Sex-specific outcome disparities in very old patients admitted to intensive care medicine: a propensity matched analysis
Wernly, Bernhard; Bruno, Raphael Romano; Kelm, Malte; Boumendil, Ariane; Morandi, Alessandro; Andersen, Finn Husøy; Artigas, Antonio; Finazzi, Stefano; Cecconi, Maurizio; Christensen, Steffen; Faraldi, Loredana; Lichtenauer, Michael; Muessig, Johanna M.; Marsh, Brian; Moreno, Rui; Oeyen, Sandra; Öhman, Christina Agwald; Pinto, Bernadro Bollen; Soliman, Ivo W.; Szczeklik, Wojciech; Niederseer, David; Valentin, Andreas; Watson, Ximena; Leaver, Susannah; Boulanger, Carole; Walther, Sten; Schefold, Jörg C.; Joannidis, Michael; Nalapko, Yuriy; Elhadi, Muhammed; Fjølner, Jesper; Zafeiridis, Tilemachos; De Lange, Dylan W.; Guidet, Bertrand; Flaatten, Hans; Jung, Christian
Journal article, Peer reviewed
MetadataShow full item record
Original versionScientific Reports. 2020, 10, 18671 10.1038/s41598-020-74910-3
Female and male very elderly intensive patients (VIPs) might differ in characteristics and outcomes. We aimed to compare female versus male VIPs in a large, multinational collective of VIPs with regards to outcome and predictors of mortality. In total, 7555 patients were included in this analysis, 3973 (53%) male and 3582 (47%) female patients. The primary endpoint was 30-day-mortality. Baseline characteristics, data on management and geriatric scores including frailty assessed by Clinical Frailty Scale (CFS) were documented. Two propensity scores (for being male) were obtained for consecutive matching, score 1 for baseline characteristics and score 2 for baseline characteristics and ICU management. Male VIPs were younger (83 ± 5 vs. 84 ± 5; p < 0.001), less often frail (CFS > 4; 38% versus 49%; p < 0.001) but evidenced higher SOFA (7 ± 6 versus 6 ± 6 points; p < 0.001) scores. After propensity score matching, no differences in baseline characteristics could be observed. In the paired analysis, the mortality in male VIPs was higher (mean difference 3.34% 95%CI 0.92–5.76%; p = 0.007) compared to females. In both multivariable logistic regression models correcting for propensity score 1 (aOR 1.15 95%CI 1.03–1.27; p = 0.007) and propensity score 2 (aOR 1.15 95%CI 1.04–1.27; p = 0.007) male sex was independently associated with higher odds for 30-day-mortality. Of note, male gender was not associated with ICU mortality (OR 1.08 95%CI 0.98–1.19; p = 0.14). Outcomes of elderly intensive care patients evidenced independent sex differences. Male sex was associated with adverse 30-day-mortality but not ICU-mortality. Further research to identify potential sex-specific risk factors after ICU discharge is warranted.