Luke · Annals of internal medicine 2006 · systematic review and meta-analysis · n=8 studies (1,703 patients)

Meta-analysis: convalescent blood products for Spanish influenza pneumonia: a future H5N1 treatment?

Cited 669 times in the scientific literature.

Level 3 - non-randomized controlled study

Systematic review and meta-analysis of non-randomized historical controlled studies

PubMed 16940336 · doi:10.7326/0003-4819-145-8-200610170-00139 · record verified 2026-08-29

What was done

Authors conducted a systematic review and meta-analysis of English-language studies published between 1918 and 1925 identified through manual searches. Eligible studies included at least 10 hospitalized patients treated with influenza-convalescent human blood products for Spanish influenza pneumonia and reported a control or comparison group. Two investigators independently extracted study characteristics, clinical outcomes, adverse events, and study quality.

What was found

Eight studies involving 1,703 patients were analyzed. The crude case-fatality rate was 16% (54 of 336) in treated patients compared with 37% (452 of 1,219) in untreated controls, with a pooled absolute risk difference in mortality of 21% (95% CI, 15% to 27%; range across studies: 8% to 26%). Early treatment (<4 days of pneumonia complications) had a crude case-fatality rate of 19% (28 of 148) compared with 59% (49 of 83) for late treatment (≥4 days), with a pooled absolute risk difference of 41% (95% CI, 29% to 54%). Adverse effects included chill reactions and possible symptom exacerbations in a few patients.

Why it matters

This synthesis provides historical evidence that convalescent blood products reduced mortality during the 1918 pandemic, supporting the rationale for clinical trials of convalescent plasma in emerging respiratory viral threats such as H5N1.

Limits

None of the included studies were randomized, blinded, or placebo-controlled. Confounding by indication was likely because treated patients were often selected due to more severe illness. The search was restricted to English-language literature from a manual search, and pertinent studies may have been missed.

Cited by