An evaluation of different dilution modes on circuit lifespan during continuous veno‐venous hemodiafiltration without anticoagulation

Circuit clotting remains a major problem during continuous kidney replacement therapy (CKRT), particularly in patients with contraindications to anticoagulant use. We hypothesized that the different options of alternative replacement fluid infusion sites might affect circuit lifespan. However, research‐based evidence supporting an optimal replacement fluid infusion strategy is limited. Therefore, we aimed to evaluate the effect of three dilution modes (pre‐dilution, post‐dilution, and pre‐ to post‐dilution) on circuit lifespan during continuous veno‐venous hemodiafiltration (CVVHDF).


Conclusion:
The pre-to post-dilution mode significantly prolonged circuit lifespan but did not reduce Scr and BUN levels, compared with the pre-dilution and postdilution modes during CVVHDF when no anticoagulants were used.

| INTRODUCTION
Continuous kidney replacement therapy (CKRT) is one of the most commonly used methods for maintaining electrolyte and acid-base homeostasis in critically ill patients, regardless of the need for kidney impairment treatment. 1During CKRT, effective anticoagulation optimizes filter and whole circuit performance, but this requirement can exaggerate the risk of bleeding in patients with active hemorrhage.CKRT without anticoagulation is recommended by the KDIGO guidelines for such patients. 2However, heparin-free hemodialysis can be challenging owing to the diminished survival of the extracorporeal circuit and the CKRT filter due to clotting, which may lead to significant blood loss in patients.
6][7] It is reasonable to believe that the choice of site for replacement fluid administration may affect the circuit lifespan despite the CKRT technique. 8There are only limited studies comparing pre-versus post-dilution strategies in CVVH, and pre-dilution strategies have been preferred for prolonging filter life, 9,10 but the underlying mechanisms have not been elucidated.A systematic review has reported that the evidence base for justifying decisions regarding the optimal dilution mode of CKRT is weak, with heterogeneous conflicting results and no overall trend. 8 our knowledge, the medical literature is lacking in terms of providing evidence for choosing a dilution-mode strategy for CVVHDF.To overcome this knowledge gap, we conducted a prospective cohort study to evaluate the effect of different dilution modes of CVVHDF without anticoagulation on the circuit lifespan.

| Study design and Ethics
A prospective cohort study was conducted on patients who received different dilution strategies during CKRT, and the circuit lifespans were compared.This study was approved by the Research Ethics Committee and was registered at the Chinese Clinical Trial Registry.
Informed consent was obtained from all the patients or their guardians before the commencement of CKRT and any sampling procedure.

| Patient recruitment
Prior to allocating patients to the study and before any study-related procedures were undertaken, informed consent was obtained from all patients.Patients who met the following criteria were enrolled between December 2019 and December 2020.The inclusion criteria were as follows: (1) patients who underwent continuous veno-venous hemodiafiltration (CVVHDF), (2) patients aged between 18 and 70 years, and (3) CVVHDF without anticoagulation.The exclusion criteria were as follows: (1) indication for systemic anticoagulation (unfractionated heparin, low molecular weight heparin, etc.) without an increased bleeding risk or impaired coagulation, (2) indication for regional citrate anticoagulation without citric acid accumulation or severe hypotension or severe liver failure, (3) pregnancy, and (4) planned surgery within 72 h.

| Data collection
Clinical and demographic characteristics such as age, gender, disease diagnosis, and filter membranes used for all the enrolled patients were collected and are summarized in Figure 1.In particular, we also collected laboratory test results, including activated partial thromboplastin time (aPTT), prothrombin time (PT), platelet count (PT), hematocrit (HCT), serum creatinine (Scr), and blood urea nitrogen (BUN) before and after CKRT.The primary outcome was circuit lifespan, which was defined as circuit survival until a nonelective circuit disconnection due to filter clotting, extracorporeal coagulation, or an effective transmembrane pressure (>300 mmHg).The second outcome was evaluated based on the clinical parameters of the patients, such as changes in Scr and BUN levels, 28-day all-cause mortality, and the length of stay.All patients in the study only accepted one treatment group, and we only collected data on the first filter lifespan.1.No significant differences were detected between the different treatment groups in terms of the baseline characteristics (p > 0.05).

| Primary outcomes of circuit lifespan
The mean circuit lifespan was significantly longer in the pre-to postdilution group (45.72 h, 95% CI, 39.75-51.69h) than in the predilution group (31.58 h, 95% CI, 26.33-36.82h) and the post-dilution group (35.20 h, 95% CI, 29.62-40.78h) (Table 2), but there was no significant difference between the pre-dilution group and the postdilution group (p > 0.05).Kaplan-Meier survival analysis revealed a significant difference between the three dilution modes (p = 0.001; Figure 2).Cox regression analysis also showed that the gender, filter membrane, and the other concomitant variables did not affect the above result (Figure 3).
There was a significant difference in the circuit clotting site among the three groups (p < 0.001).The bubble trap chamber clotting site was the most frequent location in the pre-dilution, while the filter clotting site was the most frequent location in the post-dilution and pre-to post-dilution modes.

| Secondary outcomes of clinical parameters
There was no significant difference in the change in Scr and BUN levels, length of stay, or 28-day all-cause mortality among the three dilution groups (p > 0.05; Table 2).CKRT is an extracorporeal blood purification therapy that aims to support kidney function over an extended period of time. 11Optimizing the circuit lifespan and performance efficiency of CKRT has been a hotly debated topic because a shorter circuit lifespan reduces CKRT efficacy and increases hospital costs.Although some high-quality and T A B L E 1 Baseline characteristics of the patients enrolled in the study (n = 132).3][14] Studies comparing CVVHDF with pre-to post-dilution mode to those with pre-dilution or post-dilution mode in patients with a high risk of bleeding are still limited.During CKRT, the replacement fluid can be delivered either before the filter (pre-dilution) or after the filter (postdilution). 15In recent years, combining pre-dilution and post-dilution replacement fluid strategies during CKRT appears to have been an increasingly common practice in many ICUs in China, 16,17 a practice also employed in other countries. 3,18,19Pre-dilution combined with post-dilution of replacement fluids is considered an important tool for preventing clotting during CKRT. 20r data show that the pre-dilution and post-dilution replacement fluid strategies were comparable in their impact on maintaining circuit life during CKRT, and this is in concordance with a previous study. 21However, another report indicated a reduced circuit life in post-dilution mode due to a relatively concentrated blood load compared to the pre-dilution mode during CVVH, 9 particularly when the filtration fraction was over 30%. 16The report also showed that there was a significant difference in the circuit clotting site among the  F I G U R E 2 The Kaplan-Meier curve for circuit lifespan.
the membrane.[7] The present study showed that there was no difference in terms of changes in Scr and BUN levels between the three dilution groups.
It was difficult to make a case for the basal Scr and BUN levels to not be high in the majority of patients who had excessive volume load.
However, patients who underwent pre-dilution CKRT showed reduced capability for urea clearance, leading to an increased need for ultrafiltration volume or a faster blood flow rate as compensation. 22e pre-to post-dilution mode combines the advantages of both predilution and post-dilution of replacement fluids to achieve the best circuit life during CVVHDF.Pre-to post-dilution strategy can improve the capacity for urea and BUN clearance, avoid volume overload, and extend the circuit lifespan.In clinical practice, blood flow and fluid replacement rates should be adjusted to achieve targeted plasma clearance and maintain an optimal ultrafiltration fraction volume. 22is study has several limitations.First, this was a prospective cohort study; this could, to some extent, undermine the quality of the evidence.Nevertheless, this prospective study was designed based on the best interests of patients.Second, other extraneous variables, such as the ability of the nursing staff to operate CKRT machines and to handle alarms that may affect the circuit lifespan, were not considered in this study.A randomized cross-over study should be developed to explore the same patients who randomly accepted different dilution models in the future.

| CONCLUSIONS
CKRT is intended to function continuously, and circuit clotting remains a major problem, particularly in patients with contraindications to anticoagulant use.
CVVHDF was delivered to all patients by nurses experienced in CKRT using a Prismaflex machine with AN69ST ST150 dialyzers or with the AN69 M150 dialyzers (Baxter International Inc.).The vast majority of the vascular access in all patients was the femoral vein.Double-lumen catheters (13Fr, 250 mm; GDHK-1325, Baxter International Inc., Deerfield, IL, USA) were placed in the femoral veins.Other modes of vascular access included entry through the internal jugular vein or subclavian vein.According to the theoretically prescribed dose of 25-35 mL/kg/h, our team usually set a rate of 2 L/h (dialysis:replacement fluid = 1:1) for critically ill patients of normal weight and a blood flow rate of 200 mL/min without anticoagulation.Patients were allocated to one of three treatment groups determined by the CKRT specialist according to their own preferences and the patient's condition: predilution CVVHDF, post-dilution CVVHDF, and pre-to post-dilution CVVHDF based on the replacement fluid infusion sites.The replacement fluid was administered before the filter in the pre-dilution group and after the filter in the post-dilution group.In the pre-to postdilution group, the replacement fluid was administered in a mixed manner (half before the filter and the other half after the filter).The filters were changed every 72 h, according to the manufacturer's instructions.The flow of participants throughout the study is summarized in Figure 1.
SPSS for Windows (Statistical Package for the Social Sciences, IBM SPSS Statistics 19, Chicago, IL, USA) was used for all the statistical analyses.Group comparisons were performed using χ 2 tests for categorical characteristics and either one-way ANOVA or Mann-Whitney U tests for continuous variables.Circuit lifespan was analyzed and compared using the Kaplan-Meier test and Cox regression analysis.Statistical significance was set at a two-tailed p-value of <0.05.
Patient characteristics The study was conducted from December 2019 to December 2020 in the intensive care units of a large hospital in Chengdu, China.A total of 146 patients were initially enrolled to assess eligibility, among whom five patients were excluded because of discharged during CKRT, 132 patients were eventually enrolled and allocated to receive pre-dilution (n = 40), post-dilution (n = 42), or pre-to post-dilution (n = 50; Figure 1) modes of replacement fluid addition during CVVHDF.The baseline clinical and demographic characteristics of the patients are reported in Table

F
I G U R E 1 Schematic showing participant flow in the study.In this prospective cohort study, we present evidence for the extracorporeal circuit lifespan to be a critical indicator of filter or circuit clotting under different dilution strategies during CVVHDF.Our data show that the pre-to post-dilution mode significantly prolonged the circuit lifespan as compared to the pre-dilution and post-dilution replacement fluid modes during CVVHDF.However, there was no statistically significant difference in terms of changes in Scr and BUN levels, admission day, and 28-day all-cause mortality in the three patient groups based on replacement fluid strategies.The pre-dilution and post-dilution modes did not change the circuit lifespan during CVVHDF.
three groups (p < 0.001): bubble trap chamber clotting was the most common location in the pre-dilution, while filter clotting was the most frequent site in the post-dilution and pre-to post-dilution modes.A Prismaflex machine with AN69 dialyzers during CVVHDF was used, and the size of its bubble trap chamber was smaller.Post-dilution fluids accumulating in the venous bubble chamber or the air trap chamber can create a fluid layer on top of the blood level and possibly protect the site from excessive clotting, thus reducing clot development.Therefore, post-dilution and pre-to post-dilution replacement fluid modes were particularly effective in reducing the risk of bubble trap chamber clotting.Replacement fluids administered before the membrane help to dilute the blood and reduce clot development in T A B L E 2 Clinical outcomes in the study.
The pre-to post-dilution mode significantly extends the circuit lifespan during CVVHDF for critically ill patients undergoing CKRT without anticoagulant use.Our data support the use of pre-dilution combined with post-dilution of replacement fluids to prevent circuit clotting during CKRT in the absence of anticoagulant use.