Thorsten Brenner, Annabell Skarabis, S Schaller, Thilo von Groote, Christian Putensen, Ulf Günther, Martin Sauer, Sebastian O. Decker, Fabian Dusse, Manfred Weiss, Klaudiusz Suchodolski, Tim-Philipp Simon, P B Rosenberger, Onnen Moerer, Matthias Unterberg, Jens‐Christian Schewe, Hendrik Bracht, Selina Hutzl, Manuel Feißt, Ursula Marschall, Patrick Brandenburg, Philip Stevens, Juliana Schmidt, Mathias W. Pletz, Marc M. Berger, on behalf of the German Society of Anaesthesiology and Intensive Care (GSAIC) Trials Group, Jens Brands, Alberto Antinoro, Linus Warner, M R Weigand, Dominique M. Hart, Gernot Marx, Jana Mossanen, T M Breuer, Helene Häberle, Kristina Nitzsche, Paul Thalmann, Jolanda Brezinski, Joachim Saam, Silke Grumaz, Maren Steinmann, Maxime Gaasch, Bastian Pasieka, Ulrike Jaekel, Timo Brandenburger, Friedhelm Bach, Haitham Mutlak, Stefan Utzolino, Wolfgang A. Krueger, Stefan Hagel, Christian Lanckohr
PURPOSE: Early pathogen detection is crucial in sepsis. We hypothesized that detection of microbial circulating cell-free DNA by metagenomic next-generation sequencing (mNGS) improves clinical outcomes and health-related quality of life without increasing healthcare costs. METHODS: This randomized, controlled, interventional, open-label, multicenter trial was conducted in 24 intensive care units across Germany. The intervention group (n = 200) received mNGS diagnostics in addition to standard-of-care microbiology, compared with standard-of-care microbiology alone (control group; n = 189). The primary endpoint was the Desirability of Outcome Ranking/Response Adjusted for Duration of Antibiotic Risk (DOOR/RADAR) score. RESULTS: The DOOR/RADAR score was not significantly improved at 28 days after sepsis onset (intervention group: 3.21 ± 1.54; control group: 3.49 ± 1.51; 95% CI - 0.58 to 0.03). However, other secondary endpoints were improved, including a reduced duration of mechanical ventilation (intervention group: 6.6 ± 9.4 days; control group: 9.3 ± 10.6 days; 95% CI - 5.03 to - 0.34) and faster shock resolution (intervention group: 6.9 ± 7.4 days; control group: 8.8 ± 8.5 days; 95% CI - 3.75 to - 0.04). Health-related quality of life at 90 days (EQ-5D-5L) was improved in the intervention group (0.312 ± 0.386) compared with the control group (0.208 ± 0.373; p = 0.047). In the subgroup with available claims data (33.2% of participating patients), healthcare costs over 180 days did not differ. CONCLUSION: The DOOR/RADAR score as primary endpoint was not significantly improved by mNGS. Exploratory secondary analyses revealed improvements in secondary endpoints. (Funding: German Innovation Fund; ClinicalTrials.gov number, NCT04571801, registration: 25.8.2020).