Abstract
OBJECTIVE: To investigate the causative factors, antimicrobial resistance patterns, and associated risk factors of postoperative incisional infections in patients with acute intestinal obstruction and to develop a predictive model. METHODS: A retrospective study was conducted on patients with acute intestinal obstruction (n = 329) admitted to the Emergency Surgery Department of the First Affiliated Hospital of Anhui Medical University between January 1, 2020, and December 31, 2022. Patients were included based on specific criteria. Wound drainage samples from patients with postoperative incisional infections were collected for bacterial culture and drug susceptibility testing. Patients were randomly divided into a training set (n = 231) and a validation set (n = 98) at a 7:3 ratio. Least Absolute Shrinkage and Selection Operator (LASSO) regression was employed to screen variables and select predictors. Multivariate logistic regression was utilized to analyze risk factors and develop a predictive model. The area under the curve (AUC) was calculated to assess the model's discriminatory ability, and calibration and decision curve analyses were performed. RESULTS: Among the 329 patients, 37 (11.25%) developed postoperative incisional infections. Bacterial cultures were positive in 32 of 37 infected patients (86.48%). Gram-negative bacteria, primarily Escherichia coli, accounted for 65.63% of isolates, while gram-positive bacteria, predominantly Enterococcus faecium, comprised 28.12%. Fungi, mainly Candida albicans, constituted 6.25%. Gram-negative bacteria exhibited high resistance to ceftriaxone but low resistance to imipenem. Gram-positive bacteria demonstrated higher resistance to erythromycin than ciprofloxacin, with no vancomycin-resistant strains identified. LASSO regression identified seven variables, which were further analyzed using multivariate logistic regression to identify six independent risk factors for incisional infection. A predictive model was developed based on these six factors: age ≥ 60 years, diabetes history, operative time ≥ 3 h, colorectal obstruction, enterostomy, and hemoglobin (HGB). The AUCs for the training and validation sets were 0.952 (95% CI 0.914-0.990) and 0.982 (95% CI 0.959-1.000), respectively. Hosmer-Lemeshow goodness-of-fit tests and calibration curves demonstrated good model fit. Decision curve analysis indicated a significant clinical net benefit of the predictive model. CONCLUSION: Gram-negative bacteria constitute the primary causative agents of postoperative incisional infections in patients with acute intestinal obstruction. Moreover, these bacteria exhibit significant resistance to commonly used antibiotics. To mitigate the risk of such infections, clinicians should prioritize the monitoring of gram-negative bacterial growth. Prophylactic antibiotic administration can further reduce the incidence of these infections. Additionally, a predictive model incorporating six key variables-age ≥ 60 years, diabetes mellitus, operative time ≥ 3 h, colorectal obstruction, enterostomy, and HGB-can aid in identifying high-risk patients. This model enables clinicians to implement targeted early monitoring and preventive strategies, ultimately improving patient outcomes.