[Predictive value of neutrophil-to-lymphocyte ratio combined with glycosylated haemoglobin for bloodstream infection in pyogenic liver abscesses].
Source: PubMed, NCBI / U.S. National Library of Medicine
To investigate the predictive value of neutrophil-to-lymphocyte ratio (NLR) combined with glycosylated hemoglobin A1c (HbA1c) for bloodstream infection in patients with pyogenic liver abscess (PLA). A retrospective cohort study was conducted. Patients with PLA admitted to Linyi Central Hospital from January 2020 to June 2024 were enrolled. Clinical data, comorbidities, clinical symptoms, signs, admission laboratory parameters, imaging data, and blood culture results were collected. Patients were divided into a bloodstream infection group (blood culture-positive) and a non-bloodstream infection group (blood culture-negative) according to their blood culture results, and differences in various indicators were compared between two groups. Multivariate Logistic regression analysis was used to identify independent risk factors for bloodstream infection. The receiver operator characteristic curve (ROC curve) was employed to evaluate the predictive value of NLR combined with HbA1c for bloodstream infection in patients with PLA. A total of 124 patients with PLA were enrolled, among whom 48 had positive blood cultures, yielding a positivity rate of 45.2%. The proportion of patients with diabetes, as well as levels of procalcitonin (PCT), HbA1c, and NLR, were higher in the bloodstream infection group than in the non-bloodstream infection group [diabetes proportion: 41.1% vs. 11.8%, PCT (ng/L): 18.98±12.16 vs. 14.19±11.85, HbA1c: 0.939±0.056 vs. 0.053±0.027, NLR: 28.
Abstract
To investigate the predictive value of neutrophil-to-lymphocyte ratio (NLR) combined with glycosylated hemoglobin A1c (HbA1c) for bloodstream infection in patients with pyogenic liver abscess (PLA). A retrospective cohort study was conducted. Patients with PLA admitted to Linyi Central Hospital from January 2020 to June 2024 were enrolled. Clinical data, comorbidities, clinical symptoms, signs, admission laboratory parameters, imaging data, and blood culture results were collected. Patients were divided into a bloodstream infection group (blood culture-positive) and a non-bloodstream infection group (blood culture-negative) according to their blood culture results, and differences in various indicators were compared between two groups. Multivariate Logistic regression analysis was used to identify independent risk factors for bloodstream infection. The receiver operator characteristic curve (ROC curve) was employed to evaluate the predictive value of NLR combined with HbA1c for bloodstream infection in patients with PLA. A total of 124 patients with PLA were enrolled, among whom 48 had positive blood cultures, yielding a positivity rate of 45.2%. The proportion of patients with diabetes, as well as levels of procalcitonin (PCT), HbA1c, and NLR, were higher in the bloodstream infection group than in the non-bloodstream infection group [diabetes proportion: 41.1% vs. 11.8%, PCT (ng/L): 18.98±12.16 vs. 14.19±11.85, HbA1c: 0.939±0.056 vs. 0.053±0.027, NLR: 28.08±13.64 vs. 21.69±11.37, all P<0.05]. Multivariate Logistic regression analysis revealed that elevated HbA1c at admission [odds ratio (OR)=1.289, 95% confidence interval (95%CI) was 1.097-1.513, P=0.002] and elevated NLR (OR=1.041, 95%CI was 1.007-1.075, P=0.016) were independent risk factors for bloodstream infection in patients with PLA. ROC curve analysis demonstrated that the combination of NLR and HbA1c predicted bloodstream infection with an area under the curve (AUC) of 0.824 (95%CI was 0.753-0.896, P<0.001); when the optimal cut-off value of 0.298, the sensitivity was 0.929 and the specificity was 0.603. The combination of NLR and admission HbA1c has predictive value for bloodstream infection in patients with PLA.