Ham, Kealy

Abstract Rationale Legionella species can cause severe community-acquired pneumonia requiring targeted antimicrobial therapy. The impact of time to effective Legionella-directed therapy on mortality is not well defined. Objectives To evaluate the association between time to effective Legionella-directed therapy and mortality in hospitalized patients with community-acquired Legionella pneumonia and to determine whether differences in treatment timing contribute to the excess mortality observed among immunocompromised patients. Methods A retrospective multicenter cohort study of adults hospitalized with laboratory-confirmed community-acquired Legionella pneumonia (2019–2025) was performed. Time from presentation to effective therapy (macrolide, fluoroquinolone, or doxycycline) was recorded. The primary outcome was 90-day mortality. Multivariable logistic regression analysis was performed, adjusting for age, SOFA-2 score (first 6 hours), and immunocompromised status. Secondary analyses included a 24-hour landmark Cox model for 90-day survival and multivariable logistic regression for 30-day mortality. Results Among 310 patients (median age 66.9 years, 44.8% immunocompromised), 90-day mortality was 17.4%. Every 6-hour delay in effective therapy was independently associated with increased 90-day mortality (adjusted OR per 6 hours: 1.11; 95% CI, 1.06–1.17; P < 0.001). Delays >24 hours were associated with worse 90-day survival (adjusted HR, 2.72; 95% CI, 1.65–4.77; P < 0.001). Time to effective therapy was also independently associated with increased odds of 30 day mortality (adjusted OR per 6 hours: 1.06; 95% CI: 1.01-1.11, p = 0.024). Immunocompromised patients experienced longer delays (median, 7.1 versus 3.3 hours; P < 0.001) and higher unadjusted 90-day mortality (22.3% versus 13.5%; P = 0.041), but immunocompromised status was not independently associated with mortality after adjusting for severity and treatment timing. Conclusion Longer time to effective Legionella-directed therapy was independently associated with increased mortality in a graded fashion. Delays were more common among immunocompromised patients, partially accounting for excess mortality in this group, suggesting that timely pathogen-directed therapy may represent a clinically important and potentially modifiable contributor to risk.

OBJECTIVES: Pneumocystis jirovecii pneumonia (PCP) can cause severe hypoxemic respiratory failure in immunocompromised patients. Contemporary outcomes of venovenous extracorporeal membrane oxygenation (VV ECMO) support for PCP are poorly characterized, and existing reports are limited to small case series. DESIGN: Retrospective cohort study. SETTING: International registry study using the Extracorporeal Life Support Organization (ELSO) registry. PATIENTS/SUBJECTS: Adults with PCP-associated hypoxemic respiratory failure supported with VV ECMO between 2011 and 2024. INTERVENTIONS: VV ECMO. MEASUREMENTS AND MAIN RESULTS: A total of 209 patients with PCP-associated respiratory failure supported with VV ECMO were identified. The overall in-hospital mortality was 60.8%. Survivors were younger than nonsurvivors (median age 41.9 vs. 46.8 yr; p = 0.047). Duration of mechanical ventilation before VV ECMO was longer among nonsurvivors (median 4.7 vs. 1.6 d; p = 0.019). The proportion of patients with HIV infection was similar among survivors and nonsurvivors (19.5% vs. 22.1%; p = 0.73). Pre-ECMO vasopressor use, prone positioning, and renal replacement therapy were common and did not differ between groups. Clinical course was characterized by prolonged VV ECMO support (median 18.6 d) and frequent complications, including pneumothorax (21.1%), renal replacement therapy during VV ECMO (31.6%), intracranial hemorrhage or stroke (7.7%), and major pulmonary or gastrointestinal hemorrhage (16.8%). PCP accounted for a small proportion of VV ECMO runs reported to the ELSO registry throughout the study period at 0.36% of reported cases. No significant temporal trend in mortality was observed. CONCLUSIONS: PCP-associated respiratory failure supported with VV ECMO is associated with substantial mortality and prolonged VV ECMO support. These findings provide contemporary benchmarks that may inform VV ECMO candidacy discussions and expectations in this challenging patient population.

In a large, multicenter cohort of HIV and non–HIV immunocompromised patients with Pneumocystis jirovecii pneumonia, chest CT most commonly demonstrated diffusely distributed ground-glass opacity, with no imaging features distinguishing between immunocompromise etiology.