Cardiovascular benefits of daily haemodialysis: peeling the onion

Patients with chronic kidney disease (CKD) suffer rates of cardiovascular (CV) morbidity and mortality that are far in excess of those found in the general population. This excess mortality is particularly striking amongst younger people [1]. The modification of so-called ‘classical’ CV risk factors has little impact on overall rates of morbidity and mortality, particularly in patients undergoing dialysis for end-stage renal failure (CKD5D). The pathophysiological processes underpinning these high rates are complex. CKD patients may suffer from ‘traditional’ thrombotic occlusive coronary artery disease. However, increasing severity of CKD becomes relatively less common with increasing susceptibility to sudden cardiac death and heart failure [2]. This is a result of the cumulative effects of additional factors [3]. These changes to the vasculature leave the patient vulnerable to smaller ischaemic insults particularly in the setting of the haemodynamic stresses present during haemodialysis (HD) [4]. In the last few years, and in particular since the publication of the pivotal robust randomised controlled clinical trial (RCT) Frequent Hemodialysis Network (FHN) studies of daily HD [5, 6], there has been an increasing focus on possible CV benefits related to this form of renal replacement therapy. Registry-based data are suggestive that daily HD may be associated with CV mortality rates superior to conventional therapy [7–9] and akin to those seen in patients receiving cadaveric renal transplantation [10, 11]. Current conventional HD is largely an industrialized process, with inadequate attention to the role that the dialysis treatment itself may play in the development and promulgation of uraemic-related disease states. HD is capable of exerting significant recurrent systemic circulatory stress. This predominantly haemodynamic injury can therefore result in a mixed picture of direct perfusion-related injury, local/systemic inflammation and potentiation of further cycles of injury [12]. This edition of the journal carries further analysis of the unit-based FHN study, in this case relating to effects on heart rate variability (HRV) [13]. It provides additional evidence of potential CVS benefits attendant to extended dialysis schedules. This accompanying review aims to explore potential beneficial effects on the CVS in CKD5D patients by examining the potential role that extended dialysis schedules (both in terms of dialysis hours and frequency) might have in ameliorating or abrogating the wide range of factors involved in the pathophysiology of excessive cardiovascular disease (CVD). To do this it is important to acknowledge the multi-factorial nature of the problem, but also that these factors conspire in a concerted fashion to prime for CVS injury, initiate injury and by providing additional stressors as well as propagating ongoing cycles of injury. Risk of CV death in patients receiving dialysis is strongly related to a combination of structural and functional abnormalities at all levels of the CVS (from microcirculation to major conduit arteries and into the myocardium itself). The major manifestations of these underlying vascular vulnerabilities are cardiac sudden death, heart failure and ischaemic brain injury. Abnormalities of microcirculation, peripheral arterial compliance and defective vasoregulation all increase the risk of demand ischaemia (in multiple vascular beds). Underlying ventricular morphological defects (microcalcification, fibrosis and reduced cardiac compliance) also predispose to cardiac arrhythmias. Daily dialysis may have impacts on all of these domains of importance. Arterial compliance is characteristically reduced in CKD5 patients. Arterial walls alter their structure and function in response to atherogenic and atherosclerotic factors. As a result the structural changes can include the activation and proliferation of smooth muscle cells, and rearrangement of cellular elements and extracellular matrix of the vessel wall, increasing arterial stiffness. The humoral and haemodynamic factors that predispose to this change in the non-uraemic ageing population are highly prevalent in patients with CKD. Arterial calcification with deposition of hydroxyapatite crystals on elastic fibres aggravates this increase in arterial stiffness yet further [14]. Alterations in the diastolic reinforcement of central arterial blood pressure (BP) result in an increase of the central systolic BP peak, and increased myocardial oxygen demand. This can result in myocardial ischaemia even in the absence of coronary arterial disease [15]. Data are limited at present on the effects of extended dialysis schedules on arterial stiffness. Demirci et al. reported increased dialysis hours (24 versus 12 h per week) over a 15-month period were associated with a reduction of pulse wave velocity (PWV) of around 1.5 m/s. This was apparantly not associated with any observable difference in BP between the two groups (a major determinant of PWV) [16]. Vascular calcification is well recognized as an important determinant of arterial stiffness. It is both highly prevalent and rapidly progressive in conventionally haemodialysed patients [17]. Daily dialysis has been widely (but not universally) reported to be associated with improved control of factors associated with the development of CKD MBD and vascular calcification [18]. A study of the prospective changes in coronary artery calcification in a small cohort of nocturnal HD (NHD) patients was suggestive of a beneficial effect of daily dialysis on the progression of vascular calcification [19]. The cardiac microcirculation is becoming increasingly recognized as being of importance in the development of demand-induced myocardial ischaemia [20]. Impaired microcirculatory function in CKD5D patients is associated with increased left ventricular mass (LVM) and arterial remodelling [21]. The assessment of subcutaneous dermal capillaries is the primary method for the assessment of the microcirculation. These vessels are representative of the microvascular supply to the heart and kidneys. Microcirculatory function may also be assessed indirectly [via surrogate humoral markers including asymmetric dimethylarginine (ADMA) and endothelin]. A preliminary report suggests an improvement in the endothelial function with thrice weekly extended nocturnal dialysis [22]. Conversion to more frequent NHD has also been reported to be associated with improvements in endothelial-dependant vasodilatation, but without significant effects on serum ADMA levels [23]. Daily dialysis has been reported to be associated with reductions in LVM in both ongoing observational studies and RCT [24]. Conversion to NHD is associated with an improvement in the coherent orchestration of efficient ventricular contraction and relaxation (as assessed by echocardiography). Incubation of cardiac myocytes in medium containing uraemic plasma harvested from this study resulted in a modification in the expression of a variety of genes implicated in the development of cardiac fibrosis [25]. Defective mechanics of contraction, especially relating to myocardial unwinding, directly contributes to reduced myocardial perfusion due to effects on transmural strain gradients. This produces a reduction in the suction component of the pressure wave necessary to generate myocardial perfusion. LVH itself is directly associated with a reduction in myocardial suction [26]. It is unclear the effect of an improved control of bone and mineral metabolism (via PTH, serum phosphorous, PTH and FGF23 levels) on LVM. These changes might lead to reductions in fibrotic/hypertrophic signalling, in addition to those from reductions in recurrent ischaemia and pressure/volume overload. CKD in general, and dialysis in particular, is pro-inflammatory state. It is becoming increasingly apparent that this systemic inflammatory state is a critical driver of CVD in this patient group (through a wide variety of interlinked mechanisms). Extended dialysis schedules are associated with a significant reduction of markers of the inflammatory state (such as CRP, IL 6 and TNF α) [27]. Endotoxin (without sepsis) was initially proposed as a stimulus for immune activation in the pro-inflammatory state of congestive heart failure (CHF). Endotoxin is released by bacterial cell wall breakdown, within and beyond the gut lumen, from effective host defence mechanisms and by autolysis. Endotoxin enters the circulation via translocation, with passage of intact bacteria and macro-molecules such as endotoxin across the intestinal barrier [28, 29]. Exposure to endotoxin, a profoundly pro-inflammatory stimulus, results in the release of a wide variety of pro-inflammatory cytokines, and has been implicated in a broad range of other pathophysiological responses, including oxidative stress, endothelial dysfunction and impaired circulatory autoregulation. We have reported significant incremental endotoxaemia with worsening renal function across the range of CKD [30]. In HD patients the severity of endotoxaemia was associated with both the drivers (ultrafiltration volume and rate) and the consequences (dialysis-induced myocardial injury) of dialysis-induced circulatory stress. The use of both short daily HD and NHD is associated with almost normalized levels of circulating endotoxin compared with conventional HD patients [31]. There is a high prevalence of hypertension in CKD5D patients, with estimates from 25 to 80%. The cause of this has been principally attributed to loss of renal regulation of salt and water excretion leading to extracellular volume overload with consequent adverse cardiac remodelling [32]. A detailed discussion of the interaction of salt and water overload on the CVS is beyond the scope of this review, but extended dialysis schedules by dint of both the period of time available for ultrafiltration (allowing more efficient depuration of sodium) and reduced interdialytic intervals (reducing accumulation) are associated with improved BP and volume status. In addition to the more obvious potential benefits (BP etc.), improved volume control may also impact on a number of other significant drivers of CV mortality. Obstructive sleep apnoea is a common feature in patients receiving maintenance HD and is associated with hypertension, LVH and increased CV mortality. Conversion to NHD is associated with a significant reduction in this problem, with improvement of volume control. This is associated with an attendant reduction in upper airway oedema (as measured directly by magnetic resonance imaging) whilst prone [33]. Additional effects may also include benefits attributable to non-osmotic effects of sodium accumulation and intestinal wall congestion, allowing translocation of pro-inflammatory factors from the bowel analogous to that seen in CHF. This edition of NDT carries a study from Chan et al. reporting an advantageous effect of daily dialysis schedules from the FHN trials on HRV. The FHN Daily Trial randomized 245 patients to receive 12 months of 6 versus 3 times per week in-centre HD. Two hundred and seven patients had baseline Holter recordings. From these, a variety of parameters were extracted. The key findings were daily HD increased the low frequency component of HRV (potentially related to improved LVM), and also increased beat-to-beat variability. A more detailed discussion of the drivers and importance of HRV is contained within this excellent paper. HD patients are caught in a perfect storm having acquired multiple hits on autonomic function, as a result of underlying disease aetiologies, drugs and consequences of dialysis therapy. All are present at precisely the time their circulations are challenged with circulatory stress and haemodynamic instability. The principal role for the autonomic system in this setting is the maintenance of end-organ perfusion in the face of fluxing BP. The combination of impaired sensing/response of these changes (baroreflex sensitivity), defective autonomic control and microcirculatory level issues relating to response (mediated by uraemic alterations to microcirculatory function) [34] combines with factors such as endotoxaemia to incapacitate autoregulation. This results in ischaemic challenge to a wide range of vulnerable vascular beds. It also increases the risks of the haemodynamic instability itself by reducing the CV contractile reserve and vasoregulatory accommodation in the face of HD with ultrafiltration. The possibility that extended dialysis schedules might impact on vasoregulation at multiple levels (from left ventricle and stiffening of central arteries reducing baroreceptor discharge to microcirculatory response) is a central component of the potential CV benefits being attributed to this group of therapies. The importance of factors relating to the dialysis procedure itself in understanding the pathophysiology of elevated CV risk is becoming increasingly apparent. HD-induced myocardial stunning is a reversible left ventricular systolic dysfunction which occurs during HD, returning towards normal at the end of the HD treatment. It is associated with both global and segmental cardiac ischaemia, impaired haemodynamic response to dialysis, significant elevations in cardiac troponin and increased risk of CV events and mortality. HD-induced stunning is common, occurring in at least two-thirds of HD patients, and it appears that a fixed reduction in the systolic function occurs after only 12 months of recurrent cardiac injury, setting a trajectory towards myocardial hibernation, adverse cardiac remodelling and ultimately cardiac failure. These effects appear to be important in a wide range of critical end organs beyond the heart itself. Daily dialysis schedules are associated with a reduction in ultrafiltration demands (due to reduced interdialytic period) and reduced circulatory stress (due to reduction in the ultrafiltration rate). The improved haemodynamic tolerability is associated with significant reductions in both the prevalence of dialysis-induced myocardial stunning, and (in those effected) a reduction in the severity (as assessed by a number of stunned segments and degree of reduction of fractional shortening within them) [35]. Patients receiving HD are seldom in a stable state. Although many studies have sought to categorize the CV status of subjects at a point in time this ignores the biological reality of the evolution of overall disease burden during the patient lifetime. All of the issues already discussed function in a concerted manner to conspire against the patient. Specific self-propagating cycles of injury are already emerging. For example, it is already apparent that those patients experiencing dialysis-induced cardiac stunning (driven by ultrafiltration requirements and haemodynamic instability) develop fixed reduction in ventricular contractile function. This is associated with an increasing inability to maintain BP in the face of ultrafiltration, thereby becoming more unstable and aggravating the cardiac insult [36]. Multiple cycles of injury are probably intercalated. Dialysis-induced end-organ ischaemia may be important in the pathophysiology of the rapid loss of residual renal function seen in HD, this results in increasing interdialytic weight gain (IDWG) with resulting increasing requirements for haemodynamically traumatic fluid removal. Extended dialysis schedules have the potential to directly ameliorate these kinds of positive feedback loops of injury. This assumes that their use delivers realistic reductions in circulatory stress (e.g. by delivering reduced IDWG), which may be an important discriminant of CV effects reported between the varying intensities of dialysis schedule. At present the only issue relating to CVD in HD patients that we fully understand is that the pathophysiology is complex. This complexity is due both to the range of issues involved and the way that they may interact. The central role that the dialysis procedure itself plays in this landscape is increasingly recognized. We are starting to acknowledge the ability of modified dialysis procedures (in particular extended dialysis schedules) over the current conventional norms of frequency and duration to provide a range of benefits cutting across the range of relevant domains. These are related both to physiological vulnerability and the direct negative effects of conventional thrice weekly HD. An increasingly evolved understanding of the structural and functional consequences of dialysis will allow us to identify individual components of the pathophysiology and utilize those to test refinements of dialysis procedure and patient selection. This is an essential preamble to intelligent design and execution of future RCTs of the application of daily dialysis to meaningfully impact on the objective and subjective burden of disease on these patients. I declare that the results presented in this paper have not been published previously in whole or part, except in abstract format. (See related article by Chan et al. Effects of daily hemodialysis on heart rate variability: results from the Frequent Hemodialysis Network (FHN) Daily Trial. Nephrol Dial Transplant 2014; 29: 168–178.)