Prevention and treatment of hyperphosphatemia in chronic kidney disease

review

www.kidney-international.org

Prevention and treatment of hyperphosphatemia in chronic kidney disease Marc G. Vervloet1 and Adriana J. van Ballegooijen1,2 1

Department of Nephrology, Amsterdam Cardiovascular Sciences, VU University Medical Center, Amsterdam, the Netherlands; and Department of Epidemiology and Biostatistics, Amsterdam Public Health Institute, VU University Medical Center, Amsterdam, the Netherlands

2

Hyperphosphatemia has consistently been shown to be associated with dismal outcome in a wide variety of populations, particularly in chronic kidney disease (CKD). Compelling evidence from basic and animal studies elucidated a range of mechanisms by which phosphate may exert its pathological effects and motivated interventions to treat hyperphosphatemia. These interventions consisted of dietary modifications and phosphate binders. However, the beneficial effects of these treatment methods on hard clinical outcomes have not been convincingly demonstrated in prospective clinical trials. In addition, exposure to high amounts of dietary phosphate may exert untoward actions even in the absence of overt hyperphosphatemia. Based on this concept, it has been proposed that the same interventions used in CKD patients with normal phosphate concentrations be used in the presence of hyperphosphatemia to prevent rise of phosphate concentration and as an early intervention for cardiovascular risk. This review describes conceptual models of phosphate toxicity, summarizes the evidence base for treatment and prevention of hyperphosphatemia, and identifies important knowledge gaps in the field. Kidney International (2018) 93, 1060–1072; https://doi.org/10.1016/ j.kint.2017.11.036 KEYWORDS: chronic kidney disease; dietary phosphate; hyperphosphatemia; intestinal phosphate absorption; phosphate binders Copyright ª 2018, International Society of Nephrology. Published by Elsevier Inc. All rights reserved.

Correspondence: Marc Vervloet, Department of Nephrology, VU University Medical Center, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands. E-mail: [email protected] Received 30 October 2017; revised 21 November 2017; accepted 27 November 2017; published online 23 March 2018

1060

A

bnormal phosphate metabolism is one of the key disturbances in chronic kidney disease (CKD). It is now recognized that overt hyperphosphatemia occurs rather late in the process of CKD progression, usually at stage 4 and onward.1 However, adaptive mechanisms, particularly high concentrations of parathyroid hormone (PTH) and fibroblast growth factor-23 (FGF23), antedate the development of overt hyperphosphatemia, promoting kidney phosphate excretion.2 Because these adaptive mechanisms may be directly involved in uremia-associated pathologies, it is difficult to untangle assumed phosphate toxicity from pathogenetic effects of these adaptive responses. Moreover, a reduction of phosphate exposure by dietary intervention or inhibition of intestinal phosphate absorption does not normalize elevated concentrations of PTH and FGF23. Other factors such as vitamin D deficiency, inflammation, and autonomous overproduction of these hormones may explain the limited effects of phosphate-targeted intervention on their circulating levels. Maintaining normal phosphate balance is of crucial importance for many physiological processes including bone mineralization. Phosphate homeostasis is determined by modulation of intestinal uptake of dietary phosphate, renal phosphate reabsorption of ultrafiltered phosphate, and the shift of intracellular phosphate between extracellular and bone storage pools (Figure 1).3,4 It is well established that phosphate is one of the major factors in the maintenance of bone health and that phosphate deficiency results in bone pathology, as seen in patients with specific monogenic diseases, leading to isolated renal phosphate wasting syndromes.5 Hyperphosphatemia per se usually is asymptomatic. Morbidity associated with hyperphosphatemia is the consequence of acquired structural or functional6 abnormalities, including vascular calcification, which has been recently summarized7 but is beyond the scope of the current review. This is important because the justification for treating hyperphosphatemia is based on the assumption that associated abnormalities are caused by abnormal phosphate homeostasis. The second assumption is that a reduction in phosphate concentration over time toward the normal range is accompanied by a parallel decline in morbidities and death. Comparable assumptions formed the base to restore hemoglobin concentration in renal anemia to near normal levels by epoetin, aiming to improve clinical outcome, but proved untrue.8 Kidney International (2018) 93, 1060–1072

review

MG Vervloet and AJ van Ballegooijen: Managing phosphate in CKD

Phosphate balance Total adult body stores 700 g, 85% in:

Absorption (300 mg/d)

Resorption (300 mg/d)

1200 mg/d Absorption (960 mg/d)

Blood <1% Phosphate pool

Secretion (150 mg/d) 1350 mg/d Urinary excretion (800 mg/d) Fecal excretion (400 mg/d)

Figure 1 | Phosphate homeostasis. Phosphate balance by phosphate intake, absorption, storage, and excretion. Used with permission from Hruska KA, Mathew S, Lund R, Qiu P, Pratt R. Hyperphosphatemia of chronic kidney disease. Kidney Int. 2008;74:148–157. Copyright ª 2008 International Society of Nephrology.4

Epidemiological evidence in support of the role of phosphate in clinical events

The assumptions described above are based on numerous observational studies that generally reported on the risk for mortality or cardiovascular disease of higher serum phosphate concentrations across the entire spectrum of CKD, ranging from normal or slightly decreased kidney function to dialysisdependent end-stage kidney disease. Large studies in non-CKD populations, collectively encompassing over 39,000 subjects all found an association between phosphate, even in the normal range, and all-cause mortality,9,10 cardiovascular events,11 or cardiovascular mortality.12 Also in patients with CKD, an association between phosphate and adverse events exists. In an analysis of 1203 patients, Eddington et al. found a stepwise positive association of serum phosphate concentration and mortality in CKD stages 3 and 4 but not stage 5.13 Voormolen et al.,14 however, analyzing patients with stage 5 (nondialysis) CKD did find an increased relative risk for mortality of 1.62 for patients with an average eGFR of 13 (
5.4) ml/min per 1.73 m2 and serum phosphate concentration of 4.71 (
1.16) mg/dl.14 Although an analysis of the Modification of Diet in Renal Disease (MDRD) study (n ¼ 840) could not confirm this association,15 the largest study to date, from the Veterans Affairs Medical Centers (n ¼ 3490) by Kestenbaum et al.16 found a linearly increasing mortality risk for patients with CKD, above a threshold serum phosphate concentration of 3.5 mg/dl (1.13 mmol/l).16 Finally, in patients undergoing dialysis, numerous studies have consistently reported the independent association between hyperphosphatemia and mortality risk.17–23 Phosphate pools or serum phosphate concentration?

Generally, hyperphosphatemia is considered indicative for overall phosphate burden. However, routine clinical Kidney International (2018) 93, 1060–1072

observations demonstrate that this view is an oversimplification, as hyperphosphatemia is widely prevalent in CKD, even in individuals with low bone mass, in which the tissue contains approximately 85% of total body phosphate. In fact, phosphate resides in different compartments, of which plasma and interstitial fluids represent very small fractions (Figure 2).24 Among these compartments, a rather rapid exchangeable pool must exist, which is responsible for the rebound of serum phosphate after dialysis, which is both rapid and high, reaching a 40% increase within 60 minutes after its hemodialysis-induced nadir.25 This phosphate pool of unknown residence, which is responsible for the rebound, can probably be depleted by intensified hemodialysis therapy, especially by extended duration dialysis schedules.26,27 Importantly, the magnitude of this phosphate pool is difficult to estimate, and therefore, no report has been able to demonstrate an association between serum phosphate concentration and the amount of phosphate in this or other phosphate reservoirs. The implication is that phosphate may accumulate in CKD, initially without causing hyperphosphatemia. Remarkable observations for instance from the Framingham offspring cohort, as described above, indicate that serum phosphate even within the normal range is associated with increased risk for cardiovascular events.9,11 This may be explained by the assumption that differences in the amount of stored phosphate in pools is much larger than suggested by corresponding serum concentrations. Based on these observations, authors have speculated about the potential of phosphate toxicity in the absence of overt hyperphosphatemia.24,28 It is possible that serum phosphate concentration is just an unreliable reflection of phosphate stores and that these stores are the true “phosphate culprit” causing cardiovascular disease (Figure 3, left side). This 1061

review

MG Vervloet and AJ van Ballegooijen: Managing phosphate in CKD

84%

14% Organic 1% Inorganic, exchangable Inorganic, non-exchangable

Figure 2 | Phosphate resides in different body compartments. Bone is the largest reservoir with small amounts in plasma and interstitial fluid. Clinically the quantitative amount in each compartment is difficult to estimate except the blood compartment. It is unknown which compartment contributes most to phosphate-related toxicity. Adapted with permission from Osuka S, Razzaque MS. Can features of phosphate toxicity appear in normophosphatemia? J Bone Miner Metab. 2012;30:10–18.24 Copyright ª 2012 The Japanese Society for Bone and Mineral Research and Springer.

hypothesis is supported by the observation in the third National Health and Nutrition Examination Survey (NHANES) cohort that observed that fasted phosphate concentrations, which may better reflect phosphate stores, were associated with mortality, while the higher non-fasted values were not.29 These valid, yet unproven, considerations form the evidencebase to consider clinical interventions in the absence of hyperphosphatemia, as discussed below. Biochemical context of hyperphosphatemia

Phosphate toxicity at a given concentration may differ in severity under different microenvironmental conditions (Figure 3, right side). Changes in systemic and local pH, for instance, modify transmembrane phosphate transport. In

Treatment principle

Working concept

acid–base physiology, it is well recognized that phosphate is a relevant buffer in the extracellular fluid. In this regard, it is noteworthy that trivalent phosphate (PO43 ) does not exist under physiological conditions, nor does the element phosphorus (P) itself. At a pH of 7.4 the predominant phosphate species are HPO42 and H2PO4. Because the pKa at body temperature of this phosphate buffer is 6.8, approximately 80% of total phosphate consists of the bivalent version HPO42–. This is of relevance for phosphate transport across cellular membranes but is frequently neglected. Sodiumphosphate cotransporters (NaPi) are specific for either monovalent or divalent phosphate.30 Phosphate transport into vascular smooth muscle cells (VSMC) across the inorganic sodium phosphate co-transporter NaPi-III (SLC20

Limit phosphate exposure

Restore normophosphatemia

Multitarget approach

Overloaded phosphate pools

Hyperphosphatemia

Complex phosphate interactions

Calcium Klotho deficiency

Inflammation pH

? Endothelial cells

Vascular smooth muscle cells

Figure 3 | Pathways of phosphate toxicity. Conceptual models of phosphate toxicity: as an example, the effects of phosphate on vascular smooth muscle cells are shown. As detailed in the text, exposure to phosphate, even with similar serum concentration, may induce pathology. The therapeutic intervention, indicated on top, would be to limit phosphate exposure (left-sided scenario). Currently, observational clinical studies are consistently demonstrating associations between overt hyperphosphatemia and dismal outcomes. The intervention, therefore, that follows, as suggested for instance by the Kidney Disease: Improving Global Outcomes (KDIGO) guideline, is to treat hyperphosphatemia (center scenario). Phosphate toxicity is importantly modified by other factors, like a-klotho status, pH, calcium concentration, inflammation, and possibly others as well. Targeting phosphate toxicity would be to optimize these factors, besides controlling phosphate itself (right-sided scenario). 1062

Kidney International (2018) 93, 1060–1072

MG Vervloet and AJ van Ballegooijen: Managing phosphate in CKD

subfamily member Pit-1) is considered to be among the earliest triggers in the development of arterial wall calcification.31 Interestingly, Pit-1 transports only monovalent phosphate (H2PO4).30 At a given systemic serum concentration of phosphate, the local tissue pH dictates the availability of this phosphate species to enter VSMC, as the ratio of monovalent over bivalent phosphate changes with pH, meaning that more substrate for Pit-1 becomes available at lower pH. In advanced stages of CKD, acidosis is prevalent,32,33 and the lower pH may directly promote phosphate transport into VSMCs. However, because the systemic pH is generally maintained in the normal range, the local pH of the arterial vessel wall is likely to be more important than systemic acidemia. Hypermetabolism or inflammation in the vessel wall increases local CO2 production,34 comparable to the Bohr effect on hemoglobin. Moreover, VSMCs can regulate local pH at the micro-level and increase acidity in its proximity.35 Besides pH, other conditions also influence local phosphate toxicity. In vitro studies have clearly shown that a-klotho inhibits phosphate entrance into VSMCs in a concentrationdependent manner.36 Therefore phosphate toxicity may be more pronounced in settings of a-klotho deficiency such as CKD. This was recently confirmed in a mouse CKD model with a-klotho deficiency, which demonstrated that restoring a-klotho by delivery through adeno-associated viral vector attenuated aortic calcification.37 In support of this concept, upregulation of endogenous a-klotho by treatment with an activator of peroxisome proliferator-activated receptor-g (PPAR-g), a nuclear factor that upregulates a-klotho, led to an inhibition of VSMC calcification in vitro.38 These studies indicated that phosphate-induced vascular pathology may differ depending on the a-klotho status. A final example of the relevance of the biochemical context in which phosphate may or may not exert toxic effects on the vasculature is the modifying role of calcium.39 Higher concentration of systemic or local calcium (from VSMC apoptosis or matrix vesicles) may act synergistically with phosphate to induce and propagate vascular calcification.40 Some proximal aspects of the process of vascular calcification are dependent on concentration of calcium even at high concentrations of phosphate.41,42 Apart from phenotypic changes induced by entrance of phosphate into VSMSs, spontaneously formed calcium-phosphate crystals, if the concentration of the minerals exceed their saturation product, may induce part of the pathological effects on VSMCs as well.43 Because inhibition of this crystal formation, and not lowering phosphate concentration per se, has protective properties against soft tissue calcification as shown in experimental studies,44 it is conceivable that targeting calcification propensity (a novel circulating biomarker for calcification)45 or specific features of calciprotein particles,46 is a more effective approach than solely focusing on phosphate concentrations. None of the above-mentioned aspects of potential mechanisms of phosphate-related pathology and subsequent Kidney International (2018) 93, 1060–1072

review

morbidity have been thoroughly studied in clinical trials. Despite decades of research dedicated to controlling phosphate overload in CKD, it is disappointing to conclude that an unequivocal proof of benefit of phosphate lowering strategies is still lacking. As outlined previously, the paradigm of exclusively targeting hyperphosphatemia may have to change to target phosphate-induced toxicity, an approach that encompasses more than just restricting phosphate exposure by diet or phosphate binders. With these considerations in mind, a balanced view of potential benefits and harms of current dietary and pharmacological interventions for phosphate exposure in CKD is presented as either a preventive measure or a treatment modality for established hyperphosphatemia. Dietary phosphate restriction in absence of hyperphosphatemia Phosphate balance studies. The first phosphate restriction

studies were conducted in animals decades ago. Multiple animal models (rat, dog, and cat) demonstrated a doseresponse relationship between diets with increasing phosphate content and renal toxicity, as reflected by renal tubular necrosis, nephrocalcinosis, inflammation, and albuminuria, or by the development of hyperparathyroidism.47–51 In subtotally nephrectomized animal models, restriction of phosphate intake can prevent proteinuria, kidney calcification, proximal tubular injury, and premature death.52–55 These studies demonstrated a direct relationship between phosphate load per nephron, kidney calcification, and proximal tubular injury. Human metabolic studies performed decades ago indicate that high intake of phosphate additives could lead to a positive phosphate balance.56,57 It is, however, still unclear what the clinical implication of this positive balance will be, and if ultimately a novel steady state will be reached, for instance, through an increase of FGF23 and PTH synthesis, both of which promote phosphaturia.58 Clearly, more long-term phosphate balance studies are needed to understand phosphate balance at different amounts of intake and different dietary sources of phosphate at various stages of CKD, as outlined below. Human phosphate intake studies and patient outcomes

Observational studies that examined the relationship between phosphate intake and health outcomes showed mixed results, depending on kidney function.59–61 In the general population without kidney disease, phosphate intake $1400 mg/d as estimated based on dietary recall, was prospectively associated with all-cause mortality.59 In a large multiethnic study, higher phosphate intake was associated with greater left ventricular mass.62 In stage 3 CKD patients,60 phosphate intake (based on 24-hour dietary recall) was not associated with premature death, whereas in hemodialysis patients,61 phosphate intake (based on food frequency questionnaire) was strongly associated with premature death. Results from the hemodialysis study were consistent when examining individual 1063

review

foods and food groups of soda and fast food consumption as phosphate sources. Importantly, the correlation between phosphate intake and serum phosphate concentration was very weak (r ¼ 0.1), and no association persisted after multivariate adjustment. A possible explanation for the divergent results between CKD and end-stage renal disease (ESRD) patients may be that CKD patients lowered their total protein-derived energy intake over time. In addition, these patients still possess the possibility of excreting phosphate, thereby preventing the development of hyperphosphatemia and possibly remaining in phosphate balance for a longer period of time.63 For all these intake studies, it is important to note that phosphate-containing food additives were not taken into account, and therefore these results may be an underestimation of true exposure to exogenous phosphate. Few small, human intervention studies investigated the effect of phosphate interventions by dietary modification on vascular health.64,65 A high dietary phosphate load increased serum phosphate concentration after 2 hours and significantly decreased flow-mediated dilation.64 This suggests that postprandial high phosphate concentrations impair endothelial function. Interestingly, in a recent cross-over study, 2 weeks’ exposure to high phosphate intake impaired flow-mediated vasodilation (a clinical estimate of endothelial function) in healthy volunteers but without increasing serum phosphate concentration.65 However, there was no effect on pulse wave velocity. Urinary phosphate excretion

Counter-intuitively, in individuals with stable cardiovascular disease with a mean estimated glomerular filtration rate (eGFR) of 71 ml/min per m2, higher 24-hour urinary phosphate excretion was associated with lower risk of cardiovascular events and with a similarly improved trend for all-cause mortality.66 These associations were attenuated but persisted after adjusting for multiple factors that included age and eGFR. This study also clearly showed the absence of an association between phosphate intake and its serum concentration. In an observational analysis of the Modification of Diet in Renal Disease (MDRD) trial examining the effect of baseline protein intake on CKD progression, phosphate intake as estimated by urinary phosphate excretion was not associated with the risk of premature death or incident ESRD in CKD patients without hyperphosphatemia.67 However, it should be noted that phosphate intake nowadays is much higher than it was 20 years ago, and in this analysis, it was assumed that baseline phosphate intake (prior to randomization and education in the Modification of Diet in Renal Disease [MDRD] trial) remained unchanged for many years.68 In conclusion, these data do not support a relationship between high phosphate intake and adverse outcomes in normophosphatemic patients with or without moderate CKD. Further studies are needed to determine these effects in the setting of more advanced CKD and especially in the setting of hyperphosphatemia. 1064

MG Vervloet and AJ van Ballegooijen: Managing phosphate in CKD

Protein source for phosphate

Human intervention studies focusing on the isolated effect of phosphate intake when keeping protein intake constant are scarce.69 The available evidence is focused on protein restriction, as protein-rich foods are the main sources of dietary phosphate intake. In a meta-analysis of 10 studies with moderate to severe CKD with a study duration of >1 year, lower protein intake reduced the risk of renal replacement therapy or premature death by 32%.70 In addition to the amount of protein, the source of phosphate may be equally important. Most of the studies that compared animal protein to plant-based protein found that plant-based protein was favorable for kidney function preservation and albuminuria.71–73 A balance study compared the effect of phosphate source (meat or vegetarian) in patients with CKD and found lower phosphate and FGF23 concentrations for the vegetarian diet.69 These studies provide insight into the relationship between protein sources and CKD progression; however, it remains unclear if the effect of plant-based protein is due to differences in amino acid, dietary acid load, or phosphate availability itself. Intake of phosphate additives

Some human studies examined the role of phosphate additives on outcomes (Table 1). In healthy premenopausal women, the intake of phosphate-containing food additives from processed cheese was cross-sectionally associated with higher serum PTH concentrations.74 However, in the Multi-Ethnic Study of Atherosclerosis (MESA) study, a large multiethnic cohort, no association was found between processed food intake and serum phosphate concentrations.75 In another cross-sectional study in a middle-aged population in Finland, a significant relationship was observed between higher total phosphate intake and food additive phosphate intake with higher carotid intima-media thickness.76 Some short-term human experiments have been conducted as well. A small cross-over diet study in young individuals with normal kidney function showed that high phosphate additive intake resulted in elevated concentrations of FGF23, osteopontin, and osteocalcin.58 In CKD patients, a 3-week cross-over trial comparing commercially available products with or without phosphate additives showed a trend toward increased albuminuria but was not significant.77 Collectively these studies demonstrate that a diet high in phosphate additives induces changes in hormonal systems involved in phosphate metabolism and intermediate endpoints. Importantly, the mixed results regarding the possible influence of serum phosphate concentration on outcomes suggest that this laboratory value does not reflect exposure to phosphate-containing additives. More research is needed to study long-term effects of these additives on patient-level endpoints, in particular in CKD populations at highest risk for phosphate toxicity.

Kidney International (2018) 93, 1060–1072

review

MG Vervloet and AJ van Ballegooijen: Managing phosphate in CKD

Table 1 | Phosphate additives and health outcomes Study, year

Study design

Country

Study population

Exposure

Outcome

Observational Kemi et al.,74 2009

Cross-sectional

Finland

Intake phosphate-containing food additives from processed cheese

High cheese consumption was associated with higher serum PTH: b 36.5 ng/l

Cross-sectional

USA

Cross-sectional

Finland

N ¼ 147 healthy premenopausal women aged 31–43 years 2664 MESA participants, aged 62 years N ¼ 546 middle-aged, no CKD, aged 37–47 years

Processed food intake (processed meat, soda, fast food) Intake of food additive phosphate

Not associated with serum phosphate Higher carotid intima-media thickness
40 mm P<0.05 highest versus lowest group

1000 mg of phosphate per day with foods free of phosphorus additives, followed by a diet containing identical food items; Higher versus lower phosphate intake by the addition of commercially available diet beverages and breakfast bars

FGF23þ 23% Osteopontin þ10% Osteocalcin þ 11%

Gutiérrez et al., 2012 Itkonen et al., 2013

Interventional Absence of hyperphosphatemia Gutiérrez et al., 2 week feeding 2015 trial

Chang and Grams,29 2017

3 week cross-over trial

Presence of hyperphosphatemia 3 months RCT Sullivan et al.,80 2009

De Fornasari et al., 2017

3 month RCT

USA

10 healthy individuals, normal kidney function

USA

N ¼ 31 adults with mean eGFR 45 ml/min per 1.73 m2

USA

279 ESRD patients with baseline phosphate >5.5 mg/dl

Brazil

N ¼ 34 ESRD patients with baseline phosphate >5.5 mg/dl

- Intervention (n ¼ 145) received education on avoiding foods with phosphate additives when purchasing groceries or visiting fast food restaurants. - Control participants (n ¼ 134) usual care. Intervention group (n ¼ 67): Individual orientation to replace processed foods high in phosphate additives with foods of similar nutritional value without these additives. Control group (n ¼ 67) received nutritional orientation without replacing foods

Albuminuria: þ14.3% FGF23: þ3.4% but not significant

0.6 (95% CI, 1.0 to 0.1) mg/dl phosphate favoring intervention Also intervention significantly increased in reading ingredient lists and nutrition facts labels Intervention 2.2 mg/dl lower phosphate versus 0.4 control group Intervention 70% reached the phosphate target of #5.5 mg/dl versus 18.5% in control group

CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; FGF23, fibroblast growth factor-23; MESA, Multi-Ethnic Study of Atherosclerosis study; PTH, parathyroid hormone; RCT, randomized control trial.

Diet therapy in the presence of hyperphosphatemia

The recent Kidney Disease: Improving Global Outcomes (KDIGO) guideline on the CKD-associated bone and mineral disorder (CKD-MBD) suggests in patients with CKD stages 3 to 5 that dietary phosphate intake be limited for the treatment of hyperphosphatemia, combined with other treatments.78 Few studies have investigated the effect on hyperphosphatemia of diet and phosphate additive replacement.79–81 In a 6-month trial among dialysis patients in Spain, which compared intensive dietary intervention with usual dietary recommendations, phosphate intake was significantly lower in the experimental group.81 The achieved serum phosphate concentration was 0.93 mg/dl (0.3 mmol/l) lower in the experimental group, and the prevalence of hyperphosphatemia (>5.5 mg/dl; 1.8 mmol/l) declined markedly more than in the control group (49% vs. 82%, respectively). Two randomized trials have investigated the effect of replacing foods high in phosphate additives with food without additives and found significant reductions in serum Kidney International (2018) 93, 1060–1072

phosphate concentrations and prevalence of hyperphosphatemia.79,80 In ESRD patients, education to avoid phosphate additive-rich foods, compared to usual diets, also resulted in modest improvements in hyperphosphatemia.80 Moreover, the intervention group participants read ingredient lists and nutrition fact labels more often, although they failed to achieve better food knowledge. Another recent trial indicated that replacing phosphatecontaining food additives with food without additives impressively reduced hyperphosphatemia in ESRD patients from an average of 7.2 mg/dl (2.3 mmol/l) to 5.0 mg/dl (1.6 mmol/l).79 Long-term studies of the association between high phosphate intake, especially in the form of highly absorbable food additive phosphate and health outcomes are needed. Treatment in CKD populations

The KDIGO guideline for CKD-MBD recommend that, in patients with CKD stage 5, phosphate intake should not exceed 1000 mg per day.78 Although this suggestion is made in the guideline, it is mainly based on expert opinion. 1065

review

Lowering phosphate intake in a diet is challenging. It requires intensive patient education to understand the complexity of dietary labels and avoid malnutrition.82 A systematic review among CKD patients indicates an average reduction of 0.72 mg/dl serum phosphate after any educational intervention, whereas the reduction was more pronounced in interventions for $4 months.83,84 This suggests that appropriate dietary counseling can help to keep phosphate concentrations within the recommended range and can assist in management and treatment strategies for CKD. Food composition tables usually do not take into account food additives.85 Because phosphate additive intake varies between 10% and 50% of total intake, specific attention in dietary counseling should be paid to foods high in phosphate additives. Phosphate binder therapy is expected to remove on average no more than approximately 200 to 300 mg of phosphate per day,86 which motivates the potential of dietary counseling to reduce phosphate intake by a greater amount. Growing evidence shows that a reduction in ultraprocessed foods with a high content of phosphate additives is related to better phosphate management79,80 and health outcomes.87 Control of dietary phosphate requires a specific set of skills and knowledge by trained dieticians. It should be tailored to each individual’s preference, family situation, availability of foods, purchase, and food preparation to efficiently integrate dietary phosphate control into daily life of CKD management. How to reduce phosphate intake?

Almost all foods contain phosphate, but greater amounts are found in animal products such as meat, fish, eggs, and dairy. Nuts, seeds, and many vegetables are also rich in phosphate. A regular Western diet provides between 1000 and 1500 mg/d.88 Most phosphate is absorbed in the proximal intestine, predominantly in the jejunum. The absorbed phosphate enters the extracellular fluid pool and moves into and out of bone tissue. In addition, phosphate is commonly used as an additive for preservation or thickening of foods. For example, 1 can of cola contains 65 mg of phosphate, equivalent to 20 grams of cooked chicken filet,89 which contributes to a high phosphate load. Furthermore, products like enhanced meat, flavored water, and frozen meals with high phosphate content fill our grocery shelves rapidly. Also, the bioavailability of phosphate additives (industrial phosphate; e.g., phosphoric acid, polyphosphates) is higher than phosphate from natural sources. The bioavailability of phosphate from natural sources is approximately 60% to 80%, with lowest bioavailability for legumes at 40%, whereas soda and salad dressings approach 100% bioavailability.90,91 Phosphate additives contribute 10% to 50% of total phosphate intake in Western diets.92 Furthermore, net phosphate absorption varies depending on overall intake, food source, and matrix, the relative amounts of dietary calcium, phosphate, and vitamin D sterols, and bile acid concentration,69,93 thus a large variety of factors determine daily phosphate load. 1066

MG Vervloet and AJ van Ballegooijen: Managing phosphate in CKD

The food preparation technique also influences phosphate content. Boiling reduces phosphate content due to demineralization of food, as the minerals move from the food into the boiling water.94 Reported phosphate reductions by boiling vary between 35% and 50%.95 All these details can be useful for dietary counseling aimed to lower the phosphate load from the diet. Inorganic phosphate is also added to medications and is another, generally unrecognized source of phosphate exposure. In a Canadian hemodialysis population, 11% of prescribed medication contained a phosphate salt with a median phosphate burden of 111 mg per day.96 With a median daily pill burden of 19, prescription medication can substantially contribute to the daily phosphate load in dialysis patients.97 Moreover, phosphate is usually added to multivitamin supplements with estimations between 20 and 150 mg per supplement. The use of a multivitamin supplement will further contribute to a higher dietary phosphate load. Recommendations for future research 





Perform phosphate balance studies keeping protein intake constant at different stages of CKD, initially to test feasibility followed by studies on clinical outcomes. Update phosphate additives content in food composition database. Conduct randomized controlled dietary trial interventions among CKD patients to inform evidence-based recommendations. These studies should have sufficient follow-up and clinically relevant endpoints (Table 1).

Pharmacological interventions to prevent hyperphosphatemia

It is difficult to establish when an intervention should be considered as a "preventive" or as a "therapeutic" measure, as hyperphosphatemia is not a disease. Moreover, serum phosphate exhibits circadian fluctuations not only physiologically but also in patients with CKD. For the remainder of this review, interventions for phosphate homeostasis with normal or slightly increased serum phosphate concentrations will be considered as preventive and in the setting of overt hyperphosphatemia as therapeutic. Clinically, the decision to apply preventive measures is essentially confined to predialysis CKD.1 Change in the KDIGO guideline on CKD-MBD

One of the key changes of the recently updated KDIGO guideline for CKD-MBD (guideline 4.1.2) is that it no longer suggests maintaining phosphate concentrations within the normal range.78 This change was largely based on a trial by Block et al.,98 which showed a discrepancy between the effect of phosphate binders on serum phosphate concentration, which indeed was slightly lowered, and intermediate endpoints (serum FGF23 and coronary artery calcification [CAC]) which were unchanged or even worsened.98 In that study, different phosphate binders were used, but the sample size of subgroups was too small to evaluate differences Kidney International (2018) 93, 1060–1072

MG Vervloet and AJ van Ballegooijen: Managing phosphate in CKD

between individual phosphate binders.99 The baseline phosphate concentration in that study was 1.36 mmol/l (4.2 mg/ dl); therefore that study could be considered a proof of concept trial of pharmacotherapy to prevent the development of hyperphosphatemia. Because that trial fulfilled predefined criteria to be included in the KDIGO update and, importantly also included a placebo, its impact on this section of the guideline was substantial. The question arises whether that study disqualifies future attempts to prevent phosphateinduced pathology in the absence of hyperphosphatemia. The 3 conceptual models of phosphate toxicity (Figure 2), consisting of concentrations of phosphate, the phosphate pool, and phosphate in a complex system, could be examined separately. Interestingly, in the trial by Block et al.,98 all binders used (calcium-containing binders, lanthanum, and sevelamer) had comparable effect on lowering phosphate concentration and 24-hour urine phosphate excretion. However, phosphate concentrations were measured under nonfasted conditions, and the impact of the different interventions on the total phosphate pool is difficult to estimate. The 24-hour phosphate excretion, which declined for all phosphate binders, is probably a better reflection of gastrointestinal uptake during steady state than a parameter of total phosphate content. Are calcium-based phosphate binders unsafe in predialysis CKD?

The decline of 24-hour urinary phosphate excretion in the calcium-containing binder group in the above-mentioned study by Block et al.98 contradicts findings from a meticulously performed balance study in CKD patients with a mean eGFR of 36 ml/min per m2.63 In that short-term study, no change in excretion of phosphate was accomplished in response to 1.5 g of calcium carbonate per day; however, in the study by Block et al.98, patients with similar degrees of CKD received 5.9 g of calcium acetate per day. The progression of CAC in the calcium carbonate group in the trial by Block et al.98 has been suggested to be the consequence of calcium loading.100 Notably, the noncalcium-containing phosphate binders did not lower CAC progression compared to placebo, and the statistical significance of the differences between individual treatment subgroups could not be evaluated due to insufficient sample size.99 In addition, patients without coronary calcification at baseline did not develop calcification regardless of treatment allocation.98 Different findings emerged from the head-to-head trial by Russo et al.101, where the use of calcium-containing binders (2 g of calcium carbonate daily) in CKD patients with baseline mild hyperphosphatemia (4.5 mg/dl, 1.45 mmol/l) did not induce more progression of CAC scores than no binder use at all. Furthermore, sevelamer administration, studied in the same trial, did significantly attenuate calcification progression. Both binders induced a 16% reduction in 24-hour urine phosphate excretion but had no effect on serum phosphate concentration. Importantly, this study also noted that, in the absence of baseline vascular calcification, no Kidney International (2018) 93, 1060–1072

review

calcification developed after 24 months in either treatment arm. The differences between the 2 studies may be ascribed to differences in phosphate exposure as measured by 24-hour urine excretion (higher in the trial by Block et al.98), differences in baseline serum phosphate concentrations (higher in the trial by Russo et al.101), and differences in population size and follow-up duration. In the open-label trial by Di Iorio et al.,102 baseline serum phosphate was slightly higher (4.8 mg/dl per 1.6 mmol/l) than in the 2 studies addressed previously.102 In this study, sevelamer was compared to calcium carbonate and improved survival after 3 years of follow-up. Remarkably, phosphate concentration improved only in the sevelamer group, complicating the interpretation of this trial, which could imply either an advantage of sevelamer over calcium carbonate or unsuccessful phosphate control in the calcium carbonate group. Another remarkable finding was that adjusting for time-varying covariates, which included serum levels of phosphate, C-reactive protein, and cholesterol, did not mitigate the beneficial hazard ratio for sevelamer group. This is surprising, because this argues against improved phosphate concentration (and cholesterol and C-reactive protein) as a mediating factor of the benefit of sevelamer. Unfortunately, 24-hour phosphate excretion was not measured in this trial. When trying to reconcile these data in normophosphatemic CKD, there is no proven benefit to patient-level outcomes of any pharmacological intervention in phosphate homeostasis, at least not for the currently available treatment options. However, some data, as discussed above, do suggest that, with incremental baseline phosphate concentrations beyond the upper limit of normal, benefits of intervention may outweigh the possible harm inflicted by phosphate binders. Adverse effects must be taken into account, which could include more rapid progression of established vascular calcification for calcium-containing binders, a risk that appears to be absent if there is no baseline calcification at all. Furthermore, all phosphate binders have the capacity to bind vitamin K (at least in vitro), which is a key factor in calcification defense.103 Therefore, it may be imprudent to consider noncalciumcontaining as completely safe. This should draw attention to the possibility that phosphate toxicity is dependent on the microenvironment, as outlined in the previous introductory paragraphs. This principle is generally neglected in clinical medicine, which focuses on phosphate concentrations only. Pharmacological interventions to treat hyperphosphatemia

Pharmacological treatment of hyperphosphatemia is widely used, especially in patients undergoing dialysis and contributes largely to the total pill burden for these patients.104 Costs related to phosphate binders have been estimated at $750 million US dollars globally.105 Despite the treatment burden for patient and economical impact for society, definite proof of benefit at the patient-level for this intervention is lacking, because this has never been studied appropriately. Importantly, total number of pills in and amount of phosphate 1067

review

binders prescribed in particular are associated with nonadherence.106

MG Vervloet and AJ van Ballegooijen: Managing phosphate in CKD

100

The aim of controlling hyperphosphatemia is to lower associated risks for hard outcomes, particularly in mortality and cardiovascular disease. Initially, this was based on meticulous observations of large cohorts of dialysis patients who demonstrated independent associations between both cross-sectional and time-varying concentrations of phosphate with survival, as described previously.16,17,23,107 Growing epidemiological evidence supports treatment of hyperphosphatemia because a decline in phosphate concentration is also associated with reduced mortality.108 The most recent key finding in support of targeting hyperphosphatemia is that observational data have demonstrated a survival benefit for use of phosphate binders versus not using binders.109–111 Obviously phosphate binders are prescribed more frequently to well-nourished patients, and indeed when adjusting for nutritional status, the relationship was attenuated, but the association between phosphate binders use and survival benefit persisted.109 Probably the most compelling epidemiological support in favor of using phosphate binders for hyperphosphatemia in dialysis patients comes from the Accelerated Mortality on Renal Replacement (ArMORR) cohort, which applied propensity score matching for phosphate binder prescription (Figure 4).111 After 1 year, treatment with phosphate binders was associated with a significant survival advantage compared with no treatment in incident hemodialysis patients, except for those with a baseline serum phosphate concentration below 3.7 mg/dl (1.2 mmol/l). Despite all epidemiological data supporting use of phosphate binder for hyperphosphatemia it must be kept in mind that observational studies can never provide the level of evidence of a properly conducted prospective randomized trial. Available pharmacological treatments and how to select

Phosphate binders are labeled for treatment of hyperphosphatemia, but calcimimetics also lower phosphate concentrations, at least in the setting of secondary hyperparathyroidism. In that condition, a large proportion of circulating phosphate may be derived from bone instead of from dietary sources.112–114 Intestinal phosphate uptake is accomplished by either a concentration or electrochemical gradient driving paracellular transport or active transcellular transport across ion channels. Dietary phosphate restriction and phosphate binders lower the intestinal luminal free phosphate concentration. Animals studies revealed that, in the setting of low intestinal phosphate concentrations, an upregulation of NaPi2b occurred to maintain phosphate uptake by active transcellular transport.115 This also occurs following treatment with active vitamin D compounds.116 Subsequent animal studies confirmed that inhibiting this transporter prevented the development of hyperphosphatemia.117 Also, clinically the gastrointestinal phosphate transporter NaPi2b inhibitors nicotinamide118,119 and niacin120,121 are promising treatments. However, these compounds are currently not on 1068

Survival (%)

95

What is the evidence to treat hyperphosphatemia in CKD?

90 85 80 75 0

90

180 Days

Phosphorus binder

270

365

No phosphorus binder

Figure 4 | Survival of treated and untreated patients in an overall propensity score–matched cohort of hemodialysis. Data from the prospective observational ArMORR study compared treated (n ¼ 3186) and untreated (n ¼ 3186) patients matched by their baseline serum phosphate levels and propensity score of receiving phosphorus binders during the first 90 days. After 1 year, treatment with phosphorus binders was associated with a significant survival advantage compared with no treatment in incident hemodialysis patients. Adapted with permission from Isakova T, Gutierrez OM, Chang Y, et al. Phosphorus binders and survival on hemodialysis. J Am Soc Nephrol. 2009;20:388–396.111 Copyright ª American Society of Nephrology. ArMORR, Accelerated Mortality on Renal Replacement.

the market, have not been studied in endpoints other than biochemical control of serum phosphate concentration, still face safety issues, and are therefore beyond the scope of this review. The agent tenapanor (Ardelyx, Fremont, CA) targets intestinal sodium–hydrogen exchanger and also limits absorption of phosphate. A recent clinical trial confirmed its efficacy in controlling serum phosphate concentrations.122 Currently, phosphate binders are the mainstay of orally acting phosphate-lowering treatment. It should be emphasized that all phosphate binders effectively lower phosphate.104 Therefore, treatment choices are generally based on assumed effects on intermediate endpoints like progression of vascular calcification and patient-reported tolerance, which includes gastrointestinal complaints and pill burden. For a long time, the assumed risk of calcium content of some binders (calcium carbonate and calcium acetate) has dominated this discussion. In the recently updated KDIGO guideline on CKD-MBD, there is now a more general suggestion to limit calciumcontaining binders in all patients,78 whereas the previous guideline specifically mentioned subpopulations in which to restrict calcium-containing in its phosphate binder use, like patients with known vascular calcification.123 This change was essentially based on the INDEPENDENT (Reduce Cardiovascular Calcifications to Reduce QT Interval in Dialysis) trial in hemodialysis patients.124 In that trial, incident hemodialysis patients with a mean 65 years of age were randomized to either calcium carbonate or sevelamer and were found to have a reduction in cardiovascular mortality when allocated to the sevelamer group. The study however had some drawbacks. The patients were relatively young, were not stratified for Kidney International (2018) 93, 1060–1072

review

MG Vervloet and AJ van Ballegooijen: Managing phosphate in CKD

baseline CAC scores, and phosphate control was better for those treated with sevelamer. The sevelamer arm had a remarkably favorable cardiovascular survival of 94% after 3 years’ follow-up (compared to 60% in the calcium carbonate group), which suggests that a specific population was recruited, and hence external validity might have been limited. At baseline, those with increased QT intervals were excluded, whereas especially patients with QT elongation may benefit from higher calcium concentrations.125,126 Two recent meta-analyses that compared calciumcontaining binders with noncalcium-containing binders, usually sevelamer, observed a survival advantage for noncalciumcontaining binders.127,128 These meta-analyses, as acknowledged by the authors, cannot improve the quality of the original studies on which the pooled estimates are based. Several of these studies, including the largest Dialysis Clinical Outcomes Revisited (DCOR) trial129 used higher than recommended doses of calcium containing binders and suffered from methodical issues like high drop-out rates. The DCOR trial was negative in regards to its primary endpoint, all-cause mortality after 24 months. However, the study did suggest reduced mortality for those in study longer than 24 months and those older than 65 years of age. Remarkably the suggested differences in mortality were driven by non-cardiovascular causes. From the perspective of phosphate as crucial component in a complex microenvironment (Figure 3), the Calcium Acetate Renagel Evaluation-2 (CARE-2) trial is of renewed interest. In that trial, hemodialysis patients with established vascular calcification at baseline were randomized to either calcium acetate of sevelamer, with statin therapy in both groups if needed to maintain LDL-cholesterol levels below 70 mg/dl.130 No differences were found in the progression of CAC after 1 year. This study suggests that calcification progression associated with the use of calcium-containing binders is not due to calcium loading but to a lack of control of LDL-cholesterol compared to sevelamer. This interpretation has been criticized because the baseline risk for progressive vascular calcification in the CARE-2 was exceptionally high due to comorbid conditions like high prevalence of smoking and diabetic nephropathy.131 In that setting phosphate binder selection may have no detectable impact on development of vascular calcification. However, this does not undermine the concept that the setting in which hyperphosphatemia occurs is of importance; in the CARE-2 study there might have been overrepresentation of diabetic kidney disease and smoking as accelerators of phosphate toxicity. Taken together, several lines of evidence advocate against the use of calcium containing binders. However, when used in restricted dose, and adequate phosphate control is achieved, which was recently shown to be feasible132 these cheap binders may still have a place in the treatment of hyperphosphatemia. Treatment target

Following the decision to treat hyperphosphatemia and selecting the way to do so, clinicians face the question what the treatment target should be. Unfortunately there are Kidney International (2018) 93, 1060–1072

virtually no data or guidelines that provide guidance.78 The previously discussed analysis of the Current Management of Secondary Hyperparathyroidism: A Multicenter Observational Study (COSMOS) demonstrated a benefit of a mean phosphate decline of 1 mg/dl (0.32 mmol/l) for those having a baseline value above 5.2 mg/dl (1.7 mmol/l).108 However, whether those with higher baseline values benefit from greater reductions of phosphate concentrations, a clinically relevant issue, was not studied. A recent pilot study investigated the effect of an intensive phosphate goal versus a more liberalized phosphate target in hemodialysis patients in Canada for 26 weeks. Serum phosphate concentration in the intensive group declined by 0.40 mmol/l compared with the liberalized group. This study showed that it is feasible to achieve and maintain a difference in serum phosphate concentrations by titrating calcium carbonate. Future studies, that may follow-up on this pilot study, might fill the knowledge gap of optimal phosphate targets.132 Recommendations for future research

As mentioned by many authors, the ideal trial would be to compare the effect on hard outcomes of a phosphate-lowering intervention with placebo in the setting of overt hyperphosphatemia. Given the compelling experimental and consistent epidemiological evidence pointing to a causal role of hyperphosphatemia in the occurrence of mainly cardiovascular disease and mortality, the reluctance to execute such a trial is understandable. Nevertheless, some studies may move the field forward.  Investigate the impact of different phosphate targets on patient-level outcomes, possibly combined with the use of different phosphate lowering strategies, with sufficient sample size and follow-up duration. Such a trial could identify the optimal treatment target and the optimal approach to achieve that.  Study the effect of targeting fasting serum phosphate concentration, as a reflection of phosphate pools.  Investigate the differences in outcome using multifactorial interventions addressing phosphate-toxicity modifying factors, compared to phosphate concentration only. Conclusion

The current management paradigm for treating phosphateinduced toxicity is now focused on overt hyperphosphatemia. Both dietary interventions and phosphate binder therapy are effective in lowering serum phosphate concentrations, urinary phosphate excretion, or both. Presently there is no definite proof of a beneficial effect of phosphate lowering on patient-level outcome. Moreover both dietary intervention and phosphate binder therapy may have side effects. Despite these limitations, treating hyperphosphatemia in CKD appears still appropriate but should be paralleled by ongoing research to further underpin this approach and improve therapeutic strategies. Currently, there is insufficient evidence to actively prevent the development of hyperphosphatemia by either dietary or pharmacological intervention, with a possible exception for restricting phosphate1069

review

containing additives in patients with CKD. Possible novel directions in addressing phosphate-associated pathology are to consider both phosphate concentrations and phosphate loading simultaneously. Phosphate loading is difficult to estimate clinically, but measuring phosphate concentration in the fasting state, when equilibration with phosphate pools has occurred, may be a step forward. Finally, phosphate as a component of different factors that modify its potential toxic effect may lead to an integrated approach to phosphate-induced comorbidity that is not only focused on its serum concentration. The novel paradigm for treating phosphate-induced comorbidity may be a multifactorial approach that not only targets phosphate concentrations but also consists of maintaining or increasing a-klotho, possibly lowering FGF23, and optimizing co-existing metabolic derangements such as acidemia, overt hyperparathyroidism, hypercalcemia, diabetes, and inflammation. DISCLOSURES

MGV has received research grants from AbbVie, Amgen, FMC, Sanofi, and Dutch Kidney Foundation; has received speakers fees from Baxter, Amgen, BBraun, and VFMCRP; and has consulted for Amgen, Medice, VFMCRP, Otsuka, and Astellas. MGV is member of ERA-EDTA working group on CKD-MBD and KDIGO working group on CKD-MBD. The current narrative review is not written on behalf of these working groups. The other author has declared no competing interests.

ACKNOWLEDGMENTS

AJvB is supported by a Dr. Kolff junior postdoctoral grant from the Dutch Kidney Foundation (16OKG02). REFERENCES 1. Levin A, Bakris GL, Molitch M, et al. Prevalence of abnormal serum vitamin D, PTH, calcium, and phosphorus in patients with chronic kidney disease: results of the study to evaluate early kidney disease. Kidney Int. 2007;71:31–38. 2. Isakova T, Wolf MS. FGF23 or PTH: which comes first in CKD? Kidney Int. 2010;78:947–949. 3. Marks J, Debnam ES, Unwin RJ. Phosphate homeostasis and the renalgastrointestinal axis. Am J Physiol Renal Physiol. 2010;299:F285–296. 4. Hruska KA, Mathew S, Lund R, Qiu P, Pratt R. Hyperphosphatemia of chronic kidney disease. Kidney Int. 2008;74:148–157. 5. Alizadeh Naderi AS, Reilly RF. Hereditary disorders of renal phosphate wasting. Nat Rev Nephrol. 2010;6:657–665. 6. Six I, Maizel J, Barreto FC, et al. Effects of phosphate on vascular function under normal conditions and influence of the uraemic state. Cardiovasc Res. 2012;96:130–139. 7. Vervloet MG, Sezer S, Massy ZA, et al. The role of phosphate in kidney disease. Nat Rev Nephrol. 2016;13:27–38. 8. Drueke TB, Locatelli F, Clyne N, et al. Normalization of hemoglobin level in patients with chronic kidney disease and anemia. N Engl J Med. 2006;355:2071–2084. 9. Tonelli M, Sacks F, Pfeffer M, et al. Relation between serum phosphate level and cardiovascular event rate in people with coronary disease. Circulation. 2005;112:2627–2633. 10. Foley RN, Collins AJ, Ishani A, Kalra PA. Calcium-phosphate levels and cardiovascular disease in community-dwelling adults: the Atherosclerosis Risk in Communities (ARIC) study. Am Heart J. 2008;156: 556–563. 11. Dhingra R, Sullivan LM, Fox CS, et al. Relations of serum phosphorus and calcium levels to the incidence of cardiovascular disease in the community. Arch Intern Med. 2007;167:879–885. 12. Vart P, Nigatu YT, Jaglan A, van Zon SK, Shafique K. Joint effect of hypertension and elevated serum phosphorus on the risk of mortality in national health and nutrition examination survey-III. J Am Heart Assoc. 2015;4:e001706. https://doi.org/10.1161/JAHA.114.001706.

1070

MG Vervloet and AJ van Ballegooijen: Managing phosphate in CKD

13. Eddington H, Hoefield R, Sinha S, et al. Serum phosphate and mortality in patients with chronic kidney disease. Clin J Am Soc Nephrol. 2010;5: 2251–2257. 14. Voormolen N, Noordzij M, Grootendorst DC, et al. High plasma phosphate as a risk factor for decline in renal function and mortality in predialysis patients. Nephrology Transplant. 2007;22:2909–2916. 15. Menon V, Greene T, Pereira AA, et al. Relationship of phosphorus and calcium-phosphorus product with mortality in CKD. Am J Kidney Dis. 2005;46:455–463. 16. Kestenbaum B, Sampson JN, Rudser KD, et al. Serum phosphate levels and mortality risk among people with chronic kidney disease. J Am Soc Nephrol. 2005;16:520–528. 17. Block GA, Hulbert-Shearon TE, Levin NW, Port FK. Association of serum phosphorus and calcium x phosphate product with mortality risk in chronic hemodialysis patients: a national study. Am J Kidney Dis. 1998;31:607–617. 18. Slinin Y, Foley RN, Collins AJ. Calcium, phosphorus, parathyroid hormone, and cardiovascular disease in hemodialysis patients: the USRDS waves 1, 3, and 4 study. J Am Soc Nephrol. 2005;16:1788–1793. 19. Naves-Diaz M, Passlick-Deetjen J, Guinsburg A, et al. Calcium, phosphorus, PTH and death rates in a large sample of dialysis patients from Latin America. The CORES study. Nephrology Transplant. 2011;26: 1938–1947. 20. Young EW, Albert JM, Satayathum S, et al. Predictors and consequences of altered mineral metabolism: the Dialysis Outcomes and Practice Patterns Study. Kidney Int. 2005;67:1179–1187. 21. Ganesh SK, Stack AG, Levin NW, Hulbert-Shearon T, Port FK. Association of elevated serum PO(4), Ca x PO(4) product, and parathyroid hormone with cardiac mortality risk in chronic hemodialysis patients. J Am Soc Nephrol. 2001;12:2131–2138. 22. Floege J, Kim J, Ireland E, et al. Serum iPTH, calcium and phosphate, and the risk of mortality in a European haemodialysis population. Nephrology Transplant. 2011;26:1948–1955. 23. Kalantar–Zadeh K, Kuwae N, Regidor DL, et al. Survival predictability of time-varying indicators of bone disease in maintenance hemodialysis patients. Kidney Int. 2006;70:771–780. 24. Osuka S, Razzaque MS. Can features of phosphate toxicity appear in normophosphatemia? J Bone Miner Metab. 2012;30:10–18. 25. Agar BU, Akonur A, Lo YC, Cheung AK, Leypoldt JK. Kinetic model of phosphorus mobilization during and after short and conventional hemodialysis. Clin J Am Soc Nephrol. 2011;6:2854–2860. 26. Pierratos A, Ouwendyk M, Francoeur R, et al. Nocturnal hemodialysis: three-year experience. J Am Soc Nephrol. 1998;9:859–868. 27. Cornelis T, van der Sande FM, Eloot S, et al. Acute hemodynamic response and uremic toxin removal in conventional and extended hemodialysis and hemodiafiltration: a randomized crossover study. Am J Kidney Dis. 2014;64:247–256. 28. Kuro OM. A phosphate-centric paradigm for pathophysiology and therapy of chronic kidney disease (2011). Kidney Int Suppl. 2013;3:420–426. 29. Chang AR, Grams ME. Serum phosphorus and mortality in the Third National Health and Nutrition Examination Survey (NHANES III): effect modification by fasting. Am J Kidney Dis. 2014;64:567–573. 30. Virkki LV, Biber J, Murer H, Forster IC. Phosphate transporters: a tale of two solute carrier families. Am J Physiol Renal Physiol. 2007;293. F643–564. 31. Li X, Yang HY, Giachelli CM. Role of the sodium-dependent phosphate cotransporter, Pit-1, in vascular smooth muscle cell calcification. Circ Res. 2006;98:905–912. 32. Kraut JA, Madias NE. Metabolic Acidosis of CKD: An Update. Am J Kidney Dis. 2016;67:307–317. 33. Kovesdy CP. Metabolic acidosis and kidney disease: does bicarbonate therapy slow the progression of CKD? Nephrology Transplant. 2012;27: 3056–3062. 34. Bernelot Moens SJ, Verweij SL, van der Valk FM, et al. Arterial and Cellular Inflammation in Patients with CKD. J Am Soc Nephrol. 2016. 35. Hordijk PL. Intracellular pH steers vascular wall remodelling. Cardiovasc Res. 2016;111:165–166. 36. Hu MC, Shi M, Zhang J, et al. Klotho deficiency causes vascular calcification in chronic kidney disease. J Am Soc Nephrol. 2011;22:124–136. 37. Hum JM, O’Bryan LM, Tatiparthi AK, et al. Chronic hyperphosphatemia and vascular calcification are reduced by stable delivery of soluble klotho. J Am Soc Nephrol. 2017;28:1162–1174. 38. Cheng L, Zhang L, Yang J, Hao L. Activation of peroxisome proliferatoractivated receptor gamma inhibits vascular calcification by upregulating Klotho. Exp Ther Med. 2017;13:467–474. Kidney International (2018) 93, 1060–1072

MG Vervloet and AJ van Ballegooijen: Managing phosphate in CKD

39. Lomashvili K, Garg P, O’Neill WC. Chemical and hormonal determinants of vascular calcification in vitro. Kidney Int. 2006;69:1464–1470. 40. Shanahan CM, Crouthamel MH, Kapustin A, Giachelli CM. Arterial calcification in chronic kidney disease: key roles for calcium and phosphate. Circ Res. 2011;109:697–711. 41. Reynolds JL, Joannides AJ, Skepper JN, et al. Human vascular smooth muscle cells undergo vesicle-mediated calcification in response to changes in extracellular calcium and phosphate concentrations: a potential mechanism for accelerated vascular calcification in ESRD. J Am Soc Nephrol. 2004;15:2857–2867. 42. Shroff RC, McNair R, Skepper JN, et al. Chronic mineral dysregulation promotes vascular smooth muscle cell adaptation and extracellular matrix calcification. J Am Soc Nephrol. 2010;21:103–112. 43. Sage AP, Lu J, Tintut Y, Demer LL. Hyperphosphatemia-induced nanocrystals upregulate the expression of bone morphogenetic protein-2 and osteopontin genes in mouse smooth muscle cells in vitro. Kidney Int. 2011;79:414–422. 44. Villa-Bellosta R, Sorribas V. Phosphonoformic acid prevents vascular smooth muscle cell calcification by inhibiting calcium-phosphate deposition. Arterioscler Thromb Vasc Biol. 2009;29:761–766. 45. Smith ER, Ford ML, Tomlinson LA, et al. Serum calcification propensity predicts all–cause mortality in predialysis CKD. J Am Soc Nephrol. 2014;25:339–348. 46. Smith ER, Ford ML, Tomlinson LA, Rajkumar C, McMahon LP, Holt SG. Phosphorylated fetuin-A-containing calciprotein particles are associated with aortic stiffness and a procalcific milieu in patients with predialysis CKD. Nephrology Transplant. 2012;27:1957–1966. 47. Haut LL, Alfrey AC, Guggenheim S, Buddington B, Schrier N. Renal toxicity of phosphate in rats. Kidney Int. 1980;17:722–731. 48. Brown SA, Crowell WA, Barsanti JA, White JV, Finco DR. Beneficial effects of dietary mineral restriction in dogs with marked reduction of functional renal mass. J Am Soc Nephrol. 1991;1:1169–1179. 49. Finco DR, Brown SA, Crowell WA, Groves CA, Duncan JR, Barsanti JA. Effects of phosphorus/calcium-restricted and phosphorus/calciumreplete 32% protein diets in dogs with chronic renal failure. Am J Vet Res. 1992;53:157–163. 50. Ross LA, Finco DR, Crowell WA. Effect of dietary phosphorus restriction on the kidneys of cats with reduced renal mass. Am J Vet Res. 1982;43: 1023–1026. 51. Slatopolsky E, Caglar S, Pennell JP, et al. On the pathogenesis of hyperparathyroidism in chronic experimental renal insufficiency in the dog. J Clin Invest. 1971;50:492–499. 52. Finch JL, Lee DH, Liapis H, et al. Phosphate restriction significantly reduces mortality in uremic rats with established vascular calcification. Kidney Int. 2013;84:1145–1153. 53. Ibels LS, Alfrey AC, Haut L, Huffer WE. Preservation of function in experimental renal disease by dietary restriction of phosphate. N Engl J Med. 1978;298:122–126. 54. Lumlertgul D, Burke TJ, Gillum DM, et al. Phosphate depletion arrests progression of chronic renal failure independent of protein intake. Kidney Int. 1986;29:658–666. 55. Bauer KD, Griminger P. Long-term effects of activity and of calcium and phosphorus intake on bones and kidneys of female rats. J Nutr. 1983;113:2011–2021. 56. Farquharson RF, Tibbetts DM. Studies of calcium and phosphorus metabolism: XVIII. On temporary fluctuations in the level of calcium and inorganic phosphorus in blood serum of normal individuals. J Clin Invest. 1931;10:271–286. 57. Spencer H, Menczel J, Lewin I, Samachson J. Effect of high phosphorus intake on calcium and phosphorus metabolism in man. J Nutr. 1965;86: 125–132. 58. Gutierrez OM, Luzuriaga-McPherson A, Lin Y, Gilbert LC, Ha SW, Beck GR Jr. Impact of phosphorus-based food additives on bone and mineral metabolism. J Clin Endocrinol Metab. 2015;100:4264–4271. 59. Chang AR, Lazo M, Appel LJ, Gutierrez OM, Grams ME. High dietary phosphorus intake is associated with all-cause mortality: results from NHANES III. Am J Clin Nutr. 2014;99:320–327. 60. Murtaugh MA, Filipowicz R, Baird BC, Wei G, Greene T, Beddhu S. Dietary phosphorus intake and mortality in moderate chronic kidney disease: NHANES III. Nephrology Transplant. 2012;27:990–996. 61. Noori N, Kalantar-Zadeh K, Kovesdy CP, Bross R, Benner D, Kopple JD. Association of dietary phosphorus intake and phosphorus to protein ratio with mortality in hemodialysis patients. Clin J Am Soc Nephrol. 2010;5:683–692. Kidney International (2018) 93, 1060–1072

review

62. Yamamoto KT, Robinson-Cohen C, de Oliveira MC, et al. Dietary phosphorus is associated with greater left ventricular mass. Kidney Int. 2013;83:707–714. 63. Hill KM, Martin BR, Wastney ME, et al. Oral calcium carbonate affects calcium but not phosphorus balance in stage 3-4 chronic kidney disease. Kidney Int. 2013;83:959–966. 64. Shuto E, Taketani Y, Tanaka R, et al. Dietary phosphorus acutely impairs endothelial function. J Am Soc Nephrol. 2009;20:1504–1512. 65. Stevens KK, Denby L, Patel RK, et al. Deleterious effects of phosphate on vascular and endothelial function via disruption to the nitric oxide pathway. Nephrology Transplant. 2017;32:1617–1627. 66. Palomino HL, Rifkin DE, Anderson C, Criqui MH, Whooley MA, Ix JH. 24–hour urine phosphorus excretion and mortality and cardiovascular events. Clin J Am Soc Nephrol. 2013;8:1202–1210. 67. Selamet U, Tighiouart H, Sarnak MJ, et al. Relationship of dietary phosphate intake with risk of end-stage renal disease and mortality in chronic kidney disease stages 3-5: the Modification of Diet in Renal Disease study. Kidney Int. 2016;89:176–184. 68. Evenepoel P, Vervloet MG. Dietary phosphorus restriction in predialysis chronic kidney disease: time for a cease-fire? Kidney Int. 2016;89:21–23. 69. Moe SM, Zidehsarai MP, Chambers MA, et al. Vegetarian compared with meat dietary protein source and phosphorus homeostasis in chronic kidney disease. Clin J Am Soc Nephrol. 2011;6:257–264. 70. Fouque D, Laville M. Low protein diets for chronic kidney disease in non diabetic adults. Cochrane Database Syst Rev. 2009:CD001892. 71. Gross JL, Zelmanovitz T, Moulin CC, et al. Effect of a chicken-based diet on renal function and lipid profile in patients with type 2 diabetes: a randomized crossover trial. Diabetes Care. 2002;25:645–651. 72. Jacobs DR Jr, Gross MD, Steffen L, et al. The effects of dietary patterns on urinary albumin excretion: results of the Dietary Approaches to Stop Hypertension (DASH) trial. Am J Kidney Dis. 2009;53:638–646. 73. Kontessis P, Jones S, Dodds R, et al. Renal, metabolic and hormonal responses to ingestion of animal and vegetable proteins. Kidney Int. 1990;38:136–144. 74. Kemi VE, Rita HJ, Karkkainen MU, et al. Habitual high phosphorus intakes and foods with phosphate additives negatively affect serum parathyroid hormone concentration: a cross-sectional study on healthy premenopausal women. Public Health Nutr. 2009;12: 1885–1892. 75. Gutierrez OM, Katz R, Peralta CA, et al. Associations of socioeconomic status and processed food intake with serum phosphorus concentration in community-living adults: the Multi-Ethnic Study of Atherosclerosis (MESA). J Ren Nutr. 2012;22:480–489. 76. Itkonen ST, Karp HJ, Kemi VE, et al. Associations among total and food additive phosphorus intake and carotid intima-media thickness–a cross-sectional study in a middle-aged population in Southern Finland. Nutr J. 2013;12:94. 77. Chang AR, Miller ER 3rd, Anderson CA, et al. Phosphorus additives and albuminuria in early stages of CKD: a randomized controlled trial. Am J Kidney Dis. 2017;69:200–209. 78. Ketteler M, Block GA, Evenepoel P, et al. Executive summary of the 2017 KDIGO chronic kidney disease-mineral and bone disorder (CKD-MBD) guideline update: what’s changed and why it matters. Kidney Int. 2017;92:26–36. 79. de Fornasari ML, Dos Santos Sens YA. Replacing phosphorus-containing food additives with foods without additives reduces phosphatemia in end-stage renal disease patients: a randomized clinical trial. J Ren Nutr. 2017;27:97–105. 80. Sullivan C, Sayre SS, Leon JB, et al. Effect of food additives on hyperphosphatemia among patients with end-stage renal disease: a randomized controlled trial. JAMA. 2009;301:629–635. 81. Lou LM, Caverni A, Gimeno JA, et al. Dietary intervention focused on phosphate intake in hemodialysis patients with hyperphosphoremia. Clin Nephrol. 2012;77:476–483. 82. Kuhlmann MK. Practical approaches to management of hyperphosphatemia: can we improve the current situation? Blood Purif. 2007;25:120–124. 83. Caldeira D, Amaral T, David C, Sampaio C. Educational strategies to reduce serum phosphorus in hyperphosphatemic patients with chronic kidney disease: systematic review with meta-analysis. J Ren Nutr. 2011;21:285–294. 84. Karavetian M, de Vries N, Rizk R, Elzein H. Dietary educational interventions for management of hyperphosphatemia in hemodialysis patients: a systematic review and meta-analysis. Nutr Rev. 2014;72:471–482.

1071

review

85. Benini O, D’Alessandro C, Gianfaldoni D, Cupisti A. Extra-phosphate load from food additives in commonly eaten foods: a real and insidious danger for renal patients. J Ren Nutr. 2011;21:303–308. 86. Daugirdas JT, Finn WF, Emmett M, Chertow GM. Frequent hemodialysis network trial G. The phosphate binder equivalent dose. Semin Dial. 2011;24:41–49. 87. Moreira PV, Hyseni L, Moubarac JC, et al. Effects of reducing processed culinary ingredients and ultra-processed foods in the Brazilian diet: a cardiovascular modelling study. Public Health Nutr. 2017:1–8. 88. Adatorwovor R, Roggenkamp K, Anderson JJ. Intakes of calcium and phosphorus and calculated calcium-to-phosphorus ratios of older adults: NHANES 2005-2006 data. Nutrients. 2015;7:9633–9639. 89. Westenbrink S, Jansen-van der Vliet M. Dutch Food Composition Database: National Institute for Public Health and the Environment (RIVM). Bilthoven, Kingdom of the Netherlands. 2016. 90. Calvo MS, Uribarri J. Public health impact of dietary phosphorus excess on bone and cardiovascular health in the general population. Am J Clin Nutr. 2013;98:6–15. 91. Karp H, Ekholm P, Kemi V, et al. Differences among total and in vitro digestible phosphorus content of plant foods and beverages. J Ren Nutr. 2012;22:416–422. 92. Carrigan A, Klinger A, Choquette SS, et al. Contribution of food additives to sodium and phosphorus content of diets rich in processed foods. J Ren Nutr. 2014;24:13–19. 93. Schmidt DR, Holmstrom SR, Fon Tacer K, Bookout AL, Kliewer SA, Mangelsdorf DJ. Regulation of bile acid synthesis by fat-soluble vitamins A and D. J Biol Chem. 2010;285:14486–14494. 94. Kimura M, Itokawa Y. Cooking losses of minerals in foods and its nutritional significance. J Nutr Sci Vitaminol (Tokyo). 1990;36 Suppl 1: S25–S32. discussion S33. 95. Jones WL. Demineralization of a wide variety of foods for the renal patient. J Ren Nutr. 2001;11:90–96. 96. Nelson SM, Sarabia SR, Christilaw E, et al. Phosphate-containing prescription medications contribute to the daily phosphate intake in a third of hemodialysis patients. J Ren Nutr. 2017;27:91–96. 97. Sherman RA, Ravella S, Kapoian T. A dearth of data: the problem of phosphorus in prescription medications. Kidney Int. 2015;87:1097–1099. 98. Block GA, Wheeler DC, Persky MS, et al. Effects of phosphate binders in moderate CKD. J Am Soc Nephrol. 2012;23:1407–1415. 99. Drueke TB, Massy ZA. Phosphate binders in CKD: bad news or good news? J Am Soc Nephrol. 2012;23:1277–1280. 100. Sekercioglu N, Thabane L, Diaz Martinez JP, et al. Comparative effectiveness of phosphate binders in patients with chronic kidney disease: a systematic review and network meta-analysis. PLoS One. 2016;11:e0156891. 101. Russo D, Miranda I, Ruocco C, et al. The progression of coronary artery calcification in predialysis patients on calcium carbonate or sevelamer. Kidney Int. 2007;72:1255–1261. 102. Di Iorio B, Bellasi A, Russo D, Investigators IS. Mortality in kidney disease patients treated with phosphate binders: a randomized study. Clin J Am Soc Nephrol. 2012;7:487–493. 103. Neradova A, Schumacher SP, Hubeek I, Lux P, Schurgers LJ, Vervloet MG. Phosphate binders affect vitamin K concentration by undesired binding, an in vitro study. BMC Nephrol. 2017;18:149. 104. Tonelli M, Pannu N, Manns B. Oral phosphate binders in patients with kidney failure. N Engl J Med. 2010;362:1312–1324. 105. Yusuf AA, Howell BL, Powers CA, St Peter WL. Utilization and costs of medications associated with CKD mineral and bone disorder in dialysis patients enrolled in Medicare Part D. Am J Kidney Dis. 2014;64:770–780. 106. Ghimire S, Castelino RL, Lioufas NM, Peterson GM, Zaidi ST. Nonadherence to medication therapy in haemodialysis patients: a systematic review. PLoS One. 2015;10:e0144119. 107. Tentori F, Blayney MJ, Albert JM, et al. Mortality risk for dialysis patients with different levels of serum calcium, phosphorus, and PTH: the Dialysis Outcomes and Practice Patterns study (DOPPS). Am J Kidney Dis. 2008;52:519–530. 108. Fernandez-Martin JL, Martinez-Camblor P, Dionisi MP, et al. Improvement of mineral and bone metabolism markers is associated with better survival in haemodialysis patients: the COSMOS study. Nephrology Transplant. 2015;30:1542–1551. 109. Lopes AA, Tong L, Thumma J, et al. Phosphate binder use and mortality among hemodialysis patients in the Dialysis Outcomes and Practice Patterns study (DOPPS): evaluation of possible confounding by nutritional status. Am J Kidney Dis. 2012;60:90–101.

1072

MG Vervloet and AJ van Ballegooijen: Managing phosphate in CKD

110. Cannata-Andia JB, Fernandez-Martin JL, Locatelli F, et al. Use of phosphate-binding agents is associated with a lower risk of mortality. Kidney Int. 2013;84:998–1008. 111. Isakova T, Gutierrez OM, Chang Y, et al. Phosphorus binders and survival on hemodialysis. J Am Soc Nephrol. 2009;20:388–396. 112. Investigators ET, Chertow GM, Block GA, et al. Effect of cinacalcet on cardiovascular disease in patients undergoing dialysis. N Engl J Med. 2012;367:2482–2494. 113. Zitt E, Fouque D, Jacobson SH, et al. Serum phosphorus reduction in dialysis patients treated with cinacalcet for secondary hyperparathyroidism results mainly from parathyroid hormone reduction. Clin Kidney J. 2013;6:287–294. 114. Block G, Do TP, Collins AJ, Cooper KC, Bradbury BD. Co-trending of parathyroid hormone and phosphate in patients receiving hemodialysis. Clin Nephrol. 2016;85:142–151. 115. Giral H, Caldas Y, Sutherland E, et al. Regulation of rat intestinal Na-dependent phosphate transporters by dietary phosphate. Am J Physiol Renal Physiol. 2009;297:F1466–1475. 116. Xu H, Bai L, Collins JF, Ghishan FK. Age-dependent regulation of rat intestinal type IIb sodium-phosphate cotransporter by 1,25-(OH)(2) vitamin D(3). Am J Physiol Cell Physiol. 2002;282:C487–493. 117. Eto N, Miyata Y, Ohno H, Yamashita T. Nicotinamide prevents the development of hyperphosphataemia by suppressing intestinal sodium-dependent phosphate transporter in rats with adenine-induced renal failure. Nephrology Transplant. 2005;20:1378–1384. 118. Lenglet A, Liabeuf S, El Esper N, et al. Efficacy and safety of nicotinamide in haemodialysis patients: the NICOREN study. Nephrology Transplant. 2016;32:870–879. 119. Lenglet A, Liabeuf S, Guffroy P, Fournier A, Brazier M, Massy ZA. Use of nicotinamide to treat hyperphosphatemia in dialysis patients. Drugs R D. 2013;13:165–173. 120. Cheng SC, Young DO, Huang Y, Delmez JA, Coyne DW. A randomized, double-blind, placebo-controlled trial of niacinamide for reduction of phosphorus in hemodialysis patients. Clin J Am Soc Nephrol. 2008;3: 1131–1138. 121. Edalat-Nejad M, Zameni F, Talaiei A. The effect of niacin on serum phosphorus levels in dialysis patients. Indian J Nephrol. 2012;22:174–178. 122. Block GA, Rosenbaum DP, Leonsson-Zachrisson M, et al. Effect of tenapanor on serum phosphate in patients receiving hemodialysis. J Am Soc Nephrol. 2017;28:1933–1942. 123. Kidney disease: improving global outcomes CKDMBDWG. KDIGO clinical practice guideline for the diagnosis, evaluation, prevention, and treatment of chronic kidney disease-mineral and bone disorder (CKDMBD). Kidney Int Suppl. 2009:S1–S130. 124. Di Iorio B, Molony D, Bell C, et al. Sevelamer versus calcium carbonate in incident hemodialysis patients: results of an open-label 24-month randomized clinical trial. Am J Kidney Dis. 2013;62:771–778. 125. Nappi SE, Virtanen VK, Saha HH, Mustonen JT, Pasternack AI. QTc dispersion increases during hemodialysis with low-calcium dialysate. Kidney Int. 2000;57:2117–2122. 126. Pun PH, Horton JR, Middleton JP. Dialysate calcium concentration and the risk of sudden cardiac arrest in hemodialysis patients. Clin J Am Soc Nephrol. 2013;8:797–803. 127. Patel L, Bernard LM, Elder GJ. Sevelamer versus calcium-based binders for treatment of hyperphosphatemia in CKD: a meta-analysis of randomized controlled trials. Clin J Am Soc Nephrol. 2015;11:232–234. 128. Jamal SA, Vandermeer B, Raggi P, et al. Effect of calcium-based versus non-calcium-based phosphate binders on mortality in patients with chronic kidney disease: an updated systematic review and metaanalysis. Lancet. 2013;382:1268–1277. 129. Suki WN, Zabaneh R, Cangiano JL, et al. Effects of sevelamer and calcium-based phosphate binders on mortality in hemodialysis patients. Kidney Int. 2007;72:1130–1137. 130. Qunibi W, Moustafa M, Muenz LR, et al. A 1-year randomized trial of calcium acetate versus sevelamer on progression of coronary artery calcification in hemodialysis patients with comparable lipid control: the Calcium Acetate Renagel Evaluation-2 (CARE-2) study. Am J Kidney Dis. 2008;51:952–965. 131. Floege J. Calcium-containing phosphate binders in dialysis patients with cardiovascular calcifications: should we CARE-2 avoid them? Nephrol Transplant. 2008;23:3050–3052. 132. Wald R, Rabbat CG, Girard L, et al. Two phosphAte taRGets in End-stage renal disease Trial (TARGET): a randomized controlled trial. Clin J Am Soc Nephrol. 2017;12:965–973.

Kidney International (2018) 93, 1060–1072