Marco Giani1,2, Gennaro Martucci3, Fabiana Madotto4, Mirko Belliato5, Vito Fanelli6, Eugenio Garofalo7, Clarissa Forlini1, Alberto Lucchini8, Giovanna Panarello3, Nicola Bottino9, Alberto Zanella10,11, Francesca Fossi12, Alfredo Lissoni9, Nicola Peroni5, Luca Brazzi13, Giacomo Bellani14, Paolo Navalesi15,16, Antonio Arcadipane3, Antonio Pesenti17,9, Giuseppe Foti18,8, Giacomo Grasselli19,9. 1. Università degli Studi di Milano-Bicocca, 9305, School of Medicine and Surgery, Monza, MB, Italy. 2. ASST di Monza, 189743, Emergency and Intensive Care, Monza, MB, Italy; marco.giani@unimib.it. 3. ISMETT, 18326, Department of Anesthesia and Intensive Care, Palermo, Sicilia, Italy. 4. IRCCS MultiMedica, 46842, Value-based Health Care Unit, Sesto San Giovanni, Lombardia, Italy. 5. Fondazione IRCCS Policlinico San Matteo, 18631, Pavia, Lombardia, Italy. 6. Università degli Studi di Torino Dipartimento di Scienze Chirurgiche, 202520, Department of Anesthesia and Critical Care - AOU Città della Salute e della Scienza di Torino, University of Turin, Turin, IT, Torino, Piemonte, Italy. 7. Mater Domini University Hospital, 220604, Department of Medical and Surgical Sciences, Catanzaro, Calabria, Italy. 8. ASST di Monza, 189743, Emergency and Intensive Care, Monza, MB, Italy. 9. La Fondazione IRCCS Ca' Granda Ospedale Maggiore di Milano Policlinico, 9339, Department of Anesthesia, Critical Care and Emergency, Milano, Lombardia, Italy. 10. La Fondazione IRCCS Ospedale Maggiore Policlinico, 9339, Milano, Lombardia, Italy. 11. University of Milan, 9304, Pathophysiology and Transplantation, Milano, Lombardia, Italy. 12. Azienda Ospedaliera Niguarda Ca Granda, 9338, Department of Anesthesia, Critical Care and Pain Medicine, Milano, Lombardia, Italy. 13. Università degli Studi di Torino, 9314, Department of Surgical Science, Division of Anesthesia and Critical Care Medicine, Torino, Piemonte, Italy. 14. Università degli Studi di Milano-Bicocca, 9305, Dipartimento di Scienze della Salute, Milano, Lombardia, Italy. 15. Padua University Hospital, 18624, Anesthesia and Intensive Care Unit, Padova, Veneto, Italy. 16. University of Padua, 9308, Department of Medicine, Padova, Veneto, Italy. 17. Universita degli Studi di Milano, 9304, Department of Pathophysiology and Transplantation, Milano, Italy. 18. Università degli Studi di Milano-Bicocca, 9305, School of Medicine and Surgery, Milano, Lombardia, Italy. 19. University of Milan, 9304, Department of Pathophysiology and Transplantation, Milano, Lombardia, Italy.
Abstract
RATIONALE: Prone positioning reduces mortality in severe ARDS patients. To date no evidence supports the use of prone positioning during venovenous extracorporeal oxygenation (ECMO). OBJECTIVES: Aim of the study was to assess feasibility, safety and effect on oxygenation and lung mechanics of prone positioning during ECMO. As a secondary exploratory aim, we assessed the association between PP and hospital mortality. METHODS: We performed a multicenter retrospective cohort study in six italian ECMO centers, including patients managed with prone positioning (PP) during ECMO support (prone group, four centers) and patients managed in the supine position (control group, two centers). Physiological variables were analyzed at 4 time points (supine before PP, start of PP, end of PP, supine after PP). The association between prone positioning and hospital mortality was assessed by multivariate analysis and propensity score matching. RESULTS: 240 patients were included, 107 in the prone group and 133 in the supine group. Median duration of the 326 pronation cycles was 15 [12-18] hours. Minor reversible complications were reported in 6% of prone positioning maneuvers. Prone positioning improved oxygenation and reduced intrapulmonary shunt. Unadjusted hospital mortality was lower in the prone group (34 vs 50%, p=0.017). After adjusting for covariates, prone positioning remained significantly associated with a reduction of hospital mortality (OR=0.50, 95%CI: 0.29-0.87). 66 propensity score-matched patients were identified in each group. In this matched sample, patients who underwent pronation had higher ECMO duration (16 vs10 days, p-value=0.0344) but lower hospital mortality (30% vs 53%, p=0.0241). CONCLUSION: Prone positioning during ECMO improved oxygenation and was associated with a reduction of hospital mortality.
RATIONALE: Prone positioning reduces mortality in severe ARDSpatients. To date no evidence supports the use of prone positioning during venovenous extracorporeal oxygenation (ECMO). OBJECTIVES: Aim of the study was to assess feasibility, safety and effect on oxygenation and lung mechanics of prone positioning during ECMO. As a secondary exploratory aim, we assessed the association between PP and hospital mortality. METHODS: We performed a multicenter retrospective cohort study in six italian ECMO centers, including patients managed with prone positioning (PP) during ECMO support (prone group, four centers) and patients managed in the supine position (control group, two centers). Physiological variables were analyzed at 4 time points (supine before PP, start of PP, end of PP, supine after PP). The association between prone positioning and hospital mortality was assessed by multivariate analysis and propensity score matching. RESULTS: 240 patients were included, 107 in the prone group and 133 in the supine group. Median duration of the 326 pronation cycles was 15 [12-18] hours. Minor reversible complications were reported in 6% of prone positioning maneuvers. Prone positioning improved oxygenation and reduced intrapulmonary shunt. Unadjusted hospital mortality was lower in the prone group (34 vs 50%, p=0.017). After adjusting for covariates, prone positioning remained significantly associated with a reduction of hospital mortality (OR=0.50, 95%CI: 0.29-0.87). 66 propensity score-matched patients were identified in each group. In this matched sample, patients who underwent pronation had higher ECMO duration (16 vs10 days, p-value=0.0344) but lower hospital mortality (30% vs 53%, p=0.0241). CONCLUSION: Prone positioning during ECMO improved oxygenation and was associated with a reduction of hospital mortality.