Recent advances in treatment of haemodialysis

Abstract

Haemodialysis remains the most widely used treatment for patients with end-stage renal disease. Despite the progress that has occurred in the treatment of end-stage renal disease over the last six decades, there has been a failure to translate this into the desired clinical benefits, with morbidity and mortality rates among patients on haemodialysis remaining unacceptably high. Recently, however, there have been expectations that the significant advances that took place over the last few years may result in improved outcomes. New medications for the treatment of anaemia and secondary hyperparathyroidism, as well as novel trends in the areas of iron therapy, diabetes management and physical exercise are among the most important advances which, taken together, are changing the standards of care for patients on haemodialysis. The latest advances, of relevance not only to specialists in Renal Medicine but also to general practitioners caring for these patients, are reviewed in this collaborative paper.

Introduction

Haemodialysis remains the most widely used treatment for patients with end-stage renal disease; patients on haemodialysis number 350,000 in Europe and 3 million worldwide. Despite the progress that has occurred in the treatment of end-stage renal disease over the last six decades, there has been a failure to translate this into the desired clinical benefits; the mortality rates among patients on haemodialysis remain unacceptably high.¹ Equally, patient experiences remain poor with high levels of treatment-related morbidity. In the last few years, however, there have been significant advances that are changing the standards of care for patients on haemodialysis and promise improved patient outcomes. Among them, new medications for the treatment of anaemia and secondary hyperparathyroidism, as well as novel evidence in the areas of iron therapy, diabetes management and physical exercise, are probably the most important ones that seem to be shaping already the future standards of care in patients on haemodialysis. This review presents the highlights among the therapeutical advances in haemodialysis, of relevance not only to specialists in Renal Medicine but also to general practitioners caring for these patients.

Methods

This collaborative paper reviews the latest therapeutical advances in haemodialysis as they were presented by experts at the recent meeting ‘Frontiers in Haemodialysis’, held on 12 December 2019 at the Royal Society of Medicine in London. This was an international meeting that brought together experts in haemodialysis, to present and discuss on the exciting latest advances in therapeutics as well as on the most recent technological advances that promise to change the current clinical practice in the foreseeable future. The topics, as selected by the organising committee on the grounds of their novelty and clinical relevance, focused on the new medications for the treatment of anaemia and secondary hyperparathyroidism, and the novel evidence in the areas of iron therapy, diabetes management and physical exercise. The speakers were asked to present an overview of the latest advances after performing an extensive review of the literature on their topic. Subsequently, they were asked to contribute the respective chapter in this collaborative paper on the same topic presented in the meeting, including the points raised by the audience of the meeting as well. The coordinators of the meeting (DK and CC) reviewed, refined and bound together the chapters into this review paper.

Iron treatment in the post-PIVOTAL era

For over three decades, erythropoiesis-stimulating agents have been available for the treatment of the anaemia that so commonly occurs in patients with end-stage kidney disease. Control of anaemia with erythropoiesis-stimulating agents transformed the health of haemodialysis patients, but it became clear in the mid-1990s that supplementation of iron was necessary to optimise haemoglobin response, and moreover that the intravenous route was the most efficacious to deliver iron to these patients who were vulnerable to blood loss and poor oral iron absorption. Large-scale randomised trials were undertaken to test whether normalising haemoglobin was beneficial to kidney patients, although most of the studies were performed in non-haemodialysis chronic kidney disease. Nonetheless, signals of risk but little benefit appeared in the higher haemoglobin arm of the TREAT² study, and this led to a step change in clinical practice, especially as ESA are known to associate with other complications such as vascular access thrombosis. Despite the absence of studies in patients on haemodialysis, it was recommended that erythropoiesis-stimulating agent doses should be attenuated in this group, with haemoglobin targets limited at 110 g/L (United States) and 120 g/L (Europe) in erythropoiesis-stimulating agents-treated patients. Greater dosing with intravenous iron facilitated ESA dose reduction with the net effect that average ferritin levels were targeted at 900–1000 ug/L in some countries, notably the United States. However, some held the view that this relative ‘iron overload’ might be associated with risk, especially as in vitro studies indicated that higher iron environments could reduce immune function, and basic chemistry predicted a risk of oxidative stress with the potential for cardiovascular damage. In 2015, Bailie et al.³ reported data from the Dialysis Outcomes Practice Patterns Study (DOPPS) that higher iron dosing of 400 mg/month (approximately twice the dose in an average UK haemodialysis patient) was associated with all cause and cardiac specific mortality in patients on haemodialysis.

The anaemia clinical study group of the UK Kidney Research Consortium recognised that the issue of higher iron dosing in patients on haemodialysis should be investigated further, to assess whether current clinical practices were harmful or of benefit. The Proactive IV IrOn Therapy in HaemodiALysis (PIVOTAL) trial was thus developed.⁴ In 50 UK centres, 2141 patients who had commenced haemodialysis within the preceding 12 months were randomised 1:1 to either a proactive intravenous iron regime targeting ferritin at 700 ug/L or a reactive lower dose regime with ferritin around 200 ug/L. The primary outcome of this event driven study was a composite of major cardiac events and all-cause mortality. After an average follow-up of 2.2 years, the proactive dosing arm was found not only non-inferior to the lower dose regime, but superior, with a 15% reduction in events (p = 0.04). Transfusions were reduced by 21% and erythropoiesis-stimulating agent doses by 19% with higher iron dosing, and most components of the composite outcome, fatal/non-fatal myocardial infarction, non-fatal stroke and heart failure admissions, were significantly improved. There were no signals of increased infection risk with higher iron dosing. Deducing the likely mechanism of benefit of higher dose iron can only be speculative, but reduced erythropoiesis-stimulating agent doses, reduced transfusions and improved blood rheology would appear to be the main candidates.

So has clinical practice changed after PIVOTAL? The answer is, not dramatically in most UK renal units, at least not at this stage, and PIVOTAL will certainly not discourage current US practice with intravenous iron dosing. It is well known that new findings can take time to be absorbed into practice, but we anticipate that guidelines for anaemia management in haemodialysis patients will soon be modified to encourage an even more liberal use of intravenous iron in European patients on haemodialysis.

Erythropoiesis-stimulating agents, biosimilars or Hypoxia inducible factor stabilisers?

Anaemia in chronic kidney disease starts to manifest from chronic kidney disease stage-3 due to a reduction in erythropoietin production, chronic inflammation and impaired iron utilisation. It results in a reduced quality of life and increased cardiovascular morbidity and mortality. Iron availability is controlled by hepcidin (levels increased in chronic kidney disease) from the liver which regulates dietary iron absorption and iron recycling. Data from the UK Renal Registry⁵ show that ESAs are used in 90% patients on haemodialysis and 70% patients on peritoneal dialysis (PD). The introduction of erythropoiesis-stimulating agents signalled the rise of a new era in the treatment of patients on haemodialysis, as it reduced dramatically the need for transfusion and complications from iron overload. More recently, however, studies such as the CHOIR⁶ and the TREAT² trials showed that higher Hb from erythropoiesis-stimulating agents lead to an increased risk of cardiovascular events, highlighting the need for careful use of the ESAs and avoiding Hb levels higher than 120 g/L.

Biosimilars are copies of biologic drugs developed after patent expiry and are approved partly on studies done with the originator. They are a major factor contributing to the economic stability of healthcare systems worldwide but concerns remain regarding efficacy and safety.^7,8 EPO-PASS⁹ and MONITOR-CKD5¹⁰ are two of the trials which showed consistent dosing, stable outcomes and similar rates of side effects to originators with biosimilars now licenced in Europe, offering the prospect of considerable savings.

Hypoxia-inducible factor is involved in the activation of various genes essential to the hypoxia response (e.g. Vascular endothelial growth factor) and its expression level is regulated by hypoxia-inducible factor prolyl hydroxylase. Hypoxia inducible factor stabilisers or hypoxia-inducible factor prolyl hydroxylase inhibitors are newer drugs to be used in the treatment of renal anaemia. They are all taken orally and currently undergoing Phase III trials. They cause associated dose-dependent increases in Hb. Roxadustat has been shown to increase Hb (and lower cholesterol) compared to placebo in patients on haemodialysis and be non-inferior to epoetin-alpha in haemodialysis patients with similar rates of adverse events.¹¹ Among the main concerns with the use of the hypoxia inducible factor stabilisers are concerns regarding tumour proliferation due to the diversity of genes targeted by hypoxia inducible factor.

In summary, erythropoiesis-stimulating agents and iron remain the mainstay of treatment but concerns include cost and risks including hypertension and cardiovascular events. Biosimilars may offer significant potential cost savings with similar beneficial effects on Hb thus leading to increased availability and utilisation. Hypoxia inducible factor stabilisers in offer an exciting potential new treatment strategy for renal anaemia but long-term effects are still unknown with Phase III trials ongoing.

Alfacalcidol, paricalcitol, cinacalcet or surgery?

Disturbances in mineral homeostasis are detectable at very early stages of chronic kidney disease. As a result, by the time most patients arrive on haemodialysis, there is a long vintage of mineral-bone disease. Though the ensuing increased cardiovascular risk is well-recognised, there is also a several-fold increased risk of fractures in haemodialysis patients as compared with transplant recipients, even when accounting for patient characteristics.¹² Despite the burden of disease that arises from mineral-bone disease, the optimal management strategy in secondary hyperparathyroidism for patients on haemodialysis remains unclear; this was reflected in the Kidney Disease: Improving Global Outcomes 2017 Clinical Practice Guideline¹³ which asserted no preference for ‘calcimimetics, calcitriol, or vitamin D analogs, or a combination of calcimimetics with calcitriol or vitamin D analogs’.

While equipoise may exist within guidelines, in practice alfacalcidol (non-selective vitamin D receptor activator) remains the first-line treatment option probably by the majority of the nephrologists in the UK, with selective vitamin D receptor activator (paricalcitol) and the calcimimetic cinacalcet ranking well behind; surgery seems to be no more the first-line treatment. The low take-up of calcimimetics as opposed to vitamin D receptor activators in first-line therapy may have been influenced by the PARADIGM trial.¹⁴ This multi centre randomised controlled trial compared cinacalcet against an open-label mixed-bag of vitamin D receptor activators administered either parenterally on haemodialysis or orally. The trial could not demonstrate a difference in parathormone control between these two groups over a 12-month follow-up period.

Vitamin D receptor activator choices seem to be influenced by geographical variations, as there is a predilection for the use of alfacalcidol in Northern Europe, as opposed to paricalcitol in Southern Europe. The evidence supporting the use of one over the other is limited. A well-designed Danish cross-over study¹⁵ failed to show supremacy of alfacalcidol over paricalcitol, while demonstrating the parathormone-lowering ability of both drugs along with comparable hypercalcaemia and hyperphosphataemia.

EVOLVE trial,¹⁶ the largest randomised controlled trial in haemodialysis to date (>3800 patients), has been subject to a significant amount of post hoc re-analysis. In comparing cinacalcet against placebo, EVOLVE failed to demonstrate an improvement in the clinically relevant composite primary end point. The trial could be grouped with other randomised studies in haemodialysis (q.v. HEMO, 4D and Normal Haematocrit Trial), which have also failed to show a benefit in the interventional arm. However, a number of methodological issues in EVOLVE means that therapeutic nihilism should not hide the possibility of benefit, particularly in the older patient on haemodialysis.¹⁷

Decisions around surgery as opposed to calcimetics remain difficult. United Kingdom NICE guidance from 2007 (revised in 2013)¹⁸ continues to advocate a surgical approach unless there are contraindications to parathyroidectomy; this increasingly appears out-of-step with the audience’s clinical practice based upon patients’ preferences to avoid surgery. This is understandable given the evidence of 1% mortality and 4% risk of significant complications associated with the operation,¹⁹ as well as United States evidence of a 40% increased rate of admissions after surgery.²⁰ Changes in the cost of calcimimetics over the next few years, as well as the appearance of etelcalcitide – a potentially better-tolerated intravenous preparation – may result in changes in practice.

As many questions regarding the treatment of mineral-bone disease in patients on haemodialysis remain unanswered, the need for further prospective multicentre trials particularly in older and more co-morbid is evident more than ever. To this end, the outcomes of the SIMPLIFIED trial, examining native vitamin D supplementation in haemodialysis, are awaited with great interest.

Diabetes and haemodialysis

Diabetic nephropathy is the leading cause of end-stage renal disease globally. As the number of individuals living with diabetes increases, the prevalence of those reaching end-stage renal disease and requiring renal replacement support is also set to rise. A significant number of these individuals also suffer from co-existent diabetic complications, reduced physical and cognitive functioning as well as significantly lower quality of life indices compared to those with diabetes not requiring haemodialysis.

Monitoring of glycaemic control in haemodialysis patients

Haemoglobin A1C remains the main marker of assessing glycaemic control in diabetes, including in chronic kidney disease; however, its role in haemodialysis patients is uncertain.²¹ The haemoglobin A1c assay results are affected by red cell survival, red cell turnover, the haemoglobin concentration (anaemia) and glucose variability amongst other factors.²² The presence of iron deficiency or uraemia can lead to overestimation of haemoglobin A1c. On the other hand, recent transfusions, iron infusions, use of erythropoiesis-stimulating agents and any concurrent haemolysis could lead to underestimation of the haemoglobin A1c.^21,22 The utility of alternative biomarkers such as fructosamine and glycated albumin has not been sufficiently validated outside research settings.

Self-monitoring of blood glucose is key to understanding the day-to-day glycaemic variability. However, as self-monitoring of blood glucose is user- and instrument-dependent; it is therefore important to ensure those receiving haemodialysis have regular assessments of their technique and glucometer.²³ Not all glucometers are glucose-specific; some also detect other sugars such as metabolites of icodexrtin (Extraneal) used in PD. This may lead to erroneous hyperglycaemia. Furthermore, from a practical perspective, many haemodialysis patients are unable to do their own self-monitoring of blood glucose testing; in such instances, recordings taken during haemodialysis can be critical in providing information on glycaemic control. Continuous glucose monitoring may also provide information on the glucose variability and may have a role in those with unpredictable hypoglycaemia.²³ Flash glucose sensor monitoring technology allows for the measurement of interstitial fluid glucose values using a sensor applied to the skin (usually on the upper arm) and scanned with a reader. It may facilitate estimation of glucose values in those who find it difficult to use finger-prick testing and may also allow carers to monitor glucose profiles more easily. However, the accuracy of continuous glucose monitoring and flash glucose sensor monitoring in people with diabetes on haemodialysis continues to be debated. Individuals with diabetes on haemodialysis can avail of flash glucose sensor monitoring technology on the National Health Service.

The haemoglobin A1c should be individualised but a target of 58–68 mmol/mol (7.5%–8.5%) without hypoglycaemia has been suggested.²² Given the challenges with the accuracy of glucose control metrics, ensuring regular reviews of haemodialysis blood glucose profiles and interpreting self-monitoring of blood glucose recordings alongside haemoglobin A1c measurements taken every three to four months is perhaps the most practical way of monitoring control.

Diabetes management

Starting haemodialysis can be a challenging time for those with diabetes. There is an increased requirement for monitoring as the process of haemodialysis can significantly affect the behaviour of the anti-hyperglycaemic agents and glucose levels. Blood glucose values can tend to fall during the second half of the haemodialysis sessions, while extrinsic factors such as activity, access/timing of meals, transport times and even timing of haemodialysis sessions may impact on the glucose values. In addition, the dextrose concentration in the dialysate may indirectly influence glucose readings.

Oral antihyperglycaemic medications will usually require a downwards dose adjustment, apart from Linagliptin (Table 1). Longer-acting agents such as sulfonylureas are not recommended given their potential to induce hypoglycaemia. Mechanistically, sodium glucose co‐transporter 2 inhibitors require a working kidney, and therefore are not recommended. The routine use of glucagon-like peptide 1 agonists is also not recommended currently given the limited data existing outside research settings²⁴ Insulin therapy remains the mainstay of management in many centres. Regimes have to be tailored to the specific needs of the individual, the haemodialysis schedules, need for external help to inject and the agreed haemoglobin A1c and blood glucose targets. In patients on insulin prior to haemodialysis, the deficient excretion of exogenous insulin by the kidneys and the reduction in peripheral insulin resistance can lead to a reduction in the total daily insulin requirements.²⁵ Often, a downwards adjustment of insulin doses by 25%–50% is required at the start of haemodialysis. Individuals may also require less insulin on dialysis days than on non-dialysis days. Regular input from diabetes professionals is, therefore, highly desirable.

Table 1.

Oral hypoglycaemic agents in haemodialysis individuals.

Drug class	Drug/s	Recommendation in HD individuals
Biguanides	Metformin	Not recommended for those on haemodialysis
Sulphonylureas	Gliclazide Glimepiride Glipizide Glibenclamide Tolbutamide	Use cautiously/not recommended for those on haemodialysis as may cause post-dialysis hypoglycaemia. Not recommended for those on haemodialysis
Dipeptidyl peptidase-4 inhibitors	Linagliptin Sitagliptin Saxagliptin Alogliptin Vildagliptin	Linagliptin – no dose reduction required (5 mg OD) Sitagliptin – recommend dose reduction to 25 mg Saxaglitpin – recommend dose reduction to 2.5 mg Alogliptin – recommend dose reduction to 6.25 mg Vildagliptin – recommend dose reduction to 50 mg
glucagon-like peptide 1 receptor agonists/mimetics	Exenatide (daily or once weekly) Liraglutide (daily or once weekly) Dulaglutide (once weekly) Semaglutide	Not recommended for those on haemodialysis Limited evidence to support use when eGFR < 30 mls/min Not recommended for those on haemodialysis Limited evidence to support use when eGFR < 15 mls/min. One study found haemodialysis did not affect the pharmacokinetics of Semaglutide
Sodium glucose co‐transporter 2 inhibitors	Dapagliflozin Canagliflozin Empagliflozin	Not recommended for those on haemodialysis
Other short‐acting insulin secretagogues	Nateglinide Repaglinide	Use cautiously High risk of hypoglycaemia
Alpha glucoside antagonist	Acarbose	Not recommended for those on haemodialysis
Thiazolidinedione	Pioglitazone	Not recommended for those on haemodialysis and may increase fluid retention and oedema

As new starters on haemodialysis often struggle with hypoglycaemia, especially during dialysis or within the 24 h after dialysis, early diagnosis and treatment of hypoglycaemia are important. This can be done easily with the intake of 10–20 g of a low‐glycaemic index carbohydrate during the second hour of dialysis.²¹ Individuals with diabetes on haemodialysis may require additional help to recognise and manage any hypoglycaemia in the community.

Exercise in haemodialysis: lessons learnt from the CYCLE-HD study

People on regular haemodialysis are physically inactive, with less than 50% undertaking exercise once a week. The stark reality is that such patients have significantly increased mortality compared to patients who undertake regular exercise. The reasons for physical inactivity and reduced functional capacity are complex and inter‐related, with skeletal muscle catabolism, chronic inflammation, anaemia, malnutrition, the burden of co‐morbid diseases and ‘enforced’ sedentary time during haemodialysis all contributing.²⁶ Many of these drive cardiovascular disease processes in this cohort of patients and in the general population; exercise interventions have been shown to modify many of these risk factors. While there is increasing evidence about the beneficial effects of exercise interventions on quality of life, functional capacity, aerobic fitness and muscular strength, there are few compelling data on the effects of such programs on cardiovascular outcome measures.²⁷

The CYCLE-HD Study was a prospective randomised controlled trial designed to detect a significant and clinically meaningful reduction in left ventricular mass (measured using state-of-the-art cardiac magnetic resonance imaging), a surrogate marker of cardiovascular outcomes and mortality in haemodialysis patients.²⁸ Participants were randomised to participate in either a six-month programme of cycling exercise during dialysis (exercise group) or continuing usual care dialysis (control group). Pre-specified secondary outcomes included measures of physical functioning and quality of life.

The trial recruited to target with 130 participants completing baseline assessments; 101 individuals underwent repeat cardiac magnetic resonance imaging scans and therefore completed the trial protocol (control group n = 50, exercise group n = 51). Groups were well-matched for age, sex, ethnicity, haemodialysis vintage and co-morbidities. Participants in the exercise group progressed through the exercise training programme, completing 71.7% of planned exercise sessions. At six months, left ventricular mass reduced in the exercise group and increased in the control group, representing a treatment effect of 11.1 g (p < 0.001) between groups in favour of the intervention. There was some evidence to suggest that those randomised to exercise increased their general physical activity levels, with the average daily step count being higher in the exercise group at follow-up by 732 steps (95% CI: −75, 1539; p = 0.08). Overall, the CYCLE-HD data suggest that a programme of intra-dialytic cycling can improve the cardiovascular health of patients on maintenance haemodialysis. The programme was well tolerated and deliverable with good adherence and with no adverse safety signals related to the cycling intervention. Whether this translates into a survival benefit requires further study.

Intradialytic exercise is now recommended by the UK Renal Association as a treatment for enhancing physical functioning in dialysis patients without contraindications.²⁹ A number of lessons were learnt during the CYCLE-HD trial about the potential barriers to implementation but most importantly, dedicated staff are needed to set up and maintain the programme. This is crucial both to ensure participation from patients and longevity within the National Health System infrastructure. In addition, there remains a perception from patients that exercise would either not be safe or beneficial for them at an individual level; strategies to reach those less motivated or multi-morbid patients remain a priority.

Conclusion

Recent advances in therapeutics of patients on haemodialysis have created high expectations for improved clinical outcomes in these patients in the foreseeable future. These benefits are likely to accrue in the areas of anaemia through the optimised use of iron and novel drugs such as hypoxia inducible factor stabilisers, secondary hyperparathyroidism potentially from earlier use of calcimimetics, optimised diabetes management and enhanced physical conditioning of patients on haemodialysis. Collaborative studies will help identify how these new therapies can be best used to improve patient outcomes for people requiring enduring haemodialysis.

Take home messages:

• Recent evidence encourage a more pro-active use of intravenous iron in European patients on haemodialysis.

• Whereas erythropoietin-stimulating agents and iron remain the mainstay of anaemia treatment, biosimilars may prove to be more cost-effective than prototype erythropoietin-stimulating agents and hypoxia inducible factor stabilisers may prove to be the future treatment of choice.

• Despite the burden of disease that arises from mineral-bone disease, the optimal management strategy in secondary hyperparathyroidism for patients on haemodialysis remains unclear. Alfacalcidol remain popular for control of secondary hyperparathyroidism; however, easier access to calcimetics may change practice in the foreseeable future.

• Diabetes control can be erratic in patients on dialysis especially during the starting period on haemodialysis. Close glucose monitoring is necessary yet can be challenging in these patients. Both oral antihyperglycaemic agents as well as insulin usually require a downwards dose adjustment. Sodium glucose co-transporter 2 inhibitors and glucagon-like peptide 1 agonists and long-acting agents, such as sulfonylureas, are not recommended in these patients. Individualised and multidisciplinary approach is highly recommended for these patients.

• The CYCLE-HD study suggested that intradialytic cycling can be well tolerated and can improve the cardiovascular health of patients on maintenance haemodialysis.

Declarations

Competing Interests

The Author(s) declare that there is no conflict of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Ethics approval

Not required.

Guarantor

D.K.

Contributorship

James Burton reviewed the literature and wrote the section ‘Exercise in haemodialysis: lessons learnt from the CYCLE-HD study’, reviewed and revised the manuscript.

Richard W Corbett reviewed the literature and wrote the section ‘Alfacalcidol, paricalcitol, cinacalcet or surgery?’, reviewed and revised the manuscript.

Philip A Kalra wrote the section ‘Iron treatment in the post-PIVOTAL era’, and has reviewed the manuscript.

Prashanth Vas reviewed the literature and wrote the section ‘Diabetes and haemodialysis’, reviewed and revised the manuscript.

Vivian Yiu reviewed the literature and wrote the section ‘Erythropoiesis-stimulating agents, Biosimilars or hypoxia inducible factor stabilisers?’, reviewed and revised the manuscript.

Tina Chrysochou reviewed the literature and wrote the section ‘Conclusions’, reviewed and revised the manuscript. She has also contributed to the idea of this collaborative paper and invited the authors.

Dimitrios Kirmizis reviewed the literature and wrote the section ‘Introduction’, reviewed and revised the manuscript. He has also conceived the idea of this collaborative paper, selected the topics and invited the authors.

All authors approved the version to be published.

Dimitrios Kirmizis and Tina Chrysochou accept direct responsibility for the manuscript.

ORCID iDs

Richard W Corbett https://orcid.org/0000-0002-4607-8554

Dimitrios Kirmizis https://orcid.org/0000-0002-5093-1570