Analysis of Leukocyte Recruitment in Continuous Veno-Venous Hemofiltration with Regional Citrate vs. Systemic Heparin Anticoagulation

Abstract

Acute kidney injury (AKI) is a frequent complication in critically ill patients. Supportive treatment of AKI patients is based on renal-replacement therapy, including continuous veno-venous hemofiltration (CVVH). To limit clotting events on extracorporeal surfaces, anticoagulants are administered, including systemic heparin and local citrate. The differential and comparative effects of these anticoagulants on leukocyte function in acute kidney injury patients are, so far, insufficiently understood. In this bio-add-on-study, AKI patients were randomized as part of a parallel-group trial to either systemic heparin or regional citrate anticoagulation. Patient samples were collected upon inclusion, prior to CVVH initiation at day 0, day 1, day 3 and day 5, following CVVH initiation, and one day after cessation of CVVH, then immediately analyzed. Flow cytometric assessment of surface-receptor molecules was conducted. Whole-blood-perfused human microfluidic chambers were used for the analysis of neutrophil rolling and adhesion. Acute kidney injury was associated with significant changes in the surface expression of CD182 and CD16 throughout CVVH treatment, independent of the anticoagulation regime. AKI furthermore abrogated selectin-induced slow leukocyte rolling and diminished chemokine-induced leukocyte arrest. Subgroup analyses of citrate vs. heparin treatment showed no significant differences between groups, independent of the duration of CVVH treatment. CD182 and CD16 expression remained low in both groups throughout CVVH therapy. These data confirm that AKI impairs selectin-mediated leukocyte slow rolling and chemokine-induced leukocyte arrest in vitro. Systemic heparin or local citrate anticoagulation have no differential effect on the leukocyte recruitment steps examined in this study.

Keywords: CVVH; acute kidney injury; anticoagulation; citrate; heparin; inflammation; leukocytes.

Figure 1

Neutrophil surface expression. Levels of important neutrophil surface molecules (CD11a (A), CD11b (B), CD14 (C), CD16 (D), CD29 (E), CD49d (F), CD62L (G), CD162 (H), CD182 (I)) were measured by flow cytometry in density-gradient-isolated neutrophils from critically ill patients experiencing acute kidney injury before (d0), during (d1, d3, d5) and after (d1 post) either regional-citrate- or systemic-heparin-based continuous veno-venous hemofiltration and from healthy volunteers (control). Mean fluorescence intensity (MFI) measurements of CD11b and CD14 differed between citrate and heparin groups. CD16 and CD182 surface levels were all significantly reduced in comparison to healthy controls. (n ≥ 3 individual patients; * p < 0.05; ** p < 0.01; mean ± SEM).

Figure 2

Leukocyte phenotype profiles. (A) Representative gating strategy of flow cytometry data is shown for determination of neutrophils. (B) Heat-map representation of expression levels indicative of leukocyte phenotype profiles relative to healthy control expression levels. (C) Spider-graph representation of expression levels relative to group-expression levels before CVVH initiation (d0).

Figure 3

Acute kidney injury impairs leukocyte slow rolling. (A) Slow leukocyte rolling was assessed using whole blood perfused into microfluidic chamber devices. (B) Bar-graph representation with error bars of (A): T-test was performed for analysis of integrin response for each condition (shown as black bar and asterisk). Inter-group variations were tested using ANOVA (light gray bar and asterisk) (n ≥ 3 individual patients; * p < 0.05; mean ± SEM).

Figure 4

Leukocyte slow rolling in regional-citrate- and systemic-heparin-based CVVH. Sub-group analyses for systemic heparin vs. regional citrate slow leukocyte rolling were performed on day 0 (A), day 1 (B), day 3 (C), day 5 (D) and day 1 after CVVH discontinuation (E) and revealed no significant difference. T-test was performed for analysis of integrin response for each condition (shown as black bar and asterisk), where only the control subjects exhibited a significant ICAM-1-dependent reduction in rolling velocity, whereas this slow rolling response was not present in AKI patients. Light gray significance level in graphs represents results of in-group (either heparin vs. citrate vs. control in E-selectin or heparin vs. citrate vs. control in E + ICAM-1) analyses, tested by ANOVA. (n ≥ 3 individual patients; * p < 0.05; mean ± SEM).

Figure 5

Chemokine-induced leukocyte adhesion. For analysis of chemokine-induced adhesion, whole blood was perfused into microfluidic chambers coated with either P-selectin and ICAM-1 or P-selectin, ICAM-1 and interleukin-8, and the ratio of adhering to rolling leukocytes determined. Control subjects exhibited a significant increase in ratio of adherent to rolling cells, whereas this increase was absent in all AKI-groups, independent of mode or duration of CVVH. (n ≥ 2 individual patients; * p < 0.05; mean ± SEM).