“…The following variables were collected from electronic health records: (i) demographic data (age, gender, body mass index (BMI)); (ii) therapies at home (i.e., corticosteroids or O 2 -therapy); (iii) diabetes or chronic kidney injury; (iv) Oto score [ 22 ]; (v) lung-allocation score (LAS) [ 23 ]; (vi) underlying diseases leading to LT (see full description in Table 1 ); (vii) pre-existing recipient-related Gram-negative (GN) colonization; (viii) provenience (hospital, home); (ix) surgical characteristics (time of LT, time of graft ischemia, io fluid support and peri/po surgical revisions, bleeding needing surgery and thromboembolic/ischemic events); (x) io use of ‘prophylactic’, ‘rescue’ or ‘prolonged’ ECMO [ 1 , 5 , 6 , 20 ]; (xi) immunosuppressive therapy; (xii) length of invasive mechanical ventilation; (xiii) Clavien-Dindo score [ 24 , 25 ]; and (xiv) short- and mid-term outcomes of interest (72-h PGD), perioperative blood units (transfused within 72–96 h after LT), ICU length of stay (LOS), re-tracheal intubation and/or tracheostomy, AKI (only stage 2 or 3, according to the KDIGO guidelines) and/or renal replacement therapy, multi-drug resistant (MDR)/extended-beta lactamases (ESBL) gram-negative bacteria, acute cellular rejection within 30 days after LT (according to the International Society for Heart and Lung Transplantation criteria), and hospital (H) LOS and mortality [ 14 , 15 , 24 , 25 , 26 , 27 ].…”