Effects of unfractioned heparin and low-molecular-weight heparin on osteoprotegerin and RANKL plasma levels in haemodialysis patients

Abstract

This randomized crossover study investigated the effects of unfractioned heparin (UFH) and low-molecular-weight heparin (LMWH) on intra- and post-dialytic blood levels of osteoprotegerin (OPG), receptor activator of nuclear factor kappa B ligand (RANKL) and inflammatory cytokines. Forty patients on haemodialysis for at least 12 months were selected. UFH or LMWH was randomly assigned and maintained for 1 month, and then, in the following month, each patient was switched to the other form of heparin. In the mid-week session, we determined the changes in anti-Xa activity, OPG, RANKL, IL-1β, IL-6 and TNF-α values before heparin administration and after 15 min, 4, 8 and 24 h (T0, T1, T2, T3 and T4 respectively). Since these parameters at the various experimental times showed a non-normal distribution, log transformation was applied in order to run parametric ANOVA, with Bonferroni correction for multiple comparisons. The changes in anti-Xa activity over time were similar but not the same for the UFH and LMWH. A highly significant (P<0.001) increase in anti-Xa activity was detected at T1, regardless of the type of heparin, as confirmed in the comparison of T0 vs T1 using one-way ANOVA. Moreover, with both heparins, significant differences were found in the comparisons of anti-Xa activity at T1 vs T2 (both P<0.001) and at T2 vs T3 (P=0.0003 with UFH; P<0.001 with LMWH). Conversely, the difference in anti-Xa activity at T3 vs T4 was still significant with UFH (P=0.0186) but not significant with LMWH (P=0.728). When comparing anti-Xa activity at T4 vs T0, no significant differences were found either with UFH (P=0.1996) or with LMWH (P=0.7470), thus indicating that 24 h after heparin infusion, anti-Xa activity returned back to the pre-infusion values. When we analysed the changes in OPG levels over time, we found that the administration of heparin, regardless of the type, determined an increase in circulating OPG with a zenith at 15 min (T1), with a return back to the baseline levels within the 24th hour post-infusion. One-way ANOVA revealed significant differences in OPG blood levels at T0 vs T1 with both UFH (P=0.0112) and LMWH (P=0.0288), whereas no significant difference was observed in the comparisons of OPG levels at T1 vs T2, T2 vs T3, T3 vs T4 and T4 vs T0, either with UFH or with LMWH. The circulating levels of RANKL, IL-1β, IL-6 and TNF-α at the different intra- and post-dialytic times did not show significant variations following heparin administration, either with UFH or with LMWH. One-way ANOVA performed on the log-transformed values of RANKL, IL-1β, IL-6 and TNF-α at the various experimental times (T0 vs T1, T1 vs T2, T2 vs T2, T3 vs T4 and T4 vs T0) revealed no significant intra- and post-dialytic changes in their blood levels, thus confirming that heparin infusion did not affect their blood levels. These results suggest that heparin-regulated cyclic increases of OPG might play a role in the vascular pathology of haemodialysis patients.