Venoarterial extracorporeal membrane oxygenation in immunocompromised patients with cardiogenic shock: a cohort study and propensity-weighted analysis.

PURPOSE: The outcomes of immunocompromised patients with cardiogenic shock treated with venoarterial extracorporeal membrane oxygenation (VA-ECMO) are seldom documented, making ECMO candidacy decisions challenging. This study aims (1) to report outcomes of immunocompromised patients treated with VA-ECMO, (2) to identify pre-ECMO predictors of 90-day mortality, (3) to assess the impact of immunodepression on 90-day mortality, and (4) to describe the main ECMO-related complications. METHODS: This is a retrospective, propensity-weighted study conducted in two French experienced ECMO centers. RESULTS: From January 2006 to January 2022, 177 critically ill immunocompromised patients (median (interquartile range, IQR) age 49 (32-60) years) received VA-ECMO. The main causes of immunosuppression were long-term corticosteroids/immunosuppressant treatment (29%), hematological malignancy (26%), solid organ transplant (20%), and solid tumor (13%). Overall 90-day and 1-year mortality were 70% (95% confidence interval (CI) 63-77%) and 75% (95% CI 65-79%), respectively. Older age and higher pre-ECMO lactate were independently associated with 90-day mortality. Across immunodepression causes, 1-year mortality ranged from 58% for patients with infection by human immunodeficiency virus (HIV) or asplenia, to 89% for solid organ transplant recipients. Hemorrhagic and infectious complications affected 39% and 54% of patients, while more than half the stay in intensive care unit (ICU) was spent on antibiotics. In a propensity score-weighted model comparing the 177 patients with 942 non-immunocompromised patients experiencing cardiogenic shock on VA-ECMO, immunocompromised status was independently associated with a higher 90-day mortality (odds ratio 2.53, 95% CI 1.72-3.79). CONCLUSION: Immunocompromised patients undergoing VA-ECMO treatment face an unfavorable prognosis, with higher 90-day mortality compared to non-immunocompromised patients. This underscores the necessity for thorough evaluation and careful selection of ECMO candidates within this frail population.

Systemic lupus erythematosus-related acute lung disease.

INTRODUCTION: Systemic lupus erythematosus (SLE) is non-organ specific autoimmune disease with mainly skin, joint, and kidney involvement. SLE-related acute lung disease (ALD) is rare, poorly investigated and can lead to acute respiratory failure. We conducted a retrospective study aiming to describe clinical features, treatments and outcome of SLE-related APD. METHODS: We retrospectively included all patients with SLE and ALD admitted from November 1996 and September 2018 to La Pitié-Salpêtrière Hospital, after exclusion of viral or bacterial lung infection, cardiac failure or any other alternate diagnosis. RESULTS: During the time of the study, 14 patients with 16 episodes were admitted to our center: female 79%, mean age ± SD at admission 24 ± 11 years. ALD was inaugural of the SLE in 70% cases. SLE main organ involvement were: arthritis 93%, skin 79%, serositis 79%, hematological 79%, kidney 64%, neuropsychiatric 36% and cardiac 21%. 11 episodes required ICU admission for a median time of 8 days. Chest CT-scan revealed mostly basal consolidation and ground-glass opacities. When available, bronchoalveolar lavage mostly revealed a neutrophilic alveolitis with alveolar hemorrhage in 67% cases. Symptomatic respiratory treatments were: oxygen 81%, high-flow nasal canula oxygen 27%, non-invasive ventilation 36%, mechanical ventilation 64% and venovenous extracorporeal membrane oxygenation 18%. SLE-specific treatments were: corticosteroids 100%, cyclophosphamide 56% and plasma exchange 25%. All patients but one survived to ICU and hospital discharge. Two patients had a relapse of SLE-related ALD but none had interstitial lung disease during follow-up. CONCLUSION: Systemic lupus erythematosus-related acute respiratory failure is a severe event, mostly occurring at SLE onset, typical harboring a basal consolidation pattern on chest CT-scan and alveolar hemorrhage on BAL pathological examination. Mortality in our cohort is lower than previously reported but these results needs to be confirmed in further larger studies.