Clinical Prediction of Secondary Bloodstream Infections in Patients with Cerebral Infarction: A Nomogram-Driven Risk Assessment Model Based on LASSO Regression

Abstract

Lei Zhang,1,2 Xiaojun Li,2,3 Donghao Cai,2,4 Chuangchuang Mei,2,4 Lu Lu2,5 1Department of Quality Control, Guangdong Provincial Second Hospital of Traditional Chinese Medicine, Guangzhou, Guangdong, 510095, People’s Republic of China; 2Guangdong Provincial Key Laboratory of Research and Development in Traditional Chinese Medicine Guangzhou, Guangdong, 510095, People’s Republic of China; 3Department of Nosocomial Infection, Guangdong Provincial Second Hospital of Traditional Chinese Medicine, Guangzhou, Guangdong, 510095, People’s Republic of China; 4Department of Laboratory Medicine, Guangdong Provincial Second Hospital of Traditional Chinese Medicine, Guangzhou, Guangdong, 510095, People’s Republic of China; 5Department of Lujingdong Clinic, Guangdong Provincial Second Hospital of Traditional Chinese Medicine, Guangzhou, Guangdong, 510095, People’s Republic of ChinaCorrespondence: Lu Lu, Department of Lujingdong Clinic, Guangdong Provincial Second Hospital of Traditional Chinese Medicine, Guangzhou, Guangdong, 510095, People’s Republic of China, Email [email protected]: To evaluate the impact of secondary bloodstream infections (BSI) on healthcare quality indicators in patients with cerebral infarction, and to develop a validated predictive model.Methods: This study conducted a retrospective analysis of 7,698 distinct patients with cerebral infarction (2023) from a tertiary hospital in Guangzhou. Patients were categorized into two groups: BSI-negative (n=7,573) and BSI-positive (n=125). Healthcare quality indicators were compared using Mann–Whitney U-test. A predictive model was created using Least Absolute Shrinkage and Selection Operator (LASSO) regression, based on a 7:3 training-validation split. The model’s performance was validated through the area under the Receiver Operating Characteristic curve (AUC), calibration curves, and decision curve analysis (DCA).Results: Patients with BSI had significantly prolonged hospital stays (median of 17 days versus 11 days, p< 0.001), higher costs (median of 34,859 yuan compared to 16,921 yuan, p< 0.001), and increased adverse outcomes (34.4% versus 1.6%, p< 0.001). The LASSO analysis identified four predictors: The following variables were found to have a statistically significant relationship to the occurrence of the primary complication: peripherally inserted central venous catheters (PICC) (odds ratio [OR] = 2.791, 95% confidence interval [CI] =1.514– 5.148), use of ventilators(VA) (OR = 2.771, 95% CI=1.410– 5.443), Indwelling urinary catheters(CAU) (OR = 1.800, 95% CI= 0.990– 3.276), and hypoalbuminemia (OR = 3.643, 95% CI=2.195– 6.046).The nomogram demonstrated an AUC of 0.789 in the training set and 0.778 in the test set, indicating a satisfactory model fit across data sets. Good model fit based on Hosmer-Lemeshowp-values(Hosmer-LemeshowP=0.338/0.170).DCA indicated a net clinical benefit at risk thresholds of 0– 15%.Conclusion: Secondary BSI in patients with cerebral infarction can seriously affect the quality of medical care.The developed nomogram functions as a pragmatic instrument for the preliminary identification of patients at high risk. It facilitates the implementation of targeted interventions, thereby reducing the incidence of BSI and enhancing patient outcomes.Keywords: cerebral infarction, bloodstream infection, nomogram, healthcare quality

Keywords

• cerebral infarction
• bloodstream infection;Nomogram
• healthcare quality