In recent decades, efforts in dialysis have focused on improving the clearance of larger middle molecular weight uraemic toxins, the retention of which has been associated with pathological features of uraemia [1]. Convective flow through haemodialysis membranes using online haemodiafiltration (OL-HDF) techniques has been introduced in recent years to enhance the removal of middle and large molecular weight uraemic toxins [2], resulting in better outcomes in terms of cardiovascular mortality in haemodialysis patients [3]. ‘Super high-flux’ (HF) dialysers are now commercially available, developed for the purpose of removing large amount of larger middle solutes. One of these high performance dialysers, XevontaHI23®Braun, with very high water permeability [in vitro ultrafiltration coefficient (Kuf) of 124 mL/h/mmHg] has recently been introduced, but there is still a lack of evidence on its use. We designed a transverse study to evaluate the efficacy of this very high permeability (VHP) dialyser and to compare it with another HF dialyser in OL-HDF.A total of 14 prevalent OL-HDF patients were included. Dialysers were compared in two consecutive mid-week dialysis sessions. Treatments were based on current prescription with no restriction on blood flow. OL-HDF was performed in post-dilution mode with automatic pressure control of convection and no restriction on total convective ultrafiltration volume. The efficacy of each dialyser was analysed by measuring the reduction ratios (RRs) of substances with different molecular weights. We registered total convective volume as well as hourly transmembrane pressure (TMP) with each dialyser. For detailed methodology, see Supplementary Material.Mean total convective volume per session was significantly higher with the VHP dialyser (33.5 ± 5.4 versus 30.9 ± 4.6 L/session; P = 0.013). There were no differences in in vivo Kuf, TMP or in the RR of the different molecules between the two dialysers (Table 1).
Table 1.
Comparison of dialysis parameters and RRs between the different dialysers
Dialysis parameters and RRs
VHP dialyser
HF dialyser
P-value
Mean convective volume per session (L/session)
33.5 ± 5.4
30.9 ± 4.6
0.01
Minimum TMP (mmHg)
155.9 ± 53.7
157.8 ± 34.6
0.68
Maximum TMP (mmHg)
244.85 ± 41.1
230.7 ± 38.1
0.09
Medium TMP (mmHg)
213.6 ± 47.2
205.5 ± 32
0.28
Mean Kt/V per session
1.8 ± 0.4
1.9 ± 0.6
0.31
Mean ionic dialisance per session (mL/min)
287 ± 25
284 ± 40
0.27
In vivo Kuf (mL/h/mmHg)
40.9 ± 10.8
38.2 ± 6.1
0.3
Reduction ratios (%)
Urea
85.4 ± 5
84.4 ± 44
0.17
Creatinine
77.7 ± 6
77.5 ± 4
0.34
Phosphate
65.9 ± 11
62.9 ± 8
0.22
Myoglobin
72 ± 8
73.9 ± 6
0.33
Cystatin C
78.1 ± 6
79.1 ± 4
0.25
β2-microglobulin
76.2 ± 8
81.4 ± 2
0.11
Prolactin
71.1 ± 9
71 ± 8
0.73
Values are represented as mean ± SD.
Comparison of dialysis parameters and RRs between the different dialysersValues are represented as mean ± SD.Despite higher convective volumes achieved with the VHP dialyser, we found no differences in minimum, maximum or mean TMP between the two dialysers. This could be explained by the method of TMP calculation, where only three pressure points are known and the fourth must be assumed, thus introducing great variability [4]. Secondly, it may be explained by differences between in vitro and in vivo Kuf. It is well described in the literature that in vivo Kuf is inferior to in vitro Kuf, and it progressively decreases during the dialysis session, mainly due to blood protein boundary effects and increased resistance to ultrafiltration [5]. In fact, Braun® sponsored a clinical trial in 2010 that was designed to evaluate the performance and safety profile of Xevonta high-flux dialyser with special focus on determination of in vivo Kuf (NCT01111266) but, unfortunately, no results have been posted for this study. Estimated in vivo Kuf for the different dialysers was very similar in our study, although it should be noted that estimated in vivo Kuf also includes estimated TMP, with its previously described limitations.Differences observed in total convective volume between the two dialysers did not result in significantly higher RR of different-sized uraemic toxins either. This could be explained not only by the differences between in vitro and in vivo Kuf, but also because hydraulic permeability is of utmost importance in HF convective techniques; however, it is not the only factor involved.In light of the results of our study, we can say that the VHP dialyser achieves higher convective volumes than the HF dialyser with similar removal in middle molecules. We believe that there is an in vivo infra optimization of the VHP dialyser that may explain the similar results obtained in terms of RR of middle molecules despite the higher convective volumes achieved. Future studies should assess whether the optimization on the use of this type of dialysers could improve outcomes.
SUPPLEMENTARY DATA
Supplementary data are available at ckj online.
CONFLICT OF INTEREST STATEMENT
None declared.Click here for additional data file.