Recipient-specific preservation-fluid contamination, thoracoabdominal procurement and early donor-derived infectious events after liver and kidney transplantation: a retrospective donor-recipient cohort study

Preservation-fluid (PF) cultures after organ procurement are difficult to interpret because positive cultures may represent low-risk contamination or transplant-relevant donor-derived microbial signals. We assessed whether recipient-specific clinically relevant PF positivity (PF_CP), donor respiratory-PF concordance and thoracoabdominal procurement (TAO) were associated with composite early donor-derived infectious events after liver or kidney transplantation. This single-centre retrospective donor-recipient cohort used 2024 transplant data from the Affiliated Hospital of Qingdao University. After donor deduplication, 135 deceased donors contributed 354 liver or kidney graft-level observations; 352 were evaluable for the primary endpoint. The primary endpoint was a composite early donor-derived infectious event within 7 days after transplantation, including confirmed donor-derived infection (DDI), suspected/probable DDI and treated-potential donor-derived infectious events. Confirmed or suspected/probable DDI, excluding treated-potential events, was analysed as a restricted sensitivity endpoint. Associations were estimated using donor-clustered generalized estimating equations. Seventy-one composite early donor-derived infectious events occurred: 15 confirmed, 22 suspected/probable and 34 treated-potential events. Events were more frequent in PF_CP-positive than PF_CP-negative recipients (52/113, 46.0% versus 19/239, 7.9%). PF_CP was associated with the composite endpoint in the primary adjusted model (adjusted odds ratio [aOR] 10.28, 95% confidence interval [CI] 5.63–18.74). In the model including respiratory-PF concordance, PF_CP remained associated (aOR 6.11, 95% CI 2.82–13.23), and donor respiratory-PF concordance was associated with the endpoint (aOR 2.78, 95% CI 1.25–6.21). After excluding treated-potential events, PF_CP remained associated with confirmed or suspected/probable DDI (aOR 6.44, 95% CI 2.82–14.68). Post hoc PF-recipient review showed genus/species concordance in 15/109 organism-evaluable PF_CP observations, including three Candida and four carbapenem-resistant K. pneumoniae pairs. Exploratory analyses suggested possible effect modification by TAO (PF_CP x TAO aOR 4.60, 95% CI 1.29–16.41). Recipient-specific microbiologically high-risk PF contamination was associated with composite early donor-derived infectious events. Respiratory-PF concordance may identify a donor-related microbiological signal. Exploratory analyses suggested possible effect modification by TAO, but interaction findings were imprecise and should be considered hypothesis-generating. These findings support risk-stratified PF interpretation rather than reflexive treatment based on any positive culture.