Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2026 Jan;38(1):117-123. doi: 10.3760/cma.j.cn121430-20250623-00345.
ABSTRACT
OBJECTIVE: To investigate the relationship between the pathogen load kinetics detected by droplet digital polymerase chain reaction (ddPCR) and the prognosis of patients with suspected bloodstream infection (BSI).
METHODS: A prospective observational study was conducted. Patients aged >18 years with suspected BSI admitted to intensive care unit department of Zhejiang Provincial People’s Hospital from March 1, 2022 to October 31, 2023 were consecutively enrolled. Patients with blood ddPCR detection positive for pathogenic bacteria such as Escherichia coli, Klebsiella pneumoniae, Acinetobacter baumannii, or Pseudomonas aeruginosa were included. Demographics, clinical profile, vital signs, and comorbidities data of patients upon admission were collected. Blood ddPCR testing, laboratory tests, and disease severity scoring were performed on days 0, 3, and 7. Based on whether the follow-up ddPCR test at day 7 showed a ≥50% decrease in all pathogen load compared to admission, patients were divided into the pathogen load decrease group and pathogen load non-decrease group. Baseline characteristics and the dynamic changes in perfusion indicators, inflammation indicators, and disease severity scores between two groups were compared. The correlation between ddPCR detection of pathogen load and perfusion indicators, inflammation indicators, and disease severity scores were analyzed using Spearman test. The 28-day cumulative survival rate of pathogen load decrease group and pathogen load non-decrease group was analyzed using Kaplan-Meier survival curve. Patients were divided into the survival group and death group according to the 28-day prognosis, and the clinical data were compared between the two groups. Independent risk factors for 28-day mortality in patients with suspected BSI were identified using multivariate Cox proportional hazards regression models.
RESULTS: A total of 189 suspected BSI patients with ddPCR positive for Escherichia coli, Klebsiella pneumoniae, Acinetobacter baumannii, or Pseudomonas aeruginosa infection were enrolled, of whom 121 underwent dynamic monitoring. Among these 121 patients, 82 showed a decrease in ddPCR pathogen load at day 7, while 39 did not. During the 28-day follow-up, 76 survived and 45 died, and the 28-day mortality was 37.2%. Compared to the pathogen load decrease group, the pathogen load non-decrease group had a higher proportion of patients receiving vasoactive drug support and mechanical ventilation, and a higher 28-day mortality (all P<0.05). In patients with suspected BSI, the ddPCR pathogen load on day 0 showed significant positive correlations with high-sensitivity C-reactive protein (hs-CRP), procalcitonin (PCT), blood lactate (Lac), Acute Physiology and Chronic Health Evaluation II (APACHE II), and Sequential Organ Failure Assessment (SOFA; r values were 0.150, 0.273, 0.370, 0.334, 0.311, respectively; all P<0.05). The Lac, hs-CRP, PCT and APACHE II in pathogen load decrease group decreased with time. However, hs-CRP and PCT did not decrease in pathogen load non-decrease group, and APACHE II and SOFA scores further increased. At day 7, compared to the pathogen load non-decrease group, the pathogen load decrease group showed significant reductions in hs-CRP, PCT, APACHE II and SOFA scores (all P<0.05); however, there was still no statistically significant difference in Lac between the two groups. Kaplan-Meier survival curve analysis showed that the 28-day cumulative survival rate of the pathogen load decrease group was significantly higher than that of the pathogen load non-decrease group (Log-rank test: χ 2=5.969, P=0.015). Compared to the survival group, the death group was older, had higher SOFA score, lower platelet count (PLT) and higher total pathogen load detected by ddPCR at day 7, and had a higher proportion receiving continuous renal replacement therapy (CRRT) and a higher proportion belonging to the non-decreased pathogen load. Multivariate Cox regression analysis showed that increasing age [hazard ratio (HR)=1.048, 95% confidence interval (95%CI) was 1.020-1.078, P<0.001], higher SOFA score (HR=1.127, 95%CI was 1.027-1.235, P=0.007), and pathogen load non-decrease (HR=2.165, 95%CI was 1.148-4.091, P=0.017) were independent risk factors for 28-day mortality in patients with suspected BSI.
CONCLUSIONS: Dynamic monitoring of pathogen load changes by ddPCR can reflect the prognosis of patients with suspected BSI, suggesting the role of ddPCR detection in therapeutic monitoring for patients with BSI.
PMID:41876236 | DOI:10.3760/cma.j.cn121430-20250623-00345
