.

4. Key Results

Note on data completeness: The exact absolute mortality percentages, relative risk/hazard ratio with 95% CI, and p-value were not available in the accessible source text beyond the "16% relative mortality reduction" figure reported in secondary coverage (CHEST Physician); readers should consult the primary NEJM publication for full statistical detail.

5. Internal Validity Assessment

• Randomization & Allocation: Randomized across 18 Kenyan public hospitals; specific concealment mechanics not detailed in available trial text.
• Protocol Adherence & Separation: Pragmatic, open-label design reflecting real-world general-ward administration of a simple oral medication (10-day course) — high real-world adherence plausibility given the simplicity of the intervention.
• Blinding & Detection Bias: Open-label — a genuine limitation, particularly relevant given a companion critical-care-literature critique ("Corticosteroids and community-acquired pneumonia: Africa deserves an explanatory trial," Critical Care 2026) specifically flagging the association between lack of blinding and mortality-outcome bias in critical care RCTs as an interpretive concern for this trial.
• Missing Data & Sensitivity Analyses: Not detailed in available trial text.
• Overall Internal Validity Conclusion: Moderate — pragmatic, real-world design is a genuine strength for generalizability, but the open-label design has been specifically flagged by independent commentary as a meaningful potential source of bias for this mortality-outcome trial, given known associations between lack of blinding and mortality-effect-size inflation in critical care RCTs more broadly.

6. External Validity Assessment

• Population Representativeness: Real-world Kenyan general medical ward CAP population — a population and care setting (limited diagnostics, no ICU-level monitoring) essentially unstudied by prior high-income, ICU-focused corticosteroid-for-pneumonia trials.
• Practice Context: Directly tests feasibility and effectiveness in the exact resource-limited setting where most global CAP mortality occurs — a major external validity strength distinguishing this from virtually all prior corticosteroid-CAP evidence.
• Overall External Validity Conclusion: Excellent for low-resource, general-ward CAP settings specifically — this is precisely the population and context most prior trials have failed to study, making this trial's generalizability to global (rather than just high-income) CAP burden unusually strong.

7. Strengths & Limitations

Strengths:

• First pragmatic trial of adjunctive glucocorticoids for CAP specifically in a low-resource, general-ward (non-ICU) setting
• Addresses a population bearing a disproportionate share of global CAP mortality but almost entirely absent from prior trial evidence
• Simple, cheap, widely available intervention (oral low-dose glucocorticoids) with high real-world implementation feasibility
• Locally-led research capacity building (explicitly highlighted by senior author as demonstrating Kenya's ability to conduct high-standard, practice-influencing research)

Limitations:

• Open-label design, specifically flagged by independent commentary as a potential source of mortality-outcome bias
• Exact quantitative primary-outcome statistics not accessible for this summary
• Single-country (Kenya) trial; generalizability to other resource-limited settings with different CAP etiology profiles (e.g., differing TB/HIV prevalence) requires some caution

8. Interpretation & Practice Impact

• Clinical Implications: Supports considering adjunctive low-dose oral glucocorticoids as an inexpensive, accessible mortality-reducing intervention for CAP in resource-limited general-ward settings — addressing a population where mortality is already high and treatment options are limited.
• Mechanistic Coherence: Consistent with the broader corticosteroid-in-severe-pneumonia literature (e.g., ICU-based evidence, RECOVERY dexamethasone-in-COVID-19 precedent) extending the concept to a general-ward, resource-limited context.
• Systems-Level Takeaway: As senior author Anthony Etyang noted, this trial demonstrates that locally-led African research can be conducted "to a high standard" and "influence practice around the world" — explicitly hoping the publication and accompanying NEJM editorial will help convince funders to support more such contextually relevant trials.

9. Controversies & Subsequent Evidence

• Editorial Commentary: The accompanying NEJM editorial (Kwizera, Dünser, "Glucocorticoids for pneumonia in Africa — old therapy, new context") frames this as extending a well-established high-income-setting therapy into a genuinely new context. A separate, more critical commentary in Critical Care ("Corticosteroids and community-acquired pneumonia: Africa deserves an explanatory trial") argues the open-label design and pragmatic trial format leave important mechanistic and bias-related questions unanswered, calling for a more tightly controlled explanatory (rather than purely pragmatic) follow-up trial.
• Guideline Integration: Not yet formally incorporated into WHO or regional CAP treatment guidelines at time of this handbook's compilation, but represents an influential, locally-generated evidence base directly relevant to global CAP treatment policy.

10. Summary & Executive Takeaway

Summary: This pragmatic, open-label RCT across 18 Kenyan public hospitals randomized adult CAP patients on general medical wards (not ICU) to standard care plus 10 days of oral low-dose glucocorticoids or standard care alone. Glucocorticoids reduced 30-day mortality by approximately 16% (relative reduction, per secondary reporting), extending prior high-income ICU-based corticosteroid-CAP evidence into a resource-limited general-ward context for the first time.

Overall Takeaway: SONIA provides important, locally-generated evidence that adjunctive low-dose glucocorticoids can reduce CAP mortality even in resource-limited, non-ICU general-ward settings — a potentially significant, cheap, and scalable practice change for the global majority of CAP patients who are never treated in an ICU, though the open-label design has drawn some independent methodological critique calling for a more explanatory follow-up trial.

11. Bibliography

• Kwizera A, Dünser MW. Glucocorticoids for pneumonia in Africa — old therapy, new context [editorial]. N Engl J Med. 2025;393(22):2263-2264.
• Sweeney DA, Kalil AC. Corticosteroids and community-acquired pneumonia: Africa deserves an explanatory trial [commentary]. Crit Care. 2026;30(1). doi:10.1186/s13054-026-05881-6
• RECOVERY Collaborative Group. Dexamethasone in Hospitalized Patients with Covid-19. N Engl J Med. 2021;384(8):693-704.