RESUMO
Background: Significant errors of activated partial thromboplastin time (aPTT) ratio were frequently observed in blood sampling from central venous dialysis catheter (CVC) of hemodialysis (HD) patients. Following the draw-and-return methods, initial blood withdrawal from the catheter before sampling can reduce the error, but the optimal withdrawal volume remains undetermined. Aim: The objective of this study is to determine the optimal blood withdrawal volume for the draw-and-return methods to improve aPTT ratio accuracy in hemodialysis patients with CVC. Methods: A prospective study was conducted in patients receiving HD via CVC. Four blood samples were collected from each patient, involving a peripheral venipuncture and three draw-and-return samples (10 ml, 20 ml and 25 ml groups). The aPTT ratio of a peripheral sample was used as a reference to determine the aPTT ratio accuracy for each draw-and-return group. Subsequently, the agreement was illustrated using modified Bland-Altman plot. Results: A total of 1,000 samples were obtained from 250 patients. The patients had a mean age of 59.6 ± 15.4 years, with 17.2% using citrate as the CVC's locking agent. The adjusted accuracies of the aPTT ratio varied significantly among the three withdrawal volumes (p-value <0.001). The 25 ml group demonstrated the highest accuracy (43.2%; 95%CI, 38.0-48.4), followed by the 20 ml group (30.0%; 95%CI, 24.9-35.2), and the 10 ml group (18.0%; 95%CI, 12.8-23.2). Additionally, using citrate as a locking agent provided more than 80.0% aPTT ratio accuracy, whereas heparin demonstrated inferior accuracy even in the 25 ml withdrawal group. Conclusion: The optimal blood withdraw volume for the draw-and-return methods concluded at 20 ml for citrate locked-CVC and 25 ml for heparin which significantly improved aPTT ratio accuracies. Applying citrate as a locking agent provides clear benefits for aPTT ratio monitoring, while peripheral venipuncture is recommended in cases of heparin-locked CVC.