Nurse-led respiratory care bundle and pneumonia prevention in critically ill patients with abdominal sepsis: a multicenter before-and-after controlled study

Patients with abdominal sepsis face an increased risk of pneumonia due to systemic inflammation, prolonged bed rest, impaired cough reflex, and respiratory muscle dysfunction. Pneumonia increases mortality, delays recovery, and prolongs hospital stays, underscoring the value of pulmonary care. This study evaluates the association between a nurse-led respiratory care bundle and pneumonia incidence and early prognosis in patients with abdominal sepsis. A comparative study of patients with abdominal sepsis in two stages was conducted from January 1, 2021, to June 30, 2023 (control group, which received traditional respiratory care at a single institution), and August 1, 2023, to January 31, 2026 (intervention group, which received a nurse-led respiratory care bundle across five medical centers). Multivariable logistic regression was employed to adjust for confounding factors, and sensitivity analyses were conducted using propensity score matching, inverse probability of treatment weighting, generalized estimating equations, mixed-effects models, and Bootstrap resampling. A total of 869 patients (control, 263; intervention, 606) were included. The nurse-led respiratory care bundle was associated with a lower incidence of pneumonia (42.2% versus 25.6%; P < 0.001) and remained an independent protective factor after multivariable adjustment (odds ratio [OR] = 0.39; 95% confidence interval [CI], 0.28–0.54; P < 0.001). The intervention group had a shorter hospital stay (incidence rate ratio = 0.80; 95% CI, 0.71–0.90; P < 0.001) and lower 28-day mortality (22.8% versus 18.3%; OR = 0.66; 95% CI, 0.45–0.95; P = 0.025). All five sensitivity analyses were consistent, with no heterogeneity among centers. The nurse-led respiratory care bundle was associated with a lower incidence of pneumonia, shorter hospital stays, and lower 28-day mortality in patients with abdominal sepsis.