Calcium acetate versus calcium carbonate as oral phosphate binder in pediatric and adolescent hemodialysis patients

Calcium carbonate is widely used as an oral phosphorus binder to control hyperphosphatemia in children on maintenance hemodialysis. Intestinal calcium absorption may induce hypercalcemia, particularly if calcitriol is given simultaneously. In adults, calcium acetate binds phosphorus more effectively than calcium carbonate, while reducing the frequency of hypercalcemic events. We therefore compared calcium acetate with calcium carbonate in nine pediatric patients on long-term maintenance hemodialysis. Following a 1-week withdrawal of phosphorus binders, calcium carbonate was administered for 7 weeks; after a second withdrawal, calcium acetate was given for another 7 weeks. All patients received calcitriol regularly. Both agents lowered the serum phosphorus concentration significantly (calcium carbonate 5.7±1.4 vs. 7.7±2.1 mg/dl, P<0.005; calcium acetate 5.8±1.4 vs. 7.8±2.0 mg/dl, P<0.005). Significantly less elementary calcium was ingested with calcium acetate than with calcium carbonate: 750 (375 – 1,500) vs. 1,200 (0 – 3,000) mg calcium/day, P<0.0001. With calcium carbonate serum calcium increased significantly. The number of episodes of hyperphosphatemia or hypercalcemia did not differ between treatments. Intact plasma parathyroid hormone (PTH) decreased significantly with both phosphate binders, and serum 25-hydroxyvitamin D₃ increased. There was a close relationship between serum phosphorus and PTH in prepubertal but not in pubertal patients. We conclude that hyperphosphatemia can be controlled effectively by both calcium acetate and calcium carbonate in pediatric hemodialysis patients. The oral load of elementary calcium is reduced significantly by binding phosphorus with calcium acetate instead of calcium carbonate; nevertheless, hypercalcemic episodes remain equally frequent with both phosphate binders. Received May 9, 1995; received in revised form and accepted February 23, 1996     

Calcium acetate versus calcium carbonate as phosphorus binders in patients on chronic haemodialysis: a controlled study

The first reported double-blind cross-over comparison betweenthe phosphorus binders calcium carbonate and calcium acetatewas undertaken in 15 stable patients on chronic maintenancehaemodialysis. Detailed registration of diet and analysis ofthe protein catabolic rate suggested an unchanged phosphorusintake during the study. It was found that predialytic serumphosphate concentration was significantly decreased by 0.11mmol/1 (0.34 mg/dl) (P= 0.021, 95% confidence limits 0.02–0.21mmol/1; 0.06-0.65 mg/dl) during calcium acetate treatment. Thecalcium phosphate product was insignificantly decreased duringtreatment with calcium acetate whereas we could not excludethe possibility that calcium concentration had increased.     

Calcium acetate versus calcium carbonate in chronic haemodialysis

Summary: The purpose of this study was to compare the efficacy of calcium acetate (CA) and calcium carbonate (CC) as phosphate binders in patients on maintenance haemodialysis. A randomized, blinded, crossover study was conducted for 24 weeks in 32 subjects who were stabilized on in-centre haemodialysis with low calcium (1.25 mmol/L) dialysate, were using calcium carbonate as their primary phosphate binder and had a pre-dialysis serum phosphate level of ± 1.85 mmol/L. Subjects were randomized to receive CA or CC for a period of 12 weeks, followed by the other treatment for 12 weeks; Mylanta II was added as required to achieve adequate phosphate control. There were no differences in mean serum calcium or phosphate during treatment with CA and CC, and serum parathyroid level was unchanged with both. There were four episodes of hypercalcaemia (serum Ca²⁺ > 2.70 mmol/L) while on CA compared with 11 on CC ( P = NS), and seven episodes of hyperphosphataemia (serum P₁± 2.00 mmol/L) on CA compared with 21 on CC ( P = 0.02). Signfficantly more Mylanta II was prescribed with CC than with CA after 8 or more weeks ( P <0.05). the dose of elemental Ca²⁺ ingested with CC was twice that with CA. Calcium acetate was less well tolerated than CC, and five patients withdrew from the trial because of side-effects with CA, and one with CC. In conclusion, among haemodialysis patients who can tolerate it, CA is a superior phosphate binder to CC.     

Calcium acetate versus calcium carbonate as phosphate-binding agents in chronic haemodialysis

Hyperphosphataemia plays a key role in the pathogenesis of renalosteodystrophy, and phosphate-binding agents are required inmany chronic dialysis patients. Aluminium hydroxide and calciumcarbonate are well-established phosphate binders, but theiruse is associated with toxicity or poor efficacy. Calcium acetateis known to be a potent phosphate binder, and has recently beenused successfully in chronic dialysis patients. In this randomizedcross-over trial in 31 chronic haemodialysis patients, equimolardoses of calcium acetate and calcium carbonate were administeredfor 6 weeks each. Compliance was estimated from tablet counts,and biochemical parameters were measured at the end of eachtreatment period. Of the 31 patients 23 completed both treatmentarms; of the remainder, three withdrew due to adverse symptoms,hypercalcaemia necessitated treatment withdrawal in two, andthree died. Non-compliance was significantly higher with acetate(18.3% tablets not taken) than with carbonate (8.7%). Serumphosphate was significantly lower after treatment with acetate(1.51 mmol/l) than with carbonate (1.80), as was the Ca x PO₄product (3.59 vs 4.18 respectively) and PTH (17.8 vs 25.4 pmol/lrespectively). Serum calcium was significantly higher afteracetate therapy (2.40 vs 2.32 mmol/l). No significant differencewas found for sodium, potassium, bicarbonate, urea, creatinine,and haemoglobin. This study confirms that the treatment of hyperphosphataemiais more effective with calcium acetate than with calcium carbonate.For the first time an associated beneficial effect on secondaryhyperpara thyroidism has also been demonstrated. Patient tolerability of calcium acetate was considerably poorer, probablydue in part to tablet formulation and bulkiness, as well aspossible direct gastrointestinal effects of the acetate salt.     

Calcium acetate as a phosphorus binder in hemodialysis patients.

Much interest is currently centered on the use of calcium acetate as a phosphorus binder in patients with renal failure. Therefore, this compound in subjects previously stable on calcium carbonate and undergoing high-efficiency hemodialysis with a dialysate calcium of 2.5 mEq/L was evaluated. Twenty subjects were switched from generic calcium carbonate to a single calcium carbonate preparation for a period of 2 months. This was followed by a phase (1 month) in which calcium acetate was substituted for calcium carbonate at a dose containing half the amount of elemental calcium. Subjects then continued calcium acetate for 6 months. It was found that calcium acetate allowed comparable control of immunoreactive parathyroid hormone, calcium, and phosphorus levels compared with calcium carbonate. This occurred with half the amount of elemental calcium ingested in the form of calcium acetate (349 +/- 25 versus 699 +/- 75 mmol/day; P less than 0.001). With this lower dose, the overall incidence of hypercalcemia was the same with each formulation. In the eight subjects concurrently receiving i.v. calcitriol, the incidence of hypercalcemia was significantly higher during the first month of calcium acetate compared with that in those not receiving this compound (P less than 0.05). Of those four subjects receiving the high dose of calcitriol (2 micrograms thrice weekly), all required either reduction in the dose or discontinuation of the drug. Thus, mineral metabolism could be controlled adequately with calcium acetate despite using half as much elemental calcium compared with calcium carbonate. This, however, did not result in a lower incidence of hypercalcemia, particularly in those receiving i.v. calcitriol.     

Calcium acetate versus calcium carbonate: Phosphate absorption studies in chronic renal failure

Summary: The aim of this study was to compare the alimentary phosphate-binding capacity of calcium acetate to calcium carbonate in stable chronic renal failure patients who were not on haemodialysis. Intestinal absorption of phosphate and calcium was measured on three occasions in five patients with chronic renal failure who were not on maintenance haemodialysis. During each test period they received either no drug, calcium carbonate or calcium acetate (both containing 1g elemental calcium) in a randomized manner, along with a standardized meal. Intestinal contents were recovered after 10h by whole gut lavage, and phosphorus and calcium measured in meal and intestinal contents. Faecal excretion of ingested phosphorus increased from 13.85% in the absence of drug to 29.91% after calcium carbonate administration. Phosphorus excretion was significantly higher after calcium acetate (43.92%) compared to calcium carbonate ( P <0.05). Less calcium was absorbed from calcium acetate than from equimolar amounts of calcium caronate ( P <0.05). In patients with stable renal failure, calcium acetate is a better alimentary phosphate binder than calcium carbonate and binds more phosphorus for each mol of calcium absorbed.     

Comparisons of the effects of calcium carbonate and calcium acetate on zinc tolerance test in hemodialysis patients.

Because aluminum hydroxide, as a phosphate binder, lowered intestinal zinc absorption, we studied the effects of calcium carbonate (CaCO3) and calcium acetate (CaAc), two other phosphate binders, on intestinal Zn absorption in nine patients on hemodialysis and in 11 controls by measuring 1- and 2-hour serum Zn levels after oral administration of 50 mg of elemental Zn as Zn gluconate with or without concomitant administration of 2 g CaCO3 (800 mg elemental Ca) or 3 g CaAc (750 mg elemental Ca). Fasting serum Zn levels were not different between patients and controls (14.0 +/- 2.3 v 14.1 +/- 1.2 mumol/L [91.8 +/- 14.9 v 92.3 +/- 8.0 micrograms/dL]), but the area under the curve of serum Zn increment (AUC) 2 hours after an oral Zn challenge without or with either of two of phosphate binders used was significantly smaller in patients than in controls (P less than 0.05). The AUC after concomitant administration of Zn with CaCO3 did not differ from that of Zn alone in either patients or controls, but it was significantly less in Zn with CaAc than in Zn alone or in Zn with CaCO3 in both groups. The results demonstrate that intestinal Zn absorption after an oral Zn challenge decreased in patients on hemodialysis and concomitant administration of CaAc, but CaCO3 did not decrease intestinal Zn absorption in either group.     

Effectiveness of intestinal phosphate binders in patients maintained by hemodialysis

Two formulations of aluminum phosphate binders-aluminum hydroxide suspension and aluminum carbonate powder in the form of capsules-were found to be equally effective in lowering the plasma phosphate concentration. The effective dose was 4 g/day given in divided doses before meals and at bedtime. Side effects were negligible with this dosage for both preparations, but aluminum carbonate capsules were preferred by most patients because of the greater convenience of the capsule form. Coincident with the lowering of serum phosphorus concentrations, serum parathyroid hormone concentrations decreased approximately 30% from levels obtained while patients were taking placebo.     

The effects of calcium carbonate as the sole phosphate binder in combination with low calcium dialysate and calcitriol therapy in chronic hemodialysis patients.

Alternative phosphate binders, such as CaCO3, have been shown to be effective in the control of phosphate (P) retention in hemodialysis patients (HDP). Additionally, both oral (POC) and iv (IVC) calcitriol are purported to be of benefit in the control of secondary hyperparathyroidism. This investigation was undertaken to determine: (1) the effectiveness of CaCO3 as the sole P binder in combination with low (2.5 mEq/L) Ca dialysate; (2) the effects of discontinuing Al(OH)3 binders on both unstimulated and stimulated Al concentrations; and (3) the comparative parathyroid hormone (PTH) response to both POC and IVC in a large group of hemodialysis patients. One hundred ninety-four HDP completed part 1 of the study where CaCO3 was substituted for Al(OH)3 as the sole P binder for 6 months. A cohort of 49 HDP was given desferoximine (40 mg/kg) initially and 10 months after using CaCO3. In part 2, 54 HDP were given POC and 97 HDP were given IVC in dosages of 0.25 to 0.5 micrograms/day and 1.5 to 6.0 micrograms/wk, respectively, for an additional 6 months. In part 1, Ca and P were not different from baseline values observed with Al(OH)3 therapy. Ionized Ca increased (P < 0.05) and PTH decreased (P < 0.001) during CaCO3 therapy without vitamin D. In part 2, PTH declined 23% with IVC and was unchanged with POC in equivalent dosages (P < 0.05) at 3 months. By 6 months, PTH declined a total of 54% with IVC and was unchanged with POC. Ca, ionized Ca, P, and serum calcitriol were greater (P < 0.05) in the IVC group at 6 months. Serum Al concentrations for the entire 194 HDP fell 65% (P < 0.0001) over 12 months. In the 49 HDP cohort, serum Al fell 43.6% (P < 0.001) and stimulated Al concentrations decreased 68.7% (P < 0.0001) after 10 months. We conclude that: (1) CaCO3 is as effective as Al(OH)3 in controlling P, (2) a small decrease in PTH is observed with CaCO3 alone, (3) serologic evidence of Al excess is virtually eliminated, (4) PTH suppression with IVC is superior to that seen with POC in equivalent doses.     

碳酸镧与传统磷结合剂治疗维持性血液透析患者高磷血症的Meta分析

目的 评价碳酸镧与传统磷结合剂治疗维持性血液透析患者高磷血症的疗效和安全性.方法 计算机检索MEDLINE(1996-2012.12)、EBCO (1996-2012.12)、Cochrane图书馆临床对照试验资料库和中文万方数据库(1996-2012.12).手工检索已发表或未发表的相关文献,包括会议摘要等.检索无语种限制.纳入碳酸镧与传统磷结合剂比较治疗维持性血液透析患者高磷血症的随机对照试验.由两名评价员独立评价纳入研究的质量和提取资料,并用RevMan 5.0软件进行Meta分析.结果 共纳入10项研究.Meta分析结果显示,碳酸镧与传统磷结合剂相比,降低血磷水平的疗效差异无统计学意义[WMD=-0.06,95％ CI(-0.27～0.15),P=0.57],但碳酸镧治疗组血钙水平低于含钙磷结合剂,两组间因不良反应退出情况差异无统计学意义,碳酸镧治疗组高钙血症发生率低于传统磷结合剂.结论 碳酸镧治疗对终末期肾脏疾病维持性血液透析患者高磷血症有效,且其高钙血症发生率低于传统磷结合剂.     

Bicarbonate versus acetate hemodialysis in ventilated patients

Hemodynamic tolerance to bicarbonate versus acetate hemodialysis was studied in seven ventilated, critically ill patients, suffering from acute renal failure. Both kinds of hemodialysis were carried out with a recirculating dialysate delivery system and a relatively low blood flow (180 ml/min). Each patient underwent two hemodialysis procedures, one with bicarbonate and one with acetate, lasting for four hours. Ultrafiltration rates were kept below 250 ml/h and only biocompatible membranes with a relatively small surface area (Biospal 2400, Hospal, France) were used. Despite the mild hemodialysis conditions, hypotensive episodes with a mean blood pressure below 70 mmHg were observed in 3 out of 7 bicarbonate sessions and 4 out of 7 acetate sessions. Thus, we could not demonstrate a hemodynamic advantage of bicarbonate hemodialysis in this group of ventilated patients. This contrasts with other studies conducted in non-ventilated patients. Prevention of hypoxemia by mechanical ventilation and control of vascular tone by the use of vasoactive drugs may be of more clinical relevance than the kind of hemodialysis procedure that is used.     

Calcium acetate control of serum phosphorus in hemodialysis patients

Calcium acetate has many characteristics of an ideal phosphorus binder. It is a readily soluble salt that avidly binds phosphorus in vitro at pH 5 and above. One-dose/one-meal balance studies show it to be more potent than calcium carbonate or calcium citrate. We studied chronic (3-month) phosphorus binding with calcium acetate in 91 hyperphosphatemic dialysis patients at four different centers. All phosphorus binders were stopped for 2 weeks. Calcium acetate at an initial dose of 8.11 mmol (325 mg Ca2+) per meal was then used as the only phosphorus binder. Dose was adjusted to attempt control of predialysis phosphorus level less than 1.78 mmol/L (5.5 mg/100 mL). Final calcium acetate dose was 14.6 mmol (586 mg) Ca2+ per meal. Sixteen patients developed mild transient hypercalcemia (mean, 2.84 mmol/L [11.4 mg/dL]. Initial phosphorus values in mmol/L (mg/dL) were 2.39 (7.4); at 1 month, 1.91 (5.9); and at 3 months, 1.68 (5.2). Initial calcium values in mmol/L (mg/dL) were 2.22 (8.9); at 1 month, 2.37 (9.5); and at 3 months, 2.42 (9.7). Initial aluminum values in mumol/L (micrograms/L) were 2.99 (80.7); and at 3 months were 2.54 (68.4). Initial C-terminal parathyroid hormone (C-PTH) values in ng/mL were 14.6; at 1 month, 11.9; and at 3 months, 13.2. Sixty-nine patients then entered a double-blind study. Phosphorus binders were stopped for 1 week. Calcium acetate (at a dose established in a prior study) or placebo was then administered for 2 weeks. Next, patients were crossed to the opposite regimen for 2 weeks. Initial phosphorus was 2.36 mmol/L (7.3 mg/100 mL) and calcium 2.22 mmol/L (8.9 mg/100 mL).(ABSTRACT TRUNCATED AT 250 WORDS)     

Effects of sevelamer and calcium-based phosphate binders on mortality in hemodialysis patients

Elevated serum phosphorus and calcium are associated with arterial calcification and mortality in dialysis patients. Unlike calcium-based binders, sevelamer attenuates arterial calcification but it is unknown whether sevelamer affects mortality or morbidity. In a multicenter, randomized, open-label, parallel design trial we compared sevelamer and calcium-based binders on all-cause and cause-specific mortality (cardiovascular, infection, and other) in prevalent hemodialysis patients. A total of 2103 patients were initially randomized to treatment and 1068 patients completed the study. All-cause mortality rates and cause-specific mortality rates were not significantly different. There was a significant age interaction on the treatment effect. Only in patients over 65 years of age was there a significant effect of sevelamer in lowering the mortality rate. There was a suggestion that sevelamer was associated with lower overall, but not cardiovascular-linked, mortality in older patients. We suggest that further research is needed to confirm these findings.     

Etiology of hypercalcemia in hemodialysis patients on calcium carbonate therapy

Fourteen of 39 dialysis patients (36%) became hypercalcemic after switching to calcium carbonate as their principal phosphate binder. In order to identify risk factors associated with the development of hypercalcemia, indirect parameters of intestinal calcium reabsorption and bone turnover rate in these 14 patients were compared with results in 14 eucalcemic patients matched for age, sex, length of time on dialysis, and etiology of renal disease. In addition to experiencing hypercalcemic episodes with peak calcium values of 2.7 to 3.8 mmol/L (10.7 to 15.0 mg/dL), patients in the hypercalcemic group exhibited a significant increase in the mean calcium concentration obtained during 6 months before the switch, compared with the mean value obtained during the 7 months of observation after the switch (2.4 +/- 0.03 to 2.5 +/- 0.03 mmol/L [9.7 +/- 0.2 to 10.2 +/- 0.1 mg/dL], P = 0.006). In contrast, eucalcemic patients exhibited no change in mean calcium values over the same time period (2.3 +/- 0.05 to 2.3 +/- 0.05 mmol/L [9.2 +/- 0.2 to 9.2 +/- 0.2 mg/dL]). CaCO3 dosage, calculated dietary calcium intake, and circulating levels of vitamin D metabolites were similar in both groups. Physical activity index and predialysis serum bicarbonate levels also were similar in both groups. However, there was a significant difference in parameters reflecting bone turnover rates between groups.(ABSTRACT TRUNCATED AT 250 WORDS)     

Magnesium carbonate for phosphate control in patients on hemodialysis. A randomized controlled trial

Background Magnesium salts bind dietary phosphorus, but their use in renal patients is limited due to their potential for causing side effects. The aim of this study was to evaluate the efficacy and safety of magnesium carbonate (MgCO₃) as a phosphate-binder in hemodialysis patients. Methods Forty-six stable hemodialysis patients were randomly allocated to receive either MgCO₃ (n = 25) or calcium carbonate (CaCO₃), (n = 21) for 6 months. The concentration of Mg in the dialysate bath was 0.30 mmol/l in the MgCO₃ group and 0.48 mmol/l in the CaCO₃ group. Results Only two of 25 patients (8%) discontinued ingestion of MgCO₃ due to complications: one (4%) because of persistent diarrhea, and the other (4%) because of recurrent hypermagnesemia. In the MgCO₃ and CaCO₃ groups, respectively, time-averaged (months 1–6) serum concentrations were: phosphate (P), 5.47 vs. 5.29 mg/dl, P = ns; Ca, 9.13 vs. 9.60 mg/dl, P < 0.001; Ca × P product, 50.35 vs. 50.70 (mg/dl)², P = ns; Mg, 2.57 vs. 2.41 mg/dl, P = ns; intact parathyroid hormone (iPTH), 285 vs. 235 pg/ml, P < 0.01. At month 6, iPTH levels did not differ between groups: 251 vs. 212 pg/ml, P = ns. At month 6 the percentages of patients with serum levels of phosphate, Ca × P product and iPTH that fell within the Kidney Disease Outcomes Quality Initiative (K/DOQI) guidelines were similar in both groups, whereas more patients in the MgCO₃ group (17/23; 73.91%) than in the CaCO₃ group (5/20, 25%) had serum Ca levels that fell within these guidelines, with the difference being significant at P < 0.01. Conclusion Our study shows that MgCO₃ administered for a period of 6 months is an effective and inexpensive agent to control serum phosphate levels in hemodialysis patients. The administration of MgCO₃ in combination with a low dialysate Mg concentration avoids the risk of severe hypermagnesemia.     

Decrease in thoracic vertebral bone attenuation with calcium-based phosphate binders in hemodialysis.

We performed a post hoc analysis of a 52-week randomized trial conducted in adult hemodialysis patients that compared the effects of calcium-based phosphate binders and sevelamer, a nonabsorbable polymer, on parameters of mineral metabolism and vascular calcification by electron beam tomography. In this analysis, we evaluated the relative effects of calcium and sevelamer on thoracic vertebral attenuation by CT and markers of bone turnover. Subjects randomized to calcium salts experienced a significant reduction in trabecular bone attenuation and a trend toward reduction in cortical bone attenuation, in association with higher concentrations of serum calcium, lower concentrations of PTH, and reduced total and bone-specific alkaline phosphatase. INTRODUCTION: In patients with chronic kidney disease, hyperphosphatemia is associated with osteodystrophy, vascular and soft tissue calcification, and mortality. Calcium-based phosphate binders are commonly prescribed to reduce intestinal phosphate absorption and to attenuate secondary hyperparathyroidism. Clinicians and investigators have presumed that, in hemodialysis patients, calcium exerts beneficial effects on bone. MATERIALS AND METHODS: We performed a post hoc analysis of a 52-week randomized trial conducted in adult hemodialysis patients that compared the effects of calcium-based phosphate binders and sevelamer, a nonabsorbable polymer, on parameters of mineral metabolism and vascular calcification by electron beam tomography. In this analysis, we evaluated the relative effects of calcium and sevelamer on thoracic vertebral attenuation by CT and markers of bone turnover. RESULTS AND CONCLUSIONS: The average serum phosphorus and calcium x phosphorus products were similar for both groups, although the average serum calcium concentration was significantly higher in the calcium-treated group. Compared with sevelamer-treated subjects, calcium-treated subjects showed a decrease in thoracic vertebral trabecular bone attenuation (p = 0.01) and a trend toward decreased cortical bone attenuation. More than 30% of calcium-treated subjects experienced a 10% or more decrease in trabecular and cortical bone attenuation. On study, sevelamer-treated subjects had higher concentrations of total and bone-specific alkaline phosphatase, osteocalcin, and PTH (p < 0.001). When used to correct hyperphosphatemia, calcium salts lead to a reduction in thoracic trabecular and cortical bone attenuation. Calcium salts may paradoxically decrease BMD in hemodialysis patients.     

Ranitidine reduces phosphate binding in dialysis patients receiving calcium carbonate

BACKGROUND: In a previous controlled study we showed that ranitidine significantlyreduced the phosphate binding of aluminium hydroxide in patientswith renal failure, probably by increasing intragastric pH. METHODS: In this study we have investigated the effect of ranitidineon the phosphate binding of calcium carbonate in fifteen dialysispatients. Ranitidine 300 mg or a placebo tablet was taken beforebreakfast for two 4-week periods in a double-blind crossovertrial with no washout period. The mean daily dose of calciumcarbonate was 2 g and neither the dose nor the patient's dietwas changed during the study period. Blood was taken at 2-weeklyintervals for serum phosphate, calcium, albumin, and alkalinephosphatase measurements, and at the end of each treatment periodfor parathyroid hormone (PTH) level. RESULTS: Serum phosphate concentrations were significantly higher duringthe ranitidine than the placebo phase, 1.78 (±0.43 SD)versus 1.59 (±0.49 SD) mmol/1 (P<0.001). Serum calcium,albumin, PTH, and alkaline phosphatase concentrations did notdiffer between the two treatment periods. CONCLUSION: This study shows that ranitidine has a significant adverse effecton the phosphate binding of calcium carbonate in patients withrenal failure.