Management of hyperphosphatemia in adults with chronic kidney disease

INTRODUCTION — The tendency toward phosphate retention develops early in chronic kidney disease (CKD) due to the reduction in the filtered phosphate load. Overt hyperphosphatemia develops when the estimated glomerular filtration rate (eGFR) falls below 25 to 40 mL/min/1.73 m² [1-3].

Hyperphosphatemia has been associated with increased mortality and morbidity [4-10].

This topic reviews recommendations regarding target phosphate concentration and treatment options for hyperphosphatemia for patients with CKD.

Recommended goals for serum parathyroid hormone (PTH) concentration for patients with CKD are discussed elsewhere. (See "Management of secondary hyperparathyroidism in adult patients on dialysis", section on 'Treat high parathyroid hormone'.)

Mechanisms underlying the physiologic response to phosphate retention are discussed elsewhere. (See "Overview of chronic kidney disease-mineral and bone disorder (CKD-MBD)".)

MONITORING — In patients with estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m², we routinely monitor serum phosphate, calcium, intact parathyroid hormone (iPTH), and 25-hydroxyvitamin D levels.

The frequency with which these measurements are performed depends on the eGFR and whether baseline abnormalities are present or therapeutic measures have been taken. In addition, monitoring may be performed more frequently if the eGFR is rapidly decreasing. Values should be rechecked if kidney function worsens.

The following monitoring schedule is reasonable and based upon the 2017 Kidney Disease: Improving Global Outcomes (KDIGO) guidelines [11].

Estimated GFR 30 to 59 mL/min/1.73 m2 — In patients with estimated glomerular filtration rate (eGFR) 30 to 59 mL/min/1.73 m², we measure serum phosphate and calcium every 6 to 12 months. We measure iPTH at least every 12 months and every 6 months if the baseline concentration is elevated or if the patient is being treated. We measure 25-hydroxyvitamin D every 12 months and every 6 months if the patient is being treated.

Estimated GFR 15 to 29 mL/min/1.73 m2 — In patients with estimated glomerular filtration rate (eGFR) 15 to 29 mL/min/1.73 m², we measure serum phosphate and calcium every three to six months. We measure iPTH every 6 to 12 months and every 6 months if the baseline concentration is elevated or if the patient is being treated. We measure 25-hydroxyvitamin D every 12 months and every 6 months if the patient is being treated for vitamin D deficiency.

Estimated GFR <15 mL/min/1.73 m2 (including patients on dialysis) — In patients with estimated glomerular filtration rate (eGFR) <15 mL/min/1.73 m², we measure serum phosphate and calcium every one to three months. We measure iPTH every three to six months and every three months if the baseline concentration is elevated or if the patient is being treated. We measure 25-hydroxyvitamin D every 12 months and every 6 months if the patient is being treated for vitamin D deficiency.

TREATMENT APPROACH

Patients with nondialysis chronic kidney disease — In patients with nondialysis CKD, we try to maintain the serum phosphate level in the normal range (ie, <4.5 mg/dL [1.45 mmol/L]) with dietary modification; this level is consistent with the Kidney Disease: Improving Global Outcomes (KDIGO) guidelines [11-13]. We only use phosphate binders when the serum phosphate level is persistently elevated >5.5 mg/dL despite dietary restriction [12,14-16]. In patients who present with very high phosphate (ie, >6 mg/dL), we start phosphate binders concurrently with dietary restriction because it is unlikely that dietary restriction alone will be effective. The selection of type of phosphate binder is the same for patients on dialysis and those with nondialysis CKD and is discussed below. (See 'Specific treatment' below.)

We suggest a moderate restriction of phosphate intake (approximately 900 mg/day) in patients with nondialysis CKD, provided that this can be done without compromising nutritional status (table 1A-B). These recommendations are consistent with the Kidney Disease Outcomes Quality Initiative (KDOQI) and KDIGO guidelines [11,12,14]. (See 'Phosphate restriction' below.)

We do not attempt to reduce phosphate to lower than normal values or prevent an increase in serum phosphate among patients who have a normal phosphate concentration.

Studies examining the association of hyperphosphatemia with mortality have shown mixed results. A predominant number of studies report an increased mortality in association with hyperphosphatemia among patients with nondialysis CKD [10,17-21]. This was best illustrated in a meta-analysis of three studies with nearly 5000 patients with nondialysis CKD, which showed a 35 percent increase in mortality per mg increase in phosphate above normal values (95% CI 1.16-1.57) [10]. The median follow-up in individual studies was approximately one to two years. In one study (which was included in the meta-analysis), serum phosphate >3.5 mg/dL (1.13 mmol/L) was an independent predictor of all-cause mortality [17].

Hyperphosphatemia is a stimulus for hyperparathyroidism. Studies also suggest that there is progressive cardiovascular risk associated with hyperphosphatemia in patients with normal kidney function, as well in patients with CKD with estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m² who are not on dialysis [17,20,22,23].

Some other studies have not demonstrated an association between serum phosphate and mortality among patients with nondialysis CKD [24,25]. One study published after the cited meta-analysis did not show an association between quartiles of serum phosphate and mortality over a mean follow-up of over two years [24]. Differences in study populations are a likely reason for the discrepancy between studies [3]. As an example, individuals in the later study had much milder kidney dysfunction (with mean eGFR of 44 to 48 mL/min/1.73 m²) compared with prior studies.

The effect of phosphate binders on serum phosphate levels can be variable, and the prescription of a phosphate binder increases the patient's pill burden. In a randomized trial, compared with placebo, the use of phosphate binders was associated with a greater decrease in serum phosphate and in 24-hour urine phosphate at three, six, and nine months, although the effect was small [26]. In addition, the use of phosphate binders was associated with a stable parathyroid hormone (PTH) value versus an increase in the placebo group. In another trial, use of a phosphate binder (sevelamer) led to no improvement in serum phosphate [27]. In addition, no intervention to lower serum phosphate (dietary restriction or use of phosphate binders) has been shown to affect clinically important outcomes among patients with nondialysis CKD [27,28]. Thus, some experts do not recommend phosphate binders.

An additional concern is that calcium-containing phosphate binders may induce positive calcium balance in patients with CKD, which could lead to vascular calcification. (See 'Specific treatment' below.)

Patients on dialysis

Initial therapy — In most patients on dialysis, we try to maintain the serum phosphate concentration between 3.5 and 5.5 mg/dL (1.13 and 1.78 mmol/L), although there are no data that this improves outcomes. Serum phosphate >5.5 mg/dL (1.78 mmol/L) is an indication for treatment.

Increased serum phosphate is associated with increased mortality among patients on dialysis [4-10]. This was best shown in a meta-analysis of 12 studies that included 92,345 patients with CKD, over 97 percent of whom were on dialysis [10]. Among 10 studies that were perceived to be adequately adjusted (in which seven studies were of patients on dialysis), serum phosphate >5.5 mg/dL (1.78 mmol/L) was associated with increased mortality. Based upon 13 studies that reported a continuous relative risk for each mg/dL increase in phosphate, the mortality risk increased by 18 percent (95% CI 1.12-1.25).

A target phosphate goal between 3.5 and 5.5 mg/dL (1.13 and 1.78 mmol/L) is consistent with the KDOQI guidelines [14]. This target differs from the more recent KDIGO guidelines, which recommend lowering levels toward the normal range but do not specify a target threshold [11,12]. We feel that the KDIGO recommendations do not provide a practical goal.

The initial therapy includes phosphate restriction and phosphate binders [12,29].

We suggest a moderate restriction of phosphate intake in patients on dialysis, provided that this can be done without compromising nutritional status. These recommendations are consistent with the KDOQI and KDIGO guidelines [11,14].

We restrict dietary phosphate to 900 mg/day. Among patients on dialysis, phosphate restriction should be done under the supervision of a dietician. This is because many patients on dialysis have either overt or borderline malnutrition. For such patients, protein supplementation (which contributes to phosphorus intake) rather than protein restriction is the goal. In this setting, the patient should be encouraged to avoid unnecessary dietary phosphate (such as phosphate-containing food additives, dairy products, certain vegetables, many processed foods, and colas), while increasing the intake of high-biologic-value sources of protein (such as meat and eggs) [30,31]. (See 'Phosphate restriction' below.)

While we may try dietary restriction first and only use a phosphate binder if the phosphate is persistently and progressively elevated after one or two months of dietary phosphorus restriction, most patients will also require a phosphate binder. Many clinicians will start dietary phosphate restriction and phosphate binders simultaneously in all patients with hyperphosphatemia.

Two observational studies have suggested that phosphate binders are associated with decreased mortality among patients on dialysis:

●In a prospective, observational study of patients on incident dialysis followed for one year, the use of phosphate binders was associated with a 25 percent lower one-year mortality [32].

●Among 6797 patients enrolled in an observational, prospective study (Current Management of Secondary Hyperparathyroidism: A Multicenter Observational Study [COSMOS]), patients prescribed phosphate binders had a 29 percent lower risk of all-cause mortality and a 22 percent lower risk of cardiovascular mortality [33].

Randomized trials are needed to determine whether the use of phosphate binders provides a benefit on clinically important endpoints among patients on dialysis.

The choice of phosphate binder is discussed below. (See 'Specific treatment' below.)

In addition to phosphate restriction and binders, it is important to make sure patients are adequately dialyzed and achieving recommended Kt/V targets. (See "Prescribing and assessing adequate hemodialysis".)

However, in patients who are undergoing standard in-center hemodialysis (which represents most patients on hemodialysis), we do not generally increase the dialysis above the recommended dose, as this has not been shown to improve clinically important outcomes.

Although an exception may be made for exceptionally motivated patients who may be willing to undergo more dialysis, most patients find it difficult to increase the dialysis time sufficiently to reduce the serum phosphate. Standard dialysis (ie, three times weekly for four hours per session) is limited in its ability to remove phosphate. The average standard dialysis session removes approximately 900 mg of phosphate.

Some patients may consider daily or nocturnal (ie, usually prolonged and nightly) hemodialysis. Frequent or prolonged hemodialysis leads to greater removal of phosphate and can substantially lower serum phosphate levels. In fact, in some patients undergoing frequent or prolonged hemodialysis, serum phosphate can be controlled without the use of phosphate binders. However, many factors impact a decision to undergo daily dialysis, and hyperphosphatemia alone is rarely the deciding factor. (See "Short daily hemodialysis", section on 'Phosphate' and "Outcomes associated with nocturnal hemodialysis", section on 'Phosphate'.)

Occasionally, hyperphosphatemia will be refractory to these methods. (See 'Refractory hyperphosphatemia' below.)

Refractory hyperphosphatemia — Among patients on dialysis, hyperphosphatemia is occasionally refractory to dietary restriction and phosphate binders (see 'Specific treatment' below), and patients may be unwilling or unable to undergo frequent or prolonged hemodialysis.

In such patients, our stepwise approach is as follows:

●Optimize treatment of hyperparathyroidism – The treatment of hyperparathyroidism should be reviewed. Both inadequately treated hyperparathyroidism and the use of high doses of calcitriol or active vitamin D analogs may contribute to hyperphosphatemia through different mechanisms. High levels of PTH increase bone resorption and the release of phosphate from bone [34] and active vitamin D moieties increase the gastrointestinal absorption of phosphate.

Among such patients, the phosphate may be reduced by decreasing PTH using calcimimetics instead of calcitriol or vitamin D analogs. In select patients with uncontrolled severe hyperparathyroidism, parathyroidectomy may be indicated, although this is rare. (See "Management of secondary hyperparathyroidism in adult patients on dialysis", section on 'Refractory hyperparathyroidism' and "Refractory hyperparathyroidism and indications for parathyroidectomy in adult patients on dialysis", section on 'Symptomatic patients'.)

●Add tenapanor to the medical regimen – In adequately dialyzed patients who have serum phosphate levels persistently >5.5 mg/dL despite dietary restriction, optimal treatment of hyperparathyroidism, and the appropriate use of phosphate binders (see 'Specific treatment' below), we suggest adding tenapanor to the medical regimen. Tenapanor is an inhibitor of intestinal sodium/hydrogen exchanger 3 (NHE3) that was developed to treat irritable bowel syndrome with constipation but was also found to lower serum phosphate by blocking its paracellular transport from the intestinal lumen [35]. Since no data demonstrate that lowering phosphate levels to 5.5 mg/dL or below improves clinical outcomes, some experts use higher levels of serum phosphate (eg, >6.5 mg/dL) as a threshold for tenapanor add-on therapy.

The usual dose range of tenapanor is 10 to 30 mg twice daily, taken before the first and last meals of the day. The major adverse effect of tenapanor is an increase in stool frequency, thought to be due to an increase in stool water content. Diarrhea occurs in approximately half of tenapanor-treated patients [36] but is usually not severe; diarrhea leads to drug discontinuation in approximately five percent of patients.

In patients on hemodialysis, multiple trials have demonstrated that tenapanor lowers phosphate to a modest degree [37-42]. In a meta-analysis of randomized trials, tenapanor therapy compared with placebo resulted in a mean decrease in serum phosphorus of 1.79 mg/dL [36]. In one representative trial, 169 patients on hemodialysis with hyperphosphatemia despite treatment with one or more phosphate binders were randomized to tenapanor or placebo; phosphate binders were continued in both groups [41]. Patients treated with tenapanor had a serum phosphorus 1.76 mg/dL lower than patients in the placebo group. Diarrhea occurred in 63.1 percent and 14.1 percent of patients in the tenapanor and placebo groups, respectively, but was reported as mild or moderate in all cases.

Specific treatment

Phosphate restriction — Dietary phosphate restriction can be effective in reducing the serum phosphate concentration [31], although the effect is inconsistent, and interventions designed to reduce dietary phosphate intake may have minimal effects on serum and urinary phosphate concentrations [43,44]. Moreover, no adequately powered studies have examined the efficacy of dietary phosphate restrictions on patient-important outcomes.

Approximately 900 mg/day of dietary phosphate is a level that at least some patients will find acceptable. Phosphate restriction should primarily include processed foods and colas and not high-biologic-value foods such as meat and eggs (table 1A-B). Food additives (as are found in processed foods) and medications are an important source of dietary phosphate [45]. In addition to having high phosphate content, processed foods provide a more easily absorbed form of phosphate compared with fresh animal- and plant-based foods [30,46]. We agree with the KDIGO guidelines that patients should receive education on the absorbability of phosphate from different food [11].

Some nephrologists recommend a more vegetarian-based diet in order to control phosphate. Phosphate bioavailability may be less with a vegetarian diet compared with an animal protein-based diet [47]. Many foods that have traditionally been labeled high phosphate (such as beans and nuts) may actually be acceptable (providing they are not too high in potassium) because phosphate from these sources is absorbed slowly [11,46]. This is because plant-derived phosphate found in unprocessed foods is in the form of phytate phosphorus, and the human intestine does not secrete phytase, the enzyme required for absorption [30]. In addition, such a diet rich in legumes, nuts, and whole grains may also result in higher fiber intake while offering wider food choices [46].

A randomized study of patients on hemodialysis showed that careful instruction on avoiding foods with phosphorus additives resulted in larger declines in serum phosphate compared with control patients who did not receive such instruction [31].

Very few studies have examined the efficacy of dietary phosphate restrictions on patient-important outcomes among patients on dialysis. In a post hoc analysis of data from the Hemodialysis (HEMO) study, prescribed phosphate restriction was not associated with improved survival of patients on hemodialysis [48]. In fact, there was a stepwise trend toward better survival with less restrictive prescribed phosphate intake in this study. However, the HEMO study was performed prior to an awareness of the importance of food additives as a source of phosphate; as a result, phosphate restriction may have been achieved by the restriction of nutritionally beneficial food. Supporting this possibility, phosphate restriction tended to be associated with poorer nutritional indices and a persistently greater need for nutritional supplementation in this study.

Phosphate binders — Phosphate binders are categorized as calcium-containing and non-calcium-containing. Calcium-containing binders include calcium carbonate and calcium acetate. Major non-calcium-containing binders include sevelamer and lanthanum. Other agents include ferric citrate and sucroferric oxyhydroxide. All are equivalently effective in lowering phosphate [14].

For most patients, we suggest that non-calcium-containing binders be used. Exceptions may be made where non-calcium-containing binders are not available or affordable or when the serum calcium is low and PTH is elevated, such as in patients concomitantly treated with calcimimetics. However many experts argue that calcium-containing binders should be avoided in all patients [11]. Dosing of these agents is presented below. (See 'Dose and specific agents' below.)

A number of trials and two meta-analyses have suggested that non-calcium-containing phosphate binders, compared with calcium-containing phosphate binders, decrease mortality among patients with CKD [49-59].

A meta-analysis of 11 open-label, randomized trials (4622 patients) revealed a 22 percent decrease in all-cause mortality among patients randomly assigned to receive non-calcium-based binders (sevelamer, 10 studies including 3268 patients, or lanthanum, one study including 1354 patients) compared with calcium-based binders (relative risk [RR] 0.78, 95% CI 0.61-0.98) [60]. The results of this meta-analysis were driven in large part by the study that used lanthanum carbonate not sevelamer. Analysis of patients on dialysis and those with nondialysis CKD showed similar reductions in mortality. Most studies in this meta-analysis were limited to 24 months. Analysis of studies with follow-up at 36 and 42 months showed no difference in mortality [60]. This analysis did not look at cardiovascular mortality.

A second meta-analysis also reported a decrease in all-cause mortality with sevelamer compared with calcium-based binders (13 studies, n = 3799, RR 0.54, 95% CI 0.32-0.93) [61]. There was no reduction in cardiovascular mortality demonstrated (four studies, n = 2712, RR 0.33, 95% CI 0.07-1.64). Patients receiving sevelamer had lower cholesterol, low-density lipoprotein (LDL)-cholesterol, calcium, and decreased risk of hypercalcemia. There was no difference between groups in serum phosphate. There was considerable heterogeneity among the studies.

However, a large, observational cohort study of over 4000 older patients on incident dialysis found that the risk of fatal or nonfatal cardiovascular events or all-cause mortality was not different among patients treated with either sevelamer or calcium acetate [62].

In addition to the possible effects on mortality, calcium-containing binders, but not non-calcium-containing binders, are associated with hypercalcemia, adynamic bone disease, and vascular calcification, all which could result in increased morbidity [26,49,51,53,54,61,63].

Even in the absence of hypercalcemia, a positive calcium balance may increase vascular calcification. Because normal dietary calcium intake is roughly 1000 mg per day, prescription of 1500 mg elemental calcium per day (ie, calcium carbonate 1250 mg three times daily with meals) increases calcium ingestion by roughly 2.5-fold. In addition, calcium excretion is reduced in CKD because of a reduced filtered load of calcium. The combination of increased calcium ingestion and decreased calcium excretion could lead to positive calcium balance, even in the absence of hypercalcemia. This was demonstrated in a well-conducted randomized, placebo-controlled crossover study that examined the effect of oral calcium carbonate administration on calcium and phosphate balance in eight patients with CKD with a mean eGFR of 15 to 59 mL/min/1.73 m² [64]. Subjects received a controlled diet with either a calcium carbonate supplement (1500 mg/day calcium) or placebo during two three-week balance periods. Fasting blood and urine were collected at baseline and at the end of each week. All feces and urine were collected during weeks 2 and 3 of each balance period.

An oral and intravenous calcium isotope (⁴⁵CaCl₂) was administered to determine calcium kinetics. Patients were in neutral calcium and phosphorus balance while on the placebo. Calcium carbonate supplementation caused a positive calcium balance and had no effect on phosphorus balance. In addition, compared with placebo, calcium carbonate supplementation produced a small reduction in urine phosphorus excretion.

Calcium kinetics demonstrated positive net bone balance. However, the amount of calcium that was deposited in bone was less than the overall positive calcium balance, suggesting that some degree of soft-tissue deposition occurred. Fasting blood and urine biochemistries of calcium and phosphate homeostasis were unaffected by calcium carbonate, suggesting that it is futile to rely solely on blood concentrations to determine mineral excess or accumulation.

The interpretation of these data may be limited by the short-term nature of the study. It is possible that patients were not in steady state after only one week of calcium administration. If so, the short-term positive calcium balance that was observed may have been an appropriate response to correct years of bone calcium depletion and thus may decrease over time [65]. In studies of longer duration in predialysis CKD, 24-hour phosphate excretion substantially declined when calcium-containing binders were used [26].

Dose and specific agents — For all phosphate buffers, the lowest dose that is effective should be used. If calcium-containing buffers are selected, we suggest that the total dose of elemental calcium (including dietary sources) not exceed 2000 mg per day [14]. The amount of elemental calcium contained in the phosphate binder should not exceed 1500 mg per day. This is consistent with both the KDOQI and KDIGO guidelines [12-14].

Phosphate binders are only effective if taken with meals [66].

Specific agents are discussed here.

●Sevelamer – Sevelamer hydrochloride (Renagel) and sevelamer carbonate (Renvela) are nonabsorbable cationic polymers that bind phosphate through ion exchange [67].

Sevelamer is effective in lowering serum phosphate levels [51,52,68-75]. The usual dose range of sevelamer is 800 to 2400 mg three times daily with meals.

Sevelamer hydrochloride, but not sevelamer carbonate, may induce metabolic acidosis. For this reason, sevelamer carbonate is preferred over sevelamer hydrochloride in patients with nondialysis CKD and in any patient with metabolic acidosis. (See "Pathogenesis, consequences, and treatment of metabolic acidosis in chronic kidney disease".)

Sevelamer is much more expensive than calcium-containing phosphate binders [76,77]. By one estimate in the United States, approximately 780 million dollars would be required per year in order to provide sevelamer to all who met KDOQI criteria for its use [76].

●Lanthanum – Lanthanum is a rare-earth element that is effective in lowering phosphate levels in patients on dialysis [78-85] and patients with nondialysis CKD [86]. Compared with calcium-containing phosphate binders, lanthanum appears to be associated with lower incidences of oversuppression of PTH levels [78-80,82,84,85,87-89].

An additional benefit of lanthanum, compared with other phosphate binders, may be a reduced daily tablet burden [90]. In addition, lanthanum tablets are chewed rather than swallowed whole. The usual dose range of lanthanum is 500 to 1000 mg three times daily with meals.

In studies of patients on dialysis, no significant adverse effects have been reported with lanthanum [78-80,82,84,85,87-89]. The safety of lanthanum administered for up to two years was evaluated in 1359 patients on hemodialysis randomly assigned to lanthanum (maximum dose of 3000 mg/day) or their prestudy phosphate binder [84]. The incidence of adverse effects was similar in both groups, which principally consisted of gastrointestinal effects. No evidence of hepatic toxicity was observed.

●Calcium-containing binders – Calcium-containing phosphate binders include calcium carbonate and calcium acetate [91-93]. Calcium acetate may be a more efficient phosphate binder than calcium carbonate [94-96]. However, for most patients we suggest that non-calcium-containing binders be used. (See 'Phosphate binders' above.)

The usual dosage range for calcium carbonate is 1250 to 3750 mg per day in divided doses with meals. The usual dose range for calcium acetate is 1334 to 2001 mg three times daily with meals. The higher ends of these usual dose ranges represent approximately 1500 mg daily of elemental calcium. Because dietary calcium intake is typically on the order of 1000 mg per day, patients taking calcium-containing binders at the high end of the usual dosage range may have total daily intakes of elemental calcium that exceed the recommended level of 2000 mg per day.

As noted above, calcium-containing phosphate binders may lead to extraskeletal calcium phosphate deposition, particularly in the setting of hyperphosphatemia [91,97-102]. Combined hypercalcemia and hyperphosphatemia is a particular problem among patients who are on both calcium-containing phosphate binders and active vitamin D analogs. In such patients who develop hypercalcemia, the dose of the calcium-containing phosphate binder should be decreased [14] (see 'Specific treatment' above). In addition, the dose of active vitamin D analogs should be lowered or discontinued until calcium levels return to normal.

Careful monitoring of the serum calcium concentration is essential with the chronic administration of calcium, particularly in patients on hemodialysis, where the dialysate calcium concentration can vary and therefore impact the ability to administer calcium-containing phosphate binders. (See 'Monitoring' above.)

As noted above however, even in the absence of hypercalcemia, a positive calcium balance, which is generally not detectable without careful balance studies, could increase the risk of vascular calcification. (See 'Phosphate binders' above.)

●Other agents – Other agents include sucroferric oxyhydroxide, ferric citrate, nicotinamide, aluminum hydroxide, and calcium citrate. We generally do not use these agents, either because of high cost, lack of efficacy, or concern over adverse effects.

•Sucroferric oxyhydroxide (Velphoro) – Sucroferric oxyhydroxide is a chewable phosphate binder for patients with eGFR <15 mL/min/1.73 m² [103]. Sucroferric oxyhydroxide appears to be comparable with sevelamer in efficacy and safety and may be associated with a lower pill burden [104,105]. Adverse effects are primarily gastrointestinal (diarrhea, nausea, abnormal product taste, constipation, and vomiting).

The starting dose of sucroferric oxyhydroxide is 2.5 g three times daily with meals or, if dosing is expressed in terms of elemental iron, 500 mg three times daily with meals. The majority of iron from sucroferric oxyhydroxide is not systemically absorbed, although small increases in transferrin saturation and ferritin have been observed with use [106].

•Ferric citrate – Ferric citrate is effective in reducing serum phosphate concentration by approximately the same degree as other phosphate binders [35,107-109]. In addition, ferric citrate raises hemoglobin, serum iron, transferrin saturation, and ferritin. Patients on dialysis who take ferric citrate also may be receiving parenteral iron as part of an anemia management regimen; serum iron, transferrin saturation, and ferritin should be carefully monitored in such patients to avoid iron overload.

Citrate has been shown to enhance the absorption of aluminum, which increases the risk of aluminum toxicity. Because of this, some have recommended avoidance of all citrate-containing products in patients with CKD or who are on dialysis. However, in one trial of patients on maintenance dialysis, there was no aluminum toxicity noted among recipients of ferric citrate [107].

•Nicotinamide – We do not use nicotinamide to reduce phosphate levels in patients with CKD. Nicotinamide, a metabolite of nicotinic acid (niacin, vitamin B3), may lower serum phosphate by reducing gastrointestinal tract phosphate absorption [108,109]. However, randomized trials examining nicotinamide as a treatment to lower phosphate in patients with nondialysis CKD and patients on dialysis have demonstrated limited efficacy, poor tolerance, and some safety concerns [110,111]. For example, in a randomized trial that included over 700 patients on hemodialysis who were using phosphate binders, nicotinamide as add-on therapy lowered phosphate levels compared with placebo at 24 weeks, but this effect was not maintained at 52 weeks of follow-up [111]. In addition, nicotinamide therapy was associated with higher rates of side effects, including diarrhea, pruritus, and thrombocytopenia.

•Aluminum hydroxide – We do not use aluminum hydroxide in patients with CKD. Aluminum hydroxide is effective at controlling serum phosphorus, but the safety of aluminum-based phosphate binders in CKD has not been established. Excess aluminum exposure in CKD can result in aluminum toxicity. The major manifestations of aluminum toxicity are vitamin D-resistant osteomalacia, microcytic anemia, bone and muscle pain, and dementia. (See "Aluminum toxicity in chronic kidney disease".)

•Calcium citrate – Calcium citrate has been used as a phosphate binder. Calcium citrate should be avoided in all patients with CKD since citrate can markedly increase intestinal aluminum absorption [112,113]. The use of calcium citrate has been associated with aluminum neurotoxicity and the rapid onset of symptomatic osteomalacia [114]. (See "Aluminum toxicity in chronic kidney disease".)

Citrate appears to enhance aluminum absorption both by keeping aluminum soluble (via the formation of aluminum citrate) in the intestinal lumen and by complexing with luminal calcium [112,113]. The ensuing decrease in free calcium increases permeability of tight junctions between cells, which can markedly enhance passive aluminum absorption (figure 1) [112,113].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Chronic kidney disease in adults".)

SUMMARY AND RECOMMENDATIONS

●Monitoring serum phosphate – In patients with estimated glomerular filtration rates (eGFR) <60 mL/min/1.73 m², we recommend monitoring serum phosphate, calcium, intact parathyroid hormone (iPTH), and 25-hydroxyvitamin D levels. The frequency of monitoring depends on the eGFR as outlined above. (See 'Monitoring' above.)

●Rationale for treatment – We suggest treating hyperphosphatemia in patients with chronic kidney disease (CKD) (Grade 2C). While high phosphate levels are shown to be associated with increased mortality in patients with CKD, the specific interventions suggested below that lower phosphate levels in most patients have not been shown to improve mortality or other clinical outcomes. (See 'Treatment approach' above.)

●Treatment thresholds – For patients with nondialysis CKD, we treat phosphate values that are persistently and progressively higher than normal (ie, 4.5 mg/dL), while, for patients on dialysis, we treat phosphate concentrations that are persistently >5.5 mg/dL. These somewhat arbitrary thresholds are those that have been shown to be associated with mortality in observational studies. (See 'Treatment approach' above.)

●Dietary phosphate restriction – Our treatment approach begins with dietary phosphate restriction of 900 mg/day. The patient should be encouraged to avoid unnecessary dietary phosphate while maintaining the intake of high-biologic-value sources of protein. (See 'Phosphate restriction' above.)

●Phosphate binders – If hyperphosphatemia persists with dietary restriction, we administer phosphate binders. However, alternative approaches are also reasonable:

•Many experts do not use phosphate binders among patients with nondialysis CKD, since clinically relevant benefits have not been demonstrated.

•Other experts prescribe phosphate binders simultaneously with phosphate restriction in patients on dialysis. (See 'Phosphate binders' above.)

●Preference for non-calcium-containing binders – When phosphate binders are prescribed, for most patients we suggest non-calcium-containing binders (Grade 2B). Studies have suggested lower mortality associated with the use of non-calcium-containing binders compared with calcium-containing binders. However, the increased cost and restricted availability of these agents make calcium-containing binders an acceptable alternative in some circumstances. (See 'Phosphate binders' above.)

●Refractory hyperphosphatemia – For refractory hyperphosphatemia, extended or increased hemodialysis is effective in lowering phosphate; however, this is not a practical option for many patients. Such patients may benefit from a modification of treatment of hyperparathyroidism. In adequately dialyzed patients who have serum phosphate levels persistently >5.5 mg/dL despite dietary restriction, optimal treatment of hyperparathyroidism, and the appropriate use of phosphate binders, we suggest treatment with tenapanor and phosphate binders rather than phosphate binders alone (Grade 2B). However, some experts use higher levels of serum phosphate as a threshold for tenapanor add-on therapy. (See 'Refractory hyperphosphatemia' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Robert E Cronin, MD, and Michael Berkoben, MD, who contributed to earlier versions of this topic review.