| Abstract Objective: To analyze the distribution and drug resistance of pathogenic bacteria in catheter-related bloodstream infection (CRBSI) among diabetic nephropathy (DN) patients undergoing blood purification, and to construct a Nomogram prediction model for the risk of CRBSI. Methods: A total of 185 DN patients receiving blood purification for the first time were enrolled. The incidence of CRBSI was calculated, and the distribution and drug resistance of CRBSI pathogenic bacteria were analyzed. Patients were divided into CRBSI group and non-CRBSI group according to the occurrence of CRBSI. Clinical data were compared between the two groups. Logistic regression analysis was used to screen the influencing factors of CRBSI, and a Nomogram prediction model for CRBSI risk was established with its predictive efficacy verified. Results: Among the 185 DN patients, the incidence rate of CRBSI was 24.32% (45/185). A total of 62 strains of pathogenic bacteria were isolated from blood samples of 45 patients in the CRBSI group, among which Gram-positive bacteria accounted for 56.45% (35/62) and Gram-negative bacteria accounted for 43.55% (27/62). Drug resistance analysis showed that the main Gram-positive bacteria including Staphylococcus aureus and Staphylococcus epidermidis presented varying degrees of drug resistance to erythromycin, clindamycin, levofloxacin and ciprofloxacin; the main Gram-negative bacteria including Klebsiella pneumoniae and Pseudomonas aeruginosa showed different levels of resistance to ampicillin, compound sulfamethoxazole, minocycline and meropenem. The CRBSI group had older age, higher proportion of catheter indwelling time (>7 days), higher proportion of intubation times (≥3 times), higher proportion of prophylactic use of antibiotics, as well as higher levels of serum C-reactive protein (CRP), procalcitonin (PCT), transforming growth factor-β1 (TGF-β1), Smad2 and Smad3 proteins, while the score of Adult Health Self-Management Behavior Rating Scale (AHSMSRS) and serum 25-hydroxyvitamin D3 [25(OH)D3] level were lower than those in the non-CRBSI group (all P< 0.05). Logistic regression analysis indicated that age, catheter indwelling time, intubation times, prophylactic use of antibiotics, AHSMSRS score, serum CRP, PCT, 25(OH)D3, TGF-β1, Smad2 and Smad3 were independent influencing factors for CRBSI (all P< 0.05). Based on the results of multivariate analysis, a Nomogram prediction model for CRBSI risk was constructed. The verification results showed that the area under the curve of the model for predicting CRBSI was 0.926, with obvious positive net benefit. Conclusion: The Nomogram prediction model for CRBSI risk in DN patients undergoing blood purification constructed based on pathogenic bacteria distribution, drug resistance and clinical indicators such as age and catheter indwelling time has good predictive performance and clinical applicability. |
