Intradialytic cycling does not exacerbate microparticles or circulating markers of systemic inflammation in haemodialysis patients

People with end-stage kidney disease (ESKD) receiving regular haemodialysis (HD) display elevated cardiovascular mortality, and cardiovascular disease is the leading cause of death in this population (23.3%) (UK Renal Registry 2019). This may be in part due to elevated circulating microparticle (MP) concentrations and altered MP characteristics. MPs are shed upon cellular activation or apoptosis and can act as biomarkers for inflammation and leukocyte dysfunction whilst also promoting thrombosis through tissue factor (TF) expression (Piccin et al., 2007). HD treatment causes increased platelet and neutrophil activation and thus increases MP shedding (Daniel et al. 2006). Furthermore, MPs isolated from HD patients display elevated thrombotic potential compared to the healthy population (Burton et al. 2013) and may be predictive of mortality (Amabile et al. 2011).

HD patients also display chronically elevated circulating concentrations of pro-inflammatory cytokines (e.g. interleukin-6 (IL-6), tumour necrosis factor-alpha (TNF-α)) (Rysz et al. 2006) and reduced circulating concentrations of anti-inflammatory cytokines (e.g. IL-10) (Zhang et al. 2010). Potential causes include chronic uraemia-induced leukocyte ligation, altered leukocyte subset distributions or HD treatment-induced endotoxin exposure (Zhang et al. 2010); (Hauser et al. 2008; Heine et al. 2012). Chronic leukocyte activation also increases the chemokine secretion (e.g. monocyte chemotactic protein-1 (MCP-1)), which positively associates with systemic inflammation (Papayianni et al. 2002) and may drive atherosclerosis in HD patients (Hu et al. 2016).

Both acute aerobic exercise and regular aerobic exercise training can promote favourable changes in MPs (i.e. reduced total and TF + MP concentration) (Babbitt et al. 2013; Highton et al. 2019), an effect which has also been observed in chronic kidney disease and renal transplantation (Highton et al. 2020). In addition, regular aerobic exercise training can reduce circulating pro-inflammatory cytokine and chemokine concentrations (Gleeson et al. 2011; Trøseid et al. 2004), and therefore, may be therapeutic for HD patients. However, HD patients report lack of time and opportunity, as well as safety fears as barriers to traditional exercise participation (Delgado and Johansen 2011). Intradialytic cycling (IDC), which is completed during routine HD treatment under the supervision of exercise professionals or nursing staff, circumnavigates these issues and has been shown to reduce arterial stiffness and cardiovascular risk (Toussaint et al. 2008).

Only one study has investigated the impact of IDC on MPs (Martin et al. 2018), reporting that an acute bout of IDC induced an increase in MP reactive oxygen species release but no change in MP numbers. This is theorised to promote a long-term anti-inflammatory environment through repeated exposure and adaptation, though the long-term effects have yet to be investigated.

An acute bout of moderate-intensity IDC may also prevent an HD-induced increase in IL-6 (Wong et al. 2017) (though non-significantly) or increase circulating IL-10 concentrations to a greater degree than that seen in non-exercising patients (Peres et al. 2015), though some studies investigating regular IDC training have found no impact on inflammatory markers (Dungey et al. 2017; Gołębiowski et al. 2012; Toussaint et al. 2008). However, previous research has exhibited methodological limitations, such as taking resting samples after HD initiation (Dungey et al. 2017), including inappropriate control groups (Gołębiowski et al. 2012) or using small sample sizes (Toussaint et al. 2008). Therefore, there is need for research investigating the impact of regular IDC on MPs and systemic inflammation that addresses these methodological issues. The aim of this study was to investigate how regular, moderate-intensity IDC affects resting MP concentration, TF expression and cytokine and chemokine concentrations.

This study found no evidence to suggest that a 6-month programme of thrice-weekly structured IDC training exacerbates circulating MP, cytokine, or chemokine concentrations or MP tissue factor expression.

Whilst a reduction in neutrophil-derived MP concentration was observed in the IDC group, thus suggesting a potential anti-inflammatory effect of regular IDC, there was insufficient evidence (p = 0.051) to firmly support this result, and therefore, this finding should be considered hypothesis-generating only. If future research reveals stronger evidence for this effect, possible mechanistic explanations that could be investigated include reduced uraemia-induced neutrophil activation or reduced neutrophil apoptosis resulting in reduced neutrophil-derived MP shedding. However, whilst the HD procedure can drive leukocyte apoptosis (Moser et al. 2003), the impact of IDC has not yet been investigated. As HD patients commonly experience impaired circulation (McIntyre et al. 2008), regular IDC training may restore systemic circulation and prevent hypoxia, reducing neutrophil activation. A reduction in circulating neutrophil-derived MPs would be expected to represent an anti-inflammatory effect, which could ultimately reduce thrombosis and cardiovascular risk, however, the impact of IDC in this regard can only be postulated based on the findings from this study.

Circulating concentrations of MPs derived from endothelial cells, platelets and monocytes are increased in patients with renal failure, and associate with vascular dysfunction (Daniel et al. 2008). The MP results of this study mimic those investigating acute IDC (Martin et al. 2018), and as such it may be that the exercise was not of a sufficient intensity in either case to elicit significant reductions in circulating MP concentrations, either acutely or on a long-term basis. However, as the mean power outputs achieved in this study were similar to those of previous IDC research (~ 21–22 W) (Dungey et al. 2015; Martin et al. 2018), this exercise intensity is likely to be representative of what can be reasonably expected of HD patients. As such, whilst regular aerobic exercise has been previously reported to elicit MP-mediated health benefits in other populations (Highton et al. 2018), there is as yet insufficient evidence to suggest that this can be achieved with regular IDC.

Supporting this, whilst previous research investigating acute IDC has suggested the possibility of an MP-mediated anti-inflammatory effect of long-term IDC (Martin et al. 2018), this was not clearly observed here. Therefore, it may be that any possible transient effect of acute moderate-intensity IDC on MPs is not sufficient to induce a negative feedback loop potent enough to elicit an overall anti-inflammatory effect. In addition, whilst previous research has displayed beneficial effects of regular moderate-intensity aerobic exercise on MP concentrations in the general population (Babbitt et al. 2013), this also was not observed here in this population of HD patients. This may be due in part to discrepancies in exercise intensity—Babbitt et al. incorporated objectively measured exercise intensity training (65% of VO_2max) whilst here it was subjective (RPE 12–14), which may have resulted in a, respectively, lower achieved intensity. It may also be that the potent effect of end-stage renal failure and the HD procedure on MPs (Amabile et al. 2011; Daniel et al. 2006) prevents the observation of any effects induced by exercise of this intensity. However, there is currently no evidence to suggest that regular IDC exacerbates MP status in HD patients, though further functional assays would be required to investigate this more directly, for instance a prothrombinase assay to assess prothrombotic potential.

As there was insufficient evidence here to suggest the presence of ad MP-mediated anti-inflammatory response, it is unsurprising that no IDC-dependent effects were observed in circulating cytokines. The diverse battery of cytokines assessed here typically exhibit either pro-inflammatory effects (TNF-α, IL-17a), anti-inflammatory effects (IL-2, IL-10), or both (IL-6) (Berg and Scherer 2005; Cortvrindt et al. 2017; Hoyer et al. 2008; Stenvinkel et al. 2005), though there was no evidence to suggest any exercise-dependent effects, either pro- or anti-inflammatory, despite initial evidence suggesting a reduction in TNF-α. In previous research in the general population, both acute and regular aerobic exercise participation have elicited favourable changes in these cytokines (Gleeson et al. 2011; Rhind et al. 1996), and similar anti-inflammatory effects have also been displayed following acute IDC (Peres et al. 2015). A previous study investigating regular IDC also found no observable change in circulating cytokines despite observing reductions in circulating intermediate monocyte proportion, though at a slightly lower exercise intensity (16 W) (Dungey et al. 2017). It may be that HD patients cannot routinely achieve an exercise intensity sufficient enough to impact their resting systemic inflammatory profile. MCP-1 and IL-8, pro-inflammatory chemokines that promote leukocyte migration and associate with atherosclerosis and cardiovascular risk (Hashmi and Zeng 2006; Kusano et al. 2004; Papayianni et al. 2002), also displayed no IDC-dependent effects, likely for the same reason. However, it is encouraging that these circulating mediators were not exacerbated by regular IDC, which points towards the ‘immunological safety’ of this form of exercise in this population.

Flow cytometry is considered the gold-standard technique most applicable to clinical research for the detection and characterisation of MPs due to its high throughput and the level of information it can provide concerning MP size, complexity and surface marker expression (Strasser et al. 2013; Van der Pol et al. 2014). In addition, flow cytometry allows for several steps of confirmation when detecting and measuring MPs, including confirmation based on size, internal complexity, and the expression of one or a combination of specific markers, which makes it easier to isolate the population of interest. Whilst this is possible with other techniques, such as nanoparticle-tracking analysis, it is more arduous and less comprehensive. Nanoparticle tracking analysis does have the advantage over traditional flow cytometry of being able to visualise the particles of interest (though this is possible with imaging flow cytometry (Lannigan and Erdbruegger 2017)). In addition, nanoparticle-tracking analysis typically provides a lower minimum detectable threshold (0.05 µm) than this flow cytometry technique (0.3 µm) (Highton et al. 2019), which is a previously identified limitation of flow cytometry (Burton et al. 2013). Therefore, an accepted limitation of the flow cytometry technique employed here is that it may be excluding a certain number of MPs below 0.3 µm in diameter; however, it could also be argued that using nanoparticle-tracking analysis could also incorporate unwanted exosomes in the 0.05–0.1 µm range in the analyses. Ideally, methodologies for isolating and analysing MP populations would employ two or more techniques to capitalise on the advantages of each technique and to ensure that the full size range is captured (Strasser et al. 2013; Szatanek et al. 2017). However, this was not possible in this study due to resource constraints, and therefore, flow cytometry was selected due to its advantages over other techniques (Szatanek et al. 2017). Further research should focus on harmonising the various techniques for MP detection to construct a comprehensive, standardised methodology.