Impact of lowering dialysate calcium concentration on serum bone turnover markers in hemodialysis patients

Bone 36 (2005) 909 – 916 www.elsevier.com/locate/bone

Impact of lowering dialysate calcium concentration on serum bone turnover markers in hemodialysis patients Takayuki Hamanoa, Susumu Osetoa, Naohiko Fujiia, Takahito Itoa,*, Masaya Katayamab, Masaru Horioa, Enyu Imaia, Masatsugu Horia a

Department of Internal Medicine and Therapeutics, Osaka University Graduate School of Medicine, Box A8, 2-2 Yamada-oka, Suita, Osaka 565-0871, Japan b Daini Rokushima Clinic, Amagasaki, Hyogo, Japan Received 16 November 2004; revised 29 January 2005; accepted 17 February 2005 Available online 24 March 2005

Abstract Loss of renal function perturbs bone metabolism because kidney is a vital organ maintaining homeostasis of calcium and phosphate. In hemodialysis patients, bone diseases are serious complications resulting in fractures and extraosseous calcification. The latest clinical practice guidelines by the National Kidney Foundation (New York, US) recommend a dialysate calcium concentration (D-Ca) of 2.5 mEq/L rather than 3.0 mEq/L to avoid excess calcium load and to prevent subsequent vascular calcification. However, there is no perfect agreement yet about which concentration should be chosen because lowering D-Ca might enhance uncoupled bone resorption. Here, we studied effects of lowering D-Ca from 3.0 to 2.5 mEq/L on bone metabolism in 67 patients. Doses of vitamin D and phosphate binders were kept constant for a 2-month period beginning 1 month before the change in D-Ca, and were adjusted thereafter. In group A [intact parathyroid hormone (iPTH) < 100 pg/ml before the study], serum cross-linked N-terminal telopeptide of type I collagen (NTx) increased immediately after lowering D-Ca and then remained stable. Intact osteocalcin (iOC) increased later along with iPTH, suggesting the improvement of adynamic bone disease which shows a marked decrease in bone turnover without osteoid accumulation. Vitamin D was not dosed up in this group. In group B (100  iPTH < 300), serum NTx increased transiently, which is followed by an increase of iOC but not by a change of iPTH. In group C (300  iPTH), lowering D-Ca allowed us to increase the dose of vitamin D without hypercalcemia, leading to a significant decrease in NTx and iPTH. Overall, serum phosphate increased from 5.4 T 1.6 to 6.1 T 1.6 mg/dL (P < 0.0001) and serum NTx increased by 1.5-fold ( P < 0.0001) 1 month after lowering D-Ca. Over a 3-month period after that, serum phosphate and serum NTx decreased to their basal levels. These indicate that bone resorption predominated over formation for only a short period. In conclusion, a D-Ca of 2.5 mEq/L with adjustment of vitamin D ameliorates metabolic abnormalities of bone which develop under 3.0 mEq/L. D 2005 Elsevier Inc. All rights reserved. Keywords: Renal osteodystrophy; Calcium; Cross-linked N-telopeptides of bone collagen; Bone remodeling

Introduction Kidney plays a central role in maintaining homeostasis of calcium and phosphorus in the body through urinary excretion/reabsorption and other multiple functions including 1a-hydroxylation of 25-hydroxyvitamin D. In end-stage renal disease (ESRD) patients who lost renal function, vitamin D deficiency and hyperphosphatemia lead to T Corresponding author. Fax: +81 6 6879 3639. E-mail address: [email protected] (T. Ito). 8756-3282/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2005.02.008

secondary hyperparathyroidism which enhances bone resorption. Adynamic bone disease displaying a marked decrease in bone turnover without osteoid accumulation is another serious bone disease in ESRD patients. This type of bone disease is partly due to the overuse of vitamin D or overload of calcium from dialysis solution (dialysate). Therefore, the therapeutic regimen of hemodialysis, a widely used modality to purify uremic plasma and to deliver calcium and phosphate between dialysate and plasma, should be optimized to prevent complications in ESRD patients.

910

T. Hamano et al. / Bone 36 (2005) 909 – 916

Recently, the Dialysis Outcomes and Practice Patterns Study I (DOPPS I), a prospective study using data collected from France, Germany, Italy, Japan, Spain, the United Kingdom, and the United States, has shown that elevation of dialysate calcium concentration (D-Ca) by 1 mEq/L causes a 13% elevation of mortality risk in hemodialysis patients [1]. Another recent report indicates that lowering D-Ca improves arterial compliance and decreases aldosterone levels [2]. According to DOPPS II, increased serum calcium concentration (S-Ca) in hemodialysis patients is associated with increased mortality risk, especially mortality due to cardiac events. Moreover, even patients with well-controlled S-Ca are at risk for cardiac calcification, which is closely associated with poor outcomes. In another study, intake of calcium compound to inhibit intestinal absorption of phosphate for the treatment of hyperphosphatemia (phosphate binder) is almost twice as much in patients with calcification as in patients without calcification, even though the two types of patients have similar S-Ca levels [3]. These findings support the hypothesis that high S-Ca and/or excess calcium load are risk factors for life-threatening extraosseous calcification. The new Kidney Disease Outcomes Quality Initiative (K/ DOQI) guidelines, which are evidence-based clinical practice guidelines developed by the National Kidney Foundation (NKF) for all stages of chronic kidney disease and related complications, recommend that corrected S-Ca in patients with ESRD should be maintained at the lower end of the normal range (8.4 – 9.5 mg/dL) [4]. This value is lower than that in previous recommendations. The guidelines also recommend that D-Ca should be 2.5 mEq/L rather than 3.0 mEq/L. However, 2000’s guidelines of the European Renal Association [5] and other reports [6,7] raised the concern that a D-Ca of 2.5 mEq/L may worsen secondary hyperparathyroidism due to negative calcium balance. Another concern is that lowering D-Ca may increase bone resorption which leads to bone mineral density loss [8]. Physicians in Europe and Japan still prefer 3.0 mEq/L rather than 2.5 mEq/L because such concerns are prevalent among physicians in these countries [9]. This disagreement prompted us to assess bone metabolism after lowering DCa according to the K/DOQI guidelines. In the past, there was no useful bone resorption marker for assessment of the dynamic state of bone remodeling in hemodialysis patients. Recently, sensitive assays have been developed to detect collagenous products of breakdown of the organic bone matrix [10]. One of them is serum crosslinked N-telopeptide of type I collagen (serum NTx). It has been reported that serum NTx is useful as a bone resorption marker in hemodialysis patients [11]. Use of NTx and established bone formation markers such as bone-specific alkaline phosphatase (BSAP) and intact osteocalcin (iOC) allows independent assessment of bone resorption and formation, which can provide important

insight into bone remodeling processes without invasive methods. In the present prospective study, we examined the impact of lowering D-Ca from 3.0 to 2.5 mEq/L on bone remodeling by using serum bone turnover markers.

Subjects and methods Study participants Seventy-eight of 105 adult Japanese patients on maintenance hemodialysis in Daini Rokushima Clinic were enrolled in this study after being informed of the plan to change the dialysate according to the K/DOQI guidelines recommending lower D-Ca. We excluded patients on hemodiafiltration, patients who had undergone percutaneous ethanol injection therapy or parathyroidectomy for secondary hyperparathyroidism, and patients who had previously been treated and/or were being treated with glucocorticoid which affects bone metabolism. Eleven of the enrolled 78 patients were excluded because of hospitalization, transfer, or death during the study period. The remaining 67 patients completed a 4-month follow-up, and 63 of those patients completed another 3 months of follow-up for a total of 7 months. Underlying renal diseases of the 67 followed-up patients were as follows: diabetic nephropathy, 30 patients; chronic glomerulonephritis, 22 patients; nephrosclerosis, 4 patients; polycystic kidney disease, 1 patient; chronic pyelonephritis, 1 patient; urate nephropathy, 1 patient; interstitial nephritis, 1 patient; undiagnosed, 7 patients. All the patients were oliguric or anuric (<300 mL/day). No patients were likely to have aluminum-related bone disease because aluminum-containing phosphate binders had not been prescribed to any patients and because water for dialysate was highly purified. The 67 patients who completed the 4-month follow-up comprised 35 males and 32 females (all the females were post-menopausal), who had been on hemodialysis with a D-Ca of 3.0 mEq/L for at least 1 year. Patient characteristics are shown in Table 1. Mean age (TSD) was 67 T 11 years (range, 33 –84 years). Mean duration of hemodialysis treatment was 4.7 T 8.8 years. Hemodialysis was performed 3 times a week for 3 to 4 h per session. A synthetic dialyzer membrane made of cellulose triacetate or polysulfone was used, with the type of membrane used for each patient remaining unchanged throughout the study period. In accordance with the standard practice in Japan, no dialyzer membranes were re-used. Mean Kt/Vurea, a calculated value indicating dialysis efficiency, was 1.22 T 0.16. All the patients gave well-informed consent to participate in this study including additional blood sampling. None of the patients had received estrogen, bisphosphonates, or parathyroid hormone (PTH) within the preceding 12 months or was receiving estrogen, bisphosphonates, or PTH.

T. Hamano et al. / Bone 36 (2005) 909 – 916

911

Table 1 Patient profiles

n (male/female) Age (years) DM (+/ ) Vitamin D (+/ ) Prescribed CaCO3 (g/day) Duration of hemodialysis (years) Kt/Vurea iPTH (pg/mL) Intact osteocalcin (ng/mL) BSAP (U/L) Serum NTx (nM BCE/L) Serum corrected Ca (mg/dL) Serum phosphate (mg/dL) Ca  P product (mg2/dL2) Serum albumin (g/dL) Serum magnesium (mg/dL)

Group A (iPTH < 100)

Group B (100  iPTH < 300)

Group C (300  PTH)

22 (14/8) 65.2 T 1.4 10/12 10/12 2.2 T 1.1 5.1 T 0.4 1.21 T 0.14 52 T 30*** 18 T 7* 17 T 6.5** 67 T 34* 9.2 T 0.6 5.1 T 1.5 46.2 T 13.2 3.8 T 0.3 2.9 T 0.4***

35 (15/20) 68.5 T 12.7 18/17 12/23 2.3 T 1.1 3.2 T 7.1 1.22 T 0.04 170 T 52 25 T 13 25.3 T 12.1 111 T 73 8.9 T 0.6 5.0 T 1.4 44.6 T 13.3 3.6 T 0.3 2.5 T 0.4

10 (6/4) 66.2 T 14.1 4/10 8/2 1.1 T 1.4** 8.8 T 1.4** 1.27 T 0.19 505 T 197*** 74 T 55*** 51.1 T 47.1* 261 T 195* 9.7 T 1.0* 5.9 T 2.2 58.2 T 24.4 3.6 T 0.2 2.6 T 0.3

The patients were divided into 3 groups according to iPTH level at the beginning of the study: group A (iPTH < 100 pg/mL), group B (100 pg/mL  iPTH < 300 pg/mL), and group C (300 pg/mL  iPTH). Each value is expressed as the number of patients or as mean T SD. Statistical significance, compared with group B, is indicated as follows: *P < 0.05; **P < 0.01; ***P < 0.001. BCE, bone collagen equivalents.

Prescription of vitamin D, calcium as phosphate binder At the beginning of the study, the patients were divided into 3 groups according to intact PTH (iPTH) level: group A (iPTH < 100 pg/mL), group B (100 pg/mL  iPTH < 300 pg/mL) and group C (300 pg/mL  iPTH). This grouping is based on the thresholds described in the K/ DOQI guidelines [4]: (1) the guideline 8B (based on evidence) which recommends that hemodialysis patients with iPTH levels >300 pg/mL should receive an active vitamin D sterol, targeting 150 to 300 pg/mL of iPTH; (2) the guideline 13C.1 (as an opinion) which recommends that adynamic bone disease, as determined either by bone biopsy or the iPTH level lower than 100 pg/ml, should be treated by allowing iPTH to rise in order to increase bone turnover. Among the 67 patients, 25 patients had received vitamin D or its analogs such as alfacalcidol, calcitriol, and 22-oxacalcitriol either orally or intravenously for at least 1 year prior to the study. From 1 month before the change of dialysate, the dose of vitamin D analogs and phosphate binders remained constant until 1 month after lowering DCa. Beginning 1 month after lowering D-Ca, we adjusted the dose of vitamin D or its analogs and phosphate binders using an algorithm described below, which had been used in routine practice for at least 1 year before the study. The target level of iPTH was 100 to 300 pg/ml. For patients whose iPTH was <100 pg/mL, the dose of vitamin D or its analogs was null or was not increased until iPTH was >200 pg/mL. When corrected S-Ca was <8.4 mg/dL, we first adjusted the dose of calcium carbonate to 3.0 g/day, and then added vitamin D analogs as needed to maintain an S-Ca of >8.4 mg/dL. For patients whose iPTH was in the range of group B, when alkaline phosphatase increased by >20% (compared to the value 1 month before), we gradually increased the dose of vitamin D or its analogs,

unless corrected S-Ca exceeded 10.5 mg/dL. If alkaline phosphatase did not increase much and serum-corrected calcium decreased to an abnormal level, we increased the dose of calcium carbonate. When S-P was >6.0 mg/dL after adjustment of the dose of vitamin D analogs, calcium carbonate was added as long as corrected S-Ca was <10 mg/dL. For patients whose iPTH was >300 pg/mL, when corrected S-Ca decreased to <10.5 mg/dL, we prescribed vitamin D or its analogs and gradually increased their dose until corrected S-Ca was 11 mg/dL. Fifty-five patients had been receiving calcium carbonate but not calcium acetate. Some patients were administered colestimide, a non-calcium-containing cationic polymer with chloride, whose chemical structure resembles that of sevelamer hydrochloride, a new generation of phosphate binder [12]. Colestimide was administered only to some patients whose iPTH level was >300 pg/ml and could not be controlled by calcium carbonate alone because of hypercalcemia. In such cases, the dosage of colestimide was not changed during the study. No sevelamer hydrochloride or lanthanum carbonate was prescribed, because both drugs were unavailable in Japan at the time of the study. Change of dialysate D-Ca was changed on a specific day from 3.0 mEq/L to 2.5 mEq/L, and was maintained at 2.5 mEq/L thereafter, at Daini Rokushima Clinic (Fig. 1). Dialysate with a D-Ca of 3.0 mEq/L was composed of 140 mEq/L Na, 2.0 mEq/L K, 1 mEq/L Mg, 113 mEq/L Cl, 25 mEq/L bicarbonate, 10.2 mEq/L acetate, and 100 mg/dL glucose. Dialysate with a DCa of 2.5 mEq/L had 3 slight modifications in addition to the difference in calcium concentration (110 mEq/L Cl, 28 mEq/L bicarbonate, and 10 mEq/L acetate).

912

T. Hamano et al. / Bone 36 (2005) 909 – 916

software (Dr. SPSS II, SPSS Inc., Chicago, IL, USA) for Windowsi (Microsoft Co, Redmond, WA, USA). Values are expressed as mean T SE for continuous variables unless otherwise stated. P values of <0.05 were considered to indicate statistical significance.

Results Patient profile

Fig. 1. This is the protocol for changing calcium concentration of dialysate from 3.0 mEq/L to 2.5 mEq/L. The time of the change is designated as 0month. Dosage of vitamin D, calcium, and phosphate binders remained constant from 1-month to 1-month. Thereafter, doses of those agents were adjusted according to the algorithm described in the Subjects and methods section. Some of the subjects were monitored for 7 months after lowering D-Ca.

Blood sampling and laboratory parameters Blood samples were drawn from the arterio-venous fistula immediately before the start of hemodialysis after a 2-day interval in a week. Sampling was not performed at fasting, but all the patients had their blood taken at the same time on a day throughout the study for the comparison. Routine biochemical parameters including alkaline phosphatase were measured monthly. Intact PTH, NTx, the cross-linked carboxyterminal telopeptide of type I collagen (ICTP), iOC, and bone-specific alkaline phosphatase (BSAP) were also monitored 1 month before the change of dialysate ( 1-month), on the day of the change of dialysate (0-month), and 1 (1-month) and 4 months (4month) after lowering D-Ca. Blood was centrifuged at 1000  g for 5 min, and the resultant serum was stored in aliquots at 80-C until used for assays. Serum NTx was measured using an ELIZA kit (OSTEOMARK; Mochida Pharmaceutical Co., Tokyo, Japan). Intact PTH, BSAP, ICTP, and iOC were measured using an Allegro two-site intact PTH immunoradiometric assay (IRMA) kit (Nichols Institute Diagnostics, San Juan Capistrano, CA), an Osteolinks-BAP high-sensitivity diagnostic enzyme immunoassay (EIA) kit (Sumitomo Pharmaceuticals Co., Osaka, Japan), a two-site Pyridionoline ICTP IRMA kit (Orion Diagnostica, Oulunsalo, Finland), and a BGP IRMA kit (Mitsubishi; Mitsubishi Yatron, Tokyo, Japan), respectively. Statistical analysis Comparison of basal data among groups A, B, and C was performed using a non-parametric Mann –Whitney U test. Temporal changes of parameters in each group were analyzed using a non-parametric Wilcoxon’s signed rank test. All the analyses were performed using statistical

Basal data including bone turnover markers and other biochemical parameters are listed in Table 1. There was no significant difference in age among the 3 groups. However, duration of hemodialysis was significantly longer in group C than in groups A and B. Statistical analysis showed significant differences in iOC, BSAP, and serum NTx among the groups. Serum NTx in groups A, B, and C were 67 T 34, 111 T 73, and 261 T 195 nM BCE/L, respectively. S-Ca, S-P, and Ca  P were higher in group C than in groups A and B, but only the difference in corrected S-Ca was statistically significant. Interestingly, patients in group C took significantly smaller dosage of calcium carbonate. Serum magnesium concentration in group A was significantly higher than that of groups B and C. In addition to diabetes mellitus and aging, serum magnesium may be a factor contributing to the low iPTH in group A. Clinical events related to the change of dialysate Immediately after lowering D-Ca, about 9% and 8% of the patients experienced hypotension and cramping of leg muscles, respectively, during hemodialysis sessions. These patients required close monitoring for several sessions, and in some cases required infusion of saline (up to 100 mL) to achieve remission of the symptoms. However, about 1 week after lowering D-Ca, the patients became free of symptoms and received no more infusion. We did not observe any other serious events that might be related to the change of dialysate. Temporal changes of serum calcium, serum phosphate, and the serum calcium – phosphorus product Overall temporal changes of S-Ca, S-P, and Ca  P are shown in Fig. 2. After lowering D-Ca, corrected S-Ca after a dialysis session significantly decreased (P < 0.0001) (Fig. 2D). This finding confirmed that the change of D-Ca from 3.0 mEq/L to 2.5 mEq/L resulted in less amount of calcium delivery from dialysate during each session. Despite the adjustment of vitamin D analogs and calcium carbonate, corrected S-Ca before a hemodialysis session decreased gradually from 9.4 T 0.7 to 9.1 T 0.6 mg/dL 3 months after lowering D-Ca (P < 0.001) (Fig. 2A). One month after lowering D-Ca when there was no adjustment of vitamin D or calcium, S-P significantly

T. Hamano et al. / Bone 36 (2005) 909 – 916

913

Fig. 2. Time courses of serum calcium concentration, serum phosphate concentration, and the serum calcium – phosphorus product are shown. Each bar represents the mean T SD. (A) Corrected serum calcium concentration (mg/dL) before a hemodialysis session; (B) serum phosphate concentration (mg/dL) before a hemodialysis session; (C) the calcium  phosphorus product; (D) corrected serum calcium concentration (mg/dL) after a hemodialysis session. *P < 0.05; **P < 0.001; ***P < 0.0001, compared with the value at 0-month. The black arrows indicate the period during which the doses of vitamin D and calcium carbonate were adjusted.

increased from 5.4 T 1.6 to 6.1 T 1.6 mg/dL (P < 0.0001). Over the following 3 months, S-P gradually returned to the basal level (Fig. 2B). The Ca  P product transiently increased and then returned to the basal level over a 3-month period, and thereafter remained at the basal level (Fig. 2C). Subgroup analysis revealed that this trend occurred in every group, even in group C (in which the dose of vitamin D was increased [data not shown]). Temporal change of iPTH Four months after lowering D-Ca, the number of patients with optimal iPTH level (100  iPTH < 300) increased from 35 to 42 out of 67 patients (Fig. 3). This relevant effect was achieved in different ways in different groups. In group A, iPTH had increased from 65 T 47 to 109 T 61 pg/mL 4 months after lowering D-Ca (P < 0.01) (Fig. 3). Doses of vitamin D and calcium carbonate were increased in 0% (0/22) and 27% (6/22) of group A patients, respectively. Because iPTH never exceeded 200 pg/mL, there was no need to increase the dosage of vitamin D. However, 4 months after lowering D-Ca, we increased the dosage of calcium carbonate in 6 of the 22 in this group to compensate for the decrease of S-Ca. In this group, iOC had increased significantly (by 50%) in parallel with iPTH 4 months after lowering D-Ca. These data imply that a D-Ca of 2.5 mEq/L stimulated bone remodeling in apparently hypodynamic or a

dynamic bone disease. In group B, no elevation of iPTH was observed (Fig. 3). Doses of vitamin D and calcium carbonate were increased in 40% (14/35) and 17% (6/35) of group B patients, respectively. In group C, iPTH markedly decreased from 517 T 89 to 250 T 137 pg/mL (P < 0.01) (Fig. 3). This effect was presumably due to vitamin-Ddependent suppression of parathyroid hormone because vitamin D or its analogs were dosed up in 80% (8/10) of the group C patients. The dose of calcium carbonate was increased in 20% (2/10) of group C patients. Four months after lowering D-Ca, we measured serum ionized calcium and iPTH before and after a session of hemodialysis to estimate calcium net balance. In 15 out of 67 patients, post-iPTH was lower than pre-iPTH (data not shown). Pre-serum ionized calcium was 2.26 T 0.21 mEq/L in these 15 patients. By contrast, patients whose post-iPTH was greater than pre-iPTH showed 2.46 T 0.17 mEq/L of pre-serum ionized calcium. According to literature [6], the former cases get calcium during a hemodialysis session (positive balance) but the latter lose calcium during a hemodialysis session (negative balance). Calcium net balance during a dialysis session seems to depend not only on D-Ca but also on S-Ca before a session. Temporal changes of bone turnover markers Fearing that a D-Ca of 2.5 mEq/L might cause imbalanced bone resorption, we monitored changes of bone

914

T. Hamano et al. / Bone 36 (2005) 909 – 916

Fig. 3. Temporal change of iPTH (pg/mL) in groups A (iPTH [before the change] < 100), B (100  iPTH [before the change] < 300), and C (300  iPTH [before the switch]). (A) Group A; (B) group B; (C) group C. *P < 0.05; **P < 0.01, compared with the value at 0-month.

turnover markers such as serum NTx (a bone resorption marker), and iOC and BSAP (bone formation markers). Sixty-three patients completed an extra 3 months of followup, for a total of 7 months (Figs. 4 –6). Temporal changes of serum NTx was most dynamic after lowering D-Ca when compared to iOC and BSAP. In all groups, serum NTx and BSAP converged to the ideal ranges (i.e., neither too high nor too low) (Figs. 4, 6). In group A, serum NTx increased 1 month after lowering D-Ca (P < 0.01) (Fig. 4), and thereafter remained stable. In contrast to the sharp and sensitive change of serum NTx, significant increases of iOC and BSAP were not observed until 4 months after lowering D-Ca (P < 0.05) (Figs. 5, 6). In group B, serum NTx increased 1 and 4 months after lowering D-Ca (P < 0.05), and returned to the basal level 7 months after (Fig. 4). An increase of serum NTx preceded that of iOC which reached its peak 4 months after lowering D-Ca and subsequently returned to the basal level (Fig. 5). BSAP did not show any significant change (Fig. 6). In

Fig. 4. Temporal change of serum NTx in each group. (A) Group A; (B) group B; (C) group C. *P < 0.05; **P < 0.01, compared with the value at 0-month in each group.

Fig. 5. Temporal change of intact osteocalcin (iOC) in each group. (A) Group A; (B) group B; (C) group C. *P < 0.05; **P < 0.01, compared with the value at 0-month in each group.

group C, serum NTx abruptly decreased 4 months after and then remained stable (Fig. 4). Although BSAP decreased 7 months after, changes of iOC and BSAP were not statistically significant (Figs. 5, 6). In contrast to serum NTx, levels of another collagen breakdown product, ICTP, had not changed in any of the groups 1 month after lowering D-Ca (data not shown). Bone resorptive status after lowering D-Ca, as indicated by serum NTx and S-P Overall, S-P significantly increased from 5.4 T 1.6 to 6.1 T 1.6 mg/dL 1 month after lowering D-Ca (P < 0.0001) (Fig. 7A). Serum NTx also increased by 1.5-fold for both males and females 1 month after lowering D-Ca (P < 0.0001) (Fig. 7B). In hemodialysis patients with no functional kidney, S-P is governed by oral phosphorus intake, dialysis dose, and release of phosphorus from bone. Therefore, it is likely that imbalanced bone resorption mainly accounts for the increase of S-P, because diet and

Fig. 6. Temporal change of bone-specific alkaline phosphatase (BSAP) in each group. (A) Group A; (B) group B; (C) group C. *P < 0.05, ***P < 0.001, compared with the value at 0-month.

T. Hamano et al. / Bone 36 (2005) 909 – 916

Fig. 7. Temporal change of serum phosphate and serum NTx in all the subjects. (A) Serum phosphate; (B) serum NTx. *P < 0.0001, compared with the value at 0-month.

dialysis menu were not changed throughout the study and because the doses of vitamin D and calcium compound were not changed until 1 month after lowering D-Ca. As doses of vitamin D and calcium carbonate were adjusted from 1 month after lowering D-Ca, S-P gradually decreased to the basal level in parallel with serum NTx (Fig. 7).

Discussion Our results show that a D-Ca of 2.5 mEq/L in combination with adjustment of vitamin D improves bone metabolism. One of beneficial points in group C is that this D-Ca allows us to prescribe a greater dose of active vitamin D without hypercalcemia, leading to improvement of hyperparathyroidism. Intact PTH significantly increased in group A, where bone turnover markers suggested the existence of adynamic bone or low-turnover bone disease (Fig. 3). This low bone turnover might come from the past hemodialysis therapy against a D-Ca of 3.0 mEq/L because a D-Ca of 3.0 mEq/L could maintain iPTH below the optimal range even without vitamin D by delivering calcium directly to the circulation in each hemodialysis session. There is controversy over the use of 2.5 mEq/L in patients with secondary hyperparathyroidism. Dialysate with a D-Ca of 2.5 mEq/L aggravates secondary hyperparathyroidism in both hemodialysis and continuous ambulatory peritoneal dialysis (CAPD) [6,7]. Alternatively, the introduction of 2.5 mEq/L lowers the incidence of parathyroidectomy as consumption of alfacalcidol increases [13]. With CAPD, a low-calcium peritoneal dialysate

915

supplemented with vitamin D remits high-turnover bone disease [14]. Doses of vitamin D seem to be responsible for these discrepancies among reports. ESRD patients are short of active vitamin D because dermal synthesis of vitamin D is reduced [15] and because intra-renal 1a-hydroxylation of 25-hydroxyvitamin D is impaired. A D-Ca of 3.0 mEq/L does not allow a sufficient dose of vitamin D because of a risk of hypercalcemia (group C) or of adynamic bone disease (groups A and B). Active vitamin D can reduce myocardial hypertrophy in hemodialysis patients, especially in those with hyperparathyroidism [16]. Vitamin D can lower the risk of cardiovascular mortality in hemodialysis patients [17]. Indeed, vitamin D receptor null mice fed with a highcalcium diet do not have serum electrolyte abnormality or bone metabolic abnormality, but have cardiac hypertrophy and hypertension due to an augmented renin – angiotensin system [18]. Thus, increased amount of vitamin D, which comes with the introduction of 2.5 mEq/L, might prevent cardiovascular complications in our patients. In the present study, loss of calcium during hemodialysis occurred early after the introduction of 2.5 mEq/L but dissolved over time. Response of bone turnover markers (Figs. 4 – 6) suggests that lowering D-Ca stimulates bone remodeling. Serial changes of bone turnover markers are consistent with reports that bone remodeling always begins with the bone resorption phase lasting about 1 month and is followed by the bone formation phase lasting for about 3 months [19,20]. Serum NTx reflecting earlier and/or subtler changes in bone remodeling of hemodialysis patients would be suitable for quick decision making in clinical practice. We conducted a prospective open-label clinical study of ESRD patients on maintenance hemodialysis. Analysis of serum bone turnover markers showed that new bone resorption is initiated by lowering D-Ca but is transient with the assistance of vitamin D. Lowering D-Ca from 3.0 mEq/L to 2.5 mEq/L along with the adjustment of vitamin D helped us correct abnormal secretion of PTH and abnormal bone turnover in hemodialysis patients.

Acknowledgment This work is supported in part by Chugai Pharmaceutical Co., Ltd. (Tokyo, Japan).

References [1] Young EW, Albert JM, Satayathum S, Goodkin DA. Predictors and consequences of altered mineral metabolism among hemodialysis patients in the Dialysis Outcomes and Practice Patterns Study (DOPPS). Kidney Int 2005;67:1179 – 87. [2] Yoo SJ, Oh DJ, Yu SH. The effects of low calcium dialysate on arterial compliance and vasoactive substances in patients with hemodialysis. Korean J Intern Med 2004;19:27 – 32.

916

T. Hamano et al. / Bone 36 (2005) 909 – 916

[3] Goodman WG, Goldin J, Kuizon BD, Yoon C, Gales B, Sider D, et al. Coronary-artery calcification in young adults with end-stage renal disease who are undergoing dialysis. N Engl J Med 2000;342: 1478 – 83. [4] National Kidney Foundation: K/DOQI clinical guidelines for bone metabolism and disease in chronic kidney disease. Am J Kidney Dis 2003;42(Suppl 3):S1 – 202. [5] Cannata-Andıˆa J, Passlick-Deetjen J, Ritz III E. Clinical algorithms on renal osteodystrophy. Nephrol Dial Transplant 2000;15(Suppl 5): 39 – 57. [6] Yokoyama K, Kagami S, Ohkido I, Kato N, Yamamoto H, Shigematsu T, et al. The negative Ca(2+) balance is involved in the stimulation of PTH secretion. Nephron 2002;92:86 – 90. [7] Duncan R, Cochrane T, Bhalla C, Michael J, Richards NT, Adu D. Low calcium dialysate and hyperparathyroidism. Perit Dial Int 1996;16(Suppl 1):S499 – 502. [8] Sanchez Perales MC, Garcia Cortes MJ, Borrego FJ, Fernandez Martinez S, Borrego J, Perez de lBarrio P, et al. Hemodialysis with 2.5 mEq/L of calcium in relative hypoparathyroidism: long-term effects on bone mass. Nefrologia 2000;20:254 – 61. [9] Malberti F, Ravani P. The choice of the dialysate calcium concentration in the management of patients on haemodialysis and haemodiafiltration. Nephrol Dial Transplant 2003;18(Suppl 7):vii37 [discussion vii57]. [10] Bonde M, Garnero P, Fledelius C, Qvist P, Delmas PD, Christiansen C. Measurement of bone degradation products in serum using antibodies reactive with an isomerized form of an 8 amino acid sequence of the C-telopeptide of type I collagen. J Bone Miner Res 1997;12:1028 – 34. [11] Montagnani A, Gonnelli S, Cepollaro C, Mangeri M, Martini S, Franci MB, et al. A new serum assay to measure N-terminal fragment

[12]

[13]

[14]

[15] [16]

[17]

[18]

[19]

[20]

of telopeptide of type I collagen in patients with renal osteodystrophy. Eur J Intern Med 2003;14:172 – 7. Date T, Shigematsu T, Kawashita Y, Satake N, Morita K. Colestimide can be used as a phosphate binder to treat uraemia in end-stage renal disease patients. Nephrol Dial Transplant 2003;18(Suppl 3):iii90 – 3. Heaf JG, Lokkegard H. Parathyroid hormone during maintenance dialysis: influence of low calcium dialysate, plasma albumin and age. J Nephrol 1998;11:203 – 10. Carozzi S, Nasini MG, Santoni O, Pietrucci A. Low- and highturnover bone disease in continuous ambulatory peritoneal dialysis: effects of low-Ca2+ peritoneal dialysis solution. Perit Dial Int 1993;13(Suppl 2):S473 – 5. Holick MF. Vitamin D and the kidney. Kidney Int 1987;32:912 – 29. Park CW, Oh YS, Shin YS, Kim CM, Kim YS, Kim SY, et al. Intravenous calcitriol regresses myocardial hypertrophy in hemodialysis patients with secondary hyperparathyroidism. Am J Kidney Dis 1999;33:73 – 81. Shoji T, Shinohara K, Kimoto E, Emoto M, Tahara H, Koyama H, et al. Lower risk for cardiovascular mortality in oral 1alpha-hydroxy vitamin D3 users in a haemodialysis population. Nephrol Dial Transplant 2004;19:179 – 84. Li YC, Kong J, Wei M, Chen ZF, Liu SQ, Cao LP. 1,25Dihydroxyvitamin D(3) is a negative endocrine regulator of the renin – angiotensin system. J Clin Invest 2002;110:229 – 38. Eriksen EF, Melsen F, Mosekilde L. Reconstruction of the resorptive site in iliac trabecular bone: a kinetic model for bone resorption in 20 normal individuals. Metab Bone Dis Relat Res 1984;5:235 – 42. Eriksen EF, Gundersen HJ, Melsen F, Mosekilde L. Reconstruction of the formative site in iliac trabecular bone in 20 normal individuals employing a kinetic model for matrix and mineral apposition. Metab Bone Dis Relat Res 1984;5:243 – 52.