Pathogen trajectory and prognosis of co-infection and secondary infection in hospitalized COVID-19 patients

Co-infection or secondary infection is a major risk factor for poor outcome of COVID-19 patients. However, the trajectory of pathogen accumulation and its independent effect on prognosis remain poorly characterized. We aimed to delineate the pathogen-detection trajectory of COVID−19-associated co-/secondary infections and quantify its impact on in-hospital mortality. This retrospective study reviewed the medical records of COVID-19 inpatients admitted to the Second Affiliated Hospital of Zhejiang University School of Medicine from November 2022 to February 2023. In this single-center retrospective cohort study, we included adult patients hospitalized with laboratory-confirmed COVID-19 who had microbiological surveillance data. Infections were classified as co-infection (pathogen detected within 48 h) or secondary infection (first detected after 48 h). Pathogen isolates were categorized into Gram-negative, Gram-positive and fungal groups and flagged for resistance. We defined four trajectory types (No infection, Sensitive-bacterial, Resistant-bacterial, Fungal-involved) and five burden-clusters by normalized daily detection shape. The primary outcome was in-hospital mortality; secondary outcomes were total length of stay, ICU length of stay, and duration of mechanical ventilation. Associations were assessed using multivariable logistic regression, with robustness checked by propensity score matching (PSM), inverse probability of treatment weighting (IPTW), time-fixed and time-dependent Fine-Gray competing-risk models. Among 428 patients, 333 (77.8%) developed at least one infection. Resistant Acinetobacter baumannii (55.9%), resistant Klebsiella pneumoniae (38.4%), fungi (33.0%) and resistant Pseudomonas aeruginosa (14.1%) predominated. Mortality rose monotonically across trajectory types: No infection 22.1%, Sensitive-bacterial 55.1%, Fungal-involved 60.9%, Resistant-bacterial 64.9% ( p < 0.001). After adjustment for age, SOFA, PaO2/FiO2, ICU admission and comorbidities, Sensitive-bacterial (aOR 2.87, 95%CI 1.29–6.37), Fungal-involved (aOR 3.15, 95%CI 1.56–6.39) and Resistant-bacterial (aOR 4.69, 95%CI 2.36–9.32) trajectories were all independently associated with in-hospital death (all p < 0.05). Secondary outcomes showed the same gradient: total, ICU and mechanical-ventilation length of stay were shortest in No-infection patients and longest in the Resistant-bacterial and Fungal-involved groups (all p < 0.001). The direction and magnitude were confirmed by PSM (high-risk vs. no-infection OR 2.05, 95%CI 1.29–3.27) and IPTW (Resistant-bacterial aOR 4.13, 95%CI 2.98–5.73). In time-dependent competing-risk analysis, infection onset raised instantaneous death hazard (Sensitive HR 3.23, Resistant HR 2.04, Fungal HR 1.86; all p < 0.05). Pathogen-detection trajectories dominated by resistant bacteria or fungi represent an independent, dose-dependent predictor of in-hospital mortality in COVID-19. Early recognition and targeted antimicrobial stewardship for high-burden trajectories may improve outcomes.