Biochemical markers of renal osteodystrophy in pediatric

Kidney International, Vol. 45 (1994), pp. 253 258 Biochemical markers of renal osteodystrophy in pediatric patients undergoing CAPD/CCPD ISIDRO B. SALUSKY, JORGE A. RAMIREZ, WILLIAM OPPENHEIM, BARBARA GALES, GINO V. SEGRE, and WILLIAM G. GOODMAN Departments of Pediatrics, Radiology, Medicine and Orthopedics, UCLA School of Medicine, Los Angeles, California; and Endocrine Unit, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA Biochemical markers of renal osteodystrophy in pediatric patients undergoing CAPD/CCPD. Serum intact PTH [1 84] levels were evaluated as a potential non-invasive method for the diagnosis of renal osteodystrophy in children treated with CAPD/CCPD. Sixty-eight bone biopsy samples were obtained from 55 patients, aged 13 5 (X SD) years, undergoing CAPD/CCPD for 29 13 months; osteitis fibrosa was present in 34 cases, mild lesions of secondary hyperparathyroidism in six, 15 had adynamic lesions, and 13 were classified as normal histology. Serum calcium levels were higher in patients with adynamic bone or normal bone histology than in those with secondary hyperparathyroidism, whereas serum phosphorus, alkaline phosphatase and PTH levels were greater in patients with osteitis fibrosa. The combination of a serum PTH level >200 pg/ml and a serum calcium value <10 mgldl was 85% sensitive and 100% specific for identifying patients with high-turnover lesions of bone. Serum PTH values <200 pg/ml were 100% sensitive but only 79% specific for patients with adynamic bone; specificity increased to 92%, however, using the combined criteria of a PTH level <150 pglml and a serum calcium level >10 mgldl. Higher serum calcium levels and serum PTH values within or below the normal range characterize patients with the adynamic lesion of renal osteodystrophy. Serum PTH levels of approximately 200 pg/mi are useful for distinguishing patients with low-turnover lesions of renal osteodystrophy from those with secondary hyperparathyroidism. Secondary hyperparathyroidism develops in most patients with chronic renal failure. Accordingly, osteitis fibrosa has been the most common histologic lesion of bone both in adults and in children undergoing hemodialysis or peritoneal dialysis [1 4]. Recent reports suggest, however, that the prevalence of lowturnover skeletal disorders, in particular the adynamic or aplastic lesion of renal osteodystrophy, has increased substantially. Thus, as many as 66% of adult patients receiving peritoneal dialysis have evidence of adynamic bone on histologic evaluation [5], and a substantial proportion of these do not have diabetes or evidence of bone aluminum deposition, factors previously recognized as causes of the adynamic lesion in patients with chronic renal failure [6, 7]. Although the long-term consequences of adynamic bone have yet to be fully characterized, hypercalcemia commonly occurs in this disorder, particularly during treatment with calcium Received for publication April 23, 1993 and in revised form July 27, 1993 Accepted for publication July 29, 1993 1994 by the International Society of Nephrology carbonate [5, 81. This cause of hypercalcemia in patients undergoing regular dialysis must be distinguished from that due to severe secondary hyperparathyroidism; parathyroidectomy may be required to correct hypercalcemia in patients with advanced osteitis fibrosa [3], whereas parathyroid surgery is not indicated in patients with the adynamic lesion of renal osteodystrophy since serum PTH levels are relatively low in this condition. Bone biopsy is currently the only reliable method for distinguishing pediatric patients with advanced secondary hyperparathyroidism from those with adynamic lesions of bone [2, 5]. Several reports have documented the value of PTH measurements using the immunoradiometric (IRMA) assay for intact hormone in adults undergoing dialysis [5, 9, 10]. Only one study has been done, however, using this assay in children receiving hemodialysis [11], and no data are available for children undergoing peritoneal dialysis. In the current study, the serum levels of intact PTH, calcium and alkaline phosphatase were evaluated as potential non-invasive predictors of renal osteodystrophy in children undergoing CAPD/CCPD. Methods Histomorphometric and biochemical data from all pediatric patients treated by CAPD or CCPD who had undergone bone biopsy at this institution to assess renal osteodystrophy between 1983 and 1992 were reviewed. From a total of 156 biopsy specimens, concurrent determinations of serum parathyroid hormone (PTH) levels using the immunoradiometric assay (IRMA) for intact hormone were available in 71 specimens. This subset of patients was selected for further analysis after one patient with aluminum-related bone disease and two patients with mixed lesions of renal osteodystrophy were excluded because of the small number of cases with these two disorders. None of the patients with osteomalacia had a concurrent PTH determination by IRMA assay. Sixty-eight biopsies were available from 55 patients, 29 boys and 26 girls; the mean age of patients at the time of bone biopsy was 13 5 years, and they had been treated with CAPD or CCPD for 29 13 months. All patients used dialysate solutions (Dianeal, Baxter, Deerfield, Illinois, USA) containing 3.5 meq/ liter of calcium, and glucose concentrations were adjusted according to the needs for ultrafiltration. All but eight patients 253

254 Salusky et a!: Biochemical markers of bone disease were receiving calcium carbonate exclusively as a phosphatebinding agent at the time of biopsy; eight patients had been given a combination of aluminum hydroxide and calcium carbonate. Although most patients had been treated with daily doses of oral calcitriol for the management of renal osteodystrophy, four patients had received intermittent doses of oral calcitriol three times per week, and three others had been given thrice weekly doses of calcitriol intraperitoneally. In all cases, the dose of calcitriol was adjusted to maintain serum calcium levels between 10.5 to 11.0 mg/dl [12]. All studies were approved by the UCLA Human Subject Protection Committee, and informed consent was obtained from all patients and/or their parents. Bone biopsies were obtained from the anterior iliac crest using a modified Bordier trephine after double tetracycline labeling as previously described [2]. Serum samples for measurements of calcium, phosphorus, alkaline phosphatase, and PTH levels were obtained at the time of bone biopsy. Bone histomorphometry Quantitative histomorphometry of bone was done as previously described [2], and the terminology established by the Nomenclature Committee of the American Society for Bone and Mineral Research was used for all results [13]. All histomorphometric data refer to measurements in two dimensions [13]. Reference values for normal subjects were obtained from iliac crest bone biopsies done in 27 normal children between the ages of 2.5 and 17 years who had undergone elective orthopedic or urologic surgical procedures. None of these subjects had clinical or biochemical evidence of metabolic bone disease, diabetes, or renal insufficiency. All were physically active according to age [2], and none had a history of treatment with corticosteroids, anticonvulsant medications, or vitamin D preparations. Patients with renal osteodystrophy were classified according to the following histomorphometric criteria based upon the reference values established in normal children: osteitis fibrosa osteoid area <12%, evidence of tissue fibrosis and bone formation rate (BFR, tissue referent) >613 tm2/mm2/day; mild lesion of secondary hyperparathyroidism osteoid area <12%, no tissue fibrosis and BFR >613 prn2/mm2/day; normal bone histology no fibrosis, osteoid volume >1.0% but <12% and BFR >97 but <613 m2/mm2/day; adynamic bone osteoid area <12%, no fibrosis and BFR <97 prn2/mm2/day; osteomalacia osteoid area >12%, no fibrosis and BFR <97 jsm2/mm2/ day; mixed lesion osteoid area >15% and fibrosis present [2]. Serum calcium, phosphorus and alkaline phosphatase levels were determined using a Technicon Autoanalyzer II. Serum PTH was measured by the immunoradiometric assay described by Nussbaum et al [14]; values for subjects with normal renal function range from 10 to 65 pg/ml using this assay. Statistical analysis All results are presented as the mean one standard deviation, Statistical analysis of the data was done using one-way analysis of variance with contrasts and chi-square analysis [6]. Sensitivity, specificity and positive predictive value (PPV) were determined for selected levels of PTH in serum for several types of renal bone disease. Sensitivity is defined as the ability of a biochemical parameter to identify patients with selected Table 1. Patient characteristics and serum biochemical determinations according to histologic subtype of renal osteodystrophy Osteitis Normal Adynamic/ fibrosa Mild histology aplastic (N = 34) (N = 6) (N = 13) (N = 15) Age 15±4 10±7 13±5 12±4 years Duration ofccpd 30 26 14 12 44 36 20 17 months Calcium 8.9 0.9 9.0 1.4 9.9 1,0b 10.2 0,7b mg/dl Albumin 3.7 0.5 3.9 0.3 3.9 0.4 3.8 0.4 g/dl Phosphorus 6.6 1.0w 5.2 1.4 5.7 1.4 5.7 1.9 mg/dl Alkaline 370 231" 218 181 169 67 102 40 phosphatase lu/liter PTH [1-84] 678 402C 276 ll4 180 141 58 52 pg/mi a P 0.01 vs. normal histology b p < 0.01 vs. osteitis fibrosa and mild P 0.001 vs. mild, adynamic and normal histology "P < 0.01 vs. adynamic and normal histology P < 0.05 vs. adynamic histologic lesions of bone within a population; specificity represents the ability of a biochemical test to exclude those without a given skeletal lesion. Positive predictive value (PPV) defines the probability that patients with defined biochemical criteria actually have a given disorder [15]. Results Of the 68 bone biopsies obtained with concurrent measurements of serum intact PTH, there was histologic evidence of osteitis fibrosa in 34 cases, and mild lesions of secondary hyperparathyroidism were found in six cases; adynamic lesions of renal osteodystrophy were present in 15 biopsies, and thirteen were classified as normal histology (Table 1). Patients with osteitis fibrosa were older than those with other histologic lesions of renal osteodystrophy, whereas patients with normal bone histology had been treated with CCPD for the longest period (Table 1). Only two patients had undergone parathyroidectomy in the past. A total of 17 patients had undergone prior renal transplantation; of these, nine had osteitis fibrosa, four had adynamic lesions, and four had normal bone histology. There were no differences among groups in the number of patients who had undergone previous parathyroidectomy or renal transplantation (P = 0.582, Fisher's exact test). Patients with adynamic bone and those with normal bone histology had higher serum calcium levels than patients with secondary hyperparathyroidism, whereas serum albumin levels did not differ among groups (Table 1). Serum phosphorus and alkaline phosphatase levels were greater in patients with osteitis fibrosa than in any of the other histologic subgroups (Table 1). On average, serum PTH levels were approximately ten times the upper limit of normal in patients with overt osteitis fibrosa, whereas values were less markedly elevated in those with mild lesions of secondary hyperparathyroidism (Table 1). Serum PTH levels did not differ substantially between patients with mild lesions and those with normal bone histology. For the 15

Salusky et a!: Biochemical markers of bone disease 255 2400 1500 600 400 -L 300 II E Cl) 200 100 0 I I OF M NH AP Fig. 1. Serum PTH 1-84 levels according to histologic sub-type of renal osteodytrophy in pediatric patients treated with CAPD/CCPD. Abbreviations are: OF, osteitis; M, mild lesion; NH, normal bone histology; AP, aplastic lesion. Solid lines represent the upper and lower limit of the assay for PTH. Table 2. Prevalence of histologic types of renal osteodystrophy in the current study and in patients biopsied before and after January 1987 Osteitis Normal Adynamic/ Mixed fibrosa Mild histology aplastic Osteomalacia lesion Total 1983 1992 71 26 29 21 6 3 (N 156) (46%) (17%) (19%) (13%) (4%) (2%) 1983 1986 25 15 10 7 6 0 (N = 63) (40%) (24%) (16%) (11%) (9%) (0%) 1987 1992 46 11 19 14 0 3 (N = 93) (50%) (12%) (20%) (15%) (0%) (3%) Current study 34 6 13 15 0 0 (N 68) (50%) (9%) (19%) (22%) (0%) (0%) patients with adynamic skeletal lesions, serum PTH levels were within the normal range in ten patients, whereas values fell below the lower limit of normal in one (Table 1, Fig. 1). Overall, the distribution of histologic subtypes of renal osteodystrophy in the current study was similar to that seen when all biopsy specimens obtained during the past 10 years in pediatric patients receiving CAPD/CCPD were considered (Table 2). The prevalence of histologic lesions for patients undergoing biopsy during the first and second five year intervals, that is, before or after January of 1987, also did not differ substantially (Table 2). Prior to 1987, 11% of biopsy samples demonstrated adynamic lesions of renal osteodystrophy, whereas 15% of biopsy samples had adynamic lesions after this date. The combined prevalence of adynamic bone and normal bone histology was 27% before January of 1987 and 35% thereafter. Only one patient evaluated after January of 1987 had aluminum deposition in bone documented by histochemical staining methods, and none had aluminum-related bone disease as judged by criteria previously reported from this laboratory [2]. Several biochemical determinations, including the serum levels of calcium, phosphorus and PTH, were evaluated as non-invasive predictors of the type of renal bone disease. Using the results for PTH alone, values greater than 200 pg/mi were 95% sensitive and 86% specific and for identifying patients with high-turnover lesions of bone due to secondary hyperparathyroidism (Table 3); the PPV for a serum PTH level above 200 pg/mi was 90%. When combined with a serum calcium level less Table 3. Predictive value of serum PTH and calcium levels for the diagnosis of high-turnover lesions Criteria Positive predictive Sensitivity Specificity value % PTH > 200 pg/mi (N = 38) 95 86 91 VFH > 250 pg/mi (N = 36) 90 89 92 PTH> 300 pg/mi (N = 34) 85 93 94 PTH> 200 pg/mi and 85 100 100 S0 < 10 mg/di (N = 34) PTH >200 pg/mi and 65 100 100 Sca < 9.5 mg/dl (N = 26) PTH >200 pg/mi and 43 100 100 Sca < 9.0 mg/dl (N = 17) N refers to the number of patients conforming to the specified criteria; Sca is serum calcium. than 10 mg/dl, the sensitivity of serum PTH levels above 200 pg/mi decreased to 85% as a marker of high-turnover skeletal lesions, but the two criteria together excluded all patients with normal or reduced rates of bone formation (Table 3); thus, PPV was 100% for this combination of test results. In contrast, serum alkaline phosphatase levels above 250 lu/liter were 96% specific but only 55% sensitive in predicting lesions of secondary hyperparathyroidism (Table 4). For patients with the adynamic lesion of renal osteodystrophy, serum PTH values less than 200 pg/mi were 100% sensitive but only 79% specific as an isolated diagnostic criterion; the PPV for a serum PTH level less than 200 pg/mi was 58% (Table 5). Specificity was enhanced to 89%, however, when the screening criteria included both a serum PTH level less than 200 pg/mi and a serum calcium level above 10 mgldl (Table 5). Specificity for the adynamic lesion of renal osteodystrophy was further increased using the combined criteria of a PTH level less than 150 pg/mi and a serum calcium level greater than 10 mg/dl, and the PPV increased to 82%. This approach correctly identified 11 of 15 patients with adynamic lesions and it excluded all

256 Salusky et al: Biochemical markers of bone disease Table 4. Predictive value of serum alkaline phosphatase determinations for high- and low-turnover lesions of bone Criteria Positive Sensitivity Specificity predictive value % High-turnover AP> 250 lu/liter 55 96 96 (N 22) AP> 200 lu/liter 65 82 82 (N = 26) AP> 150 lu/liter 77 61 61 (N=31) Low-turnover AP < 200 lu/liter 100 58 58 (N = 15) AP < 150 lu/liter 80 75 75 (N= 12) AP < 100 lu/liter 60 90 90 (N = 9) N refers to the number of patients conforming to the specified criteria. Table 5. Predictive value of serum PTH and calcium levels for the diagnosis of low-turnover lesions Criteria Positive Sensitivity Specificity predictive value % PTH < 200 pg/mi 100 79 58 (N= 15) PTH < 150 pg/mi 93 89 70 (N = 14) PTH < 100 pg/mi (N = 12) 80 92 75 PTH < 200 pg/mi and 60 89 60 Sca> 10 mg/dl (N = 9) PTH < 150 pg/nil and Sea> 10 mg/dl (N =9) 80 92 82 N refers to the number of patients conforming to the specified criteria; S is serum calcium. patients with either normal bone histology or secondary hyperparathyroidism (Table 5), By comparison, serum alkaline phosphatase levels less than 200 lu/liter were 100% sensitive but only 58% specific for the diagnosis of adynamic lesions (Table 4). Discussion In the past, serum biochemical determinations have been poor predictors of bone histology in pediatric patients undergoing maintenance dialysis [2, 16 19]. Although markedly elevated serum PTH levels generally suggest the presence of advanced secondary hyperparathyroidism, the use of different PTH assays with varying degrees of specificity for the intact hormone and carboxy-terminal PTH fragments can render the interpretation of results difficult; many studies have found considerable overlap among serum PTH values using midregion or carboxy-terminal assays in patients with various histologic subtypes of renal osteodystrophy [10, 12]. In contrast, PTH determinations using amino-terminal assays have generally correlated well with bone histomorphometric measurements [20], and they have also been shown to correspond to varying degrees of renal functional impairment [21]. Two-site immunoradiometric (IRMA) assays that measure intact parathyroid hormone are now widely available. These assays have high precision, are relatively free of serum effects, and provide reliable measurements of intact hormone in the serum of patients with chronic renal failure [5, 9, 10]. Mathias et al recently demonstrated the utility of an IRMA PTH assay for distinguishing between low-turnover and high-turnover lesions of renal osteodystrophy in adolescents and young adults undergoing regular hemodialysis [4]. Several other reports indicate that serum PTH levels, measured by the IRMA technique, are markedly lower in patients with adynamic lesions compared to those with histologic evidence of secondary hyperparathyroidism [5, 10]. In the current study, serum PTH levels above 200 pg/ml strongly suggested the presence of high-turnover skeletal lesions of renal osteodystrophy in pediatric patients undergoing CCPD. All patients with bone biopsy evidence of secondary hyperparathyroidism had values exceeding 200 pg/mi, findings consistent with those recently reported in children and adults receiving regular hemodialysis [4, 9]. Although some patients with normal bone histology also had serum PTH levels greater than 200 pg/mi, none of those with adynamic lesions of bone had values above this threshold. The specificity and positive predictive value of serum PTH measurements for the noninvasive diagnosis of hyperparathyroid bone disease was enhanced by including a second screening criterion; thus, the combination of a serum PTH level above 200 pg/mi and a serum calcium level less than 10 mg/dl correctly identified all patients with bone biopsy evidence of secondary hyperparathyroidism, and it excluded all patients with either normal or reduced rates of bone formation. The finding that bone formation remains normal despite modest increases in serum PTH in some patients with end-stage renal disease mdicates that PTH levels that are generally associated with parathyroid bone disease in subjects with normal renal function are not sufficient to induce this change in children with chronic renal failure. This observation is consistent with data recently reported by Quarles, Lobaugh and Murphy, and it suggests that moderately elevated serum levels of PTH are required to maintain bone formation and turnover within the normal range in patients with chronic renal failure [9] Ḟor the assessment of low-turnover skeletal lesions, serum PTH values below 200 pglml were 100% sensitive and 79% specific, but the PPV was only 58% for patients with adynamic lesions of renal osteodystrophy. Although all patients with bone biopsy evidence of secondary hyperparathyroidism were excluded using this criterion alone, most patients with normal bone histology also had serum PTH levels below 200 pg/mi. Specificity for the diagnosis of the adynamic lesion increased to 89%, however, using the combined criteria of a serum PTH level below 200 pg/mi and a serum calcium level above 10 mgldl. Serum PTH values less than 150 pg/mi in combination with serum calcium levels greater than 10 mg/dl were 80% sensitive

but 92% specific for identifying patients with adynamic bone; thus, PPV rose to 82% using these criteria. In the current study, normal bone histology was present in children with an average serum PTH level of 180 141 pg/mi, whereas histologic evidence of secondary hyperparathyroidism was not found until serum PTH levels exceeded 200 pg/mi. In contrast, all patients with adynamic skeletal lesions had serum PTH levels below 200 pg/mi and 11 of these had values below 65 pg/mi. Such findings not only provide an estimate of the skeletal resistance to the actions of PTH in chronic renal failure but also suggest a suitable therapeutic target range for PTH in children with end-stage renal disease. Serum PTH levels approximately three times the upper limit of normal appear to be functionally appropriate with respect to the integrity of bone in pediatric patients undergoing regular dialysis. Lowering PTH levels further, however, by aggressive therapy with calcium carbonate and 1,25(OH)2D may be associated with the development of adynamic lesions. Whether this disorder adversely influences longitudinal bone growth, the incidence of bone fractures or the prevalence of skeletal deformity in children undergoing regular dialysis, remains to be determined. The results of the current investigation indicate that the prevalence of the adynamic lesion of renal osteodystrophy has increased only modestly during the last five years in children undergoing regular peritoneal dialysis. Before 1987, when the use of aluminum-containing, phosphate-binding antacids was sharply curtailed, 7 of 63, or 11%, of pediatric patients at our institution had histologic evidence of adynamic bone, and one of these cases was due to excess bone aluminum deposition. Since the beginning of 1987, when calcium carbonate was adopted as the primary phosphate-binding agent, 14 of 93 patients undergoing bone biopsy, or 15%, had adynamic lesions of renal osteodystrophy. Overall, the proportion of patients with either normal bone histology or adynamic lesions was 27% before 1987 and 35% thereafter. These results are in striking contrast to data recently reported by Sherrard et al, who found that 66% of adult patients undergoing peritoneal dialysis had adynamic skeletal lesions [5]. This discrepancy may be attributable, at least in part, to the substantial number of adult patients with diabetes and/or evidence of aluminum deposition in bone. In contrast, none of the patients in the current study was diabetic, and aluminumcontaining, phosphate-binding antacids have been used infrequently during the past five to seven years at this facility in children receiving regular dialysis. Moreover, some older adult patients undergoing regular dialysis may have the syndrome of osteoporosis with reduced rates of bone formation and turnover unrelated to the presence of renal disease [22]. Adequate distinction between adults with osteoporosis undergoing regular dialysis and those with adynamic lesions of renal osteodystrophy may be quite difficult, even when assessments include bone biopsy and histomorphometric analysis. It has been suggested that adults receiving regular peritoneai dialysis have a higher incidence of adynamic bone than those managed by hemodialysis [5]. In the current study, the moderate rise in the prevalence of adynamic lesions in children evaluated after 1987 suggests that factors unrelated to peritoneal dialysis per se account for this change since all biopsy data for the current report were obtained from patients undergoing CAPD or CCPD. The increased use of large doses of oral Salusky et al: Biochemical markers of bone disease 257 calcium carbonate and/or more aggressive treatment with active vitamin D sterols, particularly calcitriol, are more likely contributors to the changing pattern of renal osteodystrophy in patients managed by peritoneal dialysis [5, 8, 10]. The system of classification of renal osteodystrophy used in the current and previous studies from this laboratory includes a subgroup with normal bone histology [2]; such patients have no overt histologic evidence of renal osteodystrophy, and tetracycline-based measurements of the rate of bone formation fall within the range seen in normal subjects without renal failure. Other investigators have generally not described a similar subgroup of patients [3, 5, 10]. The distinction between the adynamic lesion of renal osteodystrophy and normal bone histology requires adequate normative data for both histologic indices and tetracycline-based measurements of bone formation. Differences between the current results in children and those described in adults may, in part, be due to variations in the criteria used for histologic classification [3, 5, 10]. In the past, the finding of relatively low serum PTH levels in patients with chronic renal failure suggested the presence of aluminum toxicity and aluminum-related bone disease [23]. With recognition that the long-term use of oral aluminumcontaining, phosphate-binding antacids can lead to the development of aluminum intoxication [24, 25], alternative phosphate-binding agents which do not contain aluminum have been widely utilized [26 28]. The results of the current investigation indicate that relatively low serum PTH levels are a feature of adynamic bone, even in the absence of aluminum intoxication. Such patients commonly develop hypercalcemia which cannot be ascribed to excess PTH secretion; accordingly, parathyroidectomy is not indicated in such patients. The diagnostic criteria described in the current report should be of value for the non-invasive assessment of children with adynamic skeletal lesions of renal osteodystrophy; these criteria may also serve to distinguish between patients with adynamic lesions of bone and those with overt secondary hyperparathyroidism. Acknowledgments This work was supported by USPHS grants DK-35423 and RR-00865. Dr. J.A. Ramirez was a fellow of the National Kidney Foundation of Southern California. The authors are grateful to Ms. Jeanenne O'Connor for technical assistance with histomorphometric analyses. Reprint requests to I. B. Salusky, M.D., UCLA Medical Center, MDCC A2-383, 10833 Le Conte Ave., Los Angeles, California 90024, USA. References 1. SHERRARD DJ, BAYLINK Di, WERGEDAL JE, MALONEY NA: Quantitative histological studies on the pathogenesis of uremic bone disease. J Cliii Endocrinol Metab 39:119 135, 1974 2. SALUSKY IB, COBURN JW, BRILL J, FOLEY J, SLATOPOLSKY E, FINE RN, GOODMAN WG: Bone disease in pediatric patients undergoing dialysis with CAPD or CCPD. Kidney In: 33:975 982, 1988 3. MALLUCHE H, FAUGERE MC: Renal bone disease 1990: An unmet challenge for the nephrologist. Kidney mt 38:193 211, 1990 4. MATHIAS RS, SALUSKY IB, HARMON WH, PAREDES A, EMAN5 J, SEGRE GV, GOODMAN WG: Renal bone disease in pediatric patients and young adults treated by hemodialysis in a childrens hospital. JAm Soc Nephrol 12:1938 1946, 1993 5. SHERRARD Di, HERCZ G, PEI Y, MALONEY N, GREENWOOD C, MANUEL A, SAIPH00 C, FENTON SS, SnoRE GV: The spectrum of

258 Salusky et a!: Biochemical markers of bone disease bone disease in end-stage renal failure An evolving disorder. Kidney mt 43:436 442, 1993 6. ANDRESS DL, MALONEY NA, COBURN JW, ENDRES DB, SHER- RARD Di: Osteomalacia and aplastic bone disease in aluminumrelated osteodystrophy. J Clin Endocrinol Metab 65:11 16, 1987 7. ANDRESS DL, Kop JB, MALONEY NA, COBURN JW, SHERRARD Di: Early deposition of aluminum in bone in diabetic patients on hemodialysis. N Engl J Med 3 16:292 296, 1987 8. MALLUCHE RH, FAUGERE MC: Risk of adynamic bone disease in dialyzed patients. Kidney mt 42:S-62 S-67, 1992 9. QUARLES LD, LOBAUGH B, MURPHY G: Intact parathyroid hormone overestimates the presence and severity of parathyroidmediated osseous abnormalities in uremia. I Clin Endocrinol Metab 75: 145 150, 1992 10. COHEN-SOLAL ME, SEBERT JL, BOUDAILLIEZ B, MARIE A, M0RINIERE P, GUERIS J, BouILLoN R, FOURNIER A: Comparison of intact, midregion, and carboxy-terminal assays of parathyroid hormone for the diagnosis of bone disease in hemodialyzed patients. I Clin Endocrinol Metab 73:516 524, 1991 11. RAMIREZ JA, GOODMAN WG, GORNBEIN J, MENEZES C, MOUL- TON L, SEGRE GV, SALUSKY IB: Direct in vivo comparison of calcium-regulated parathyroid hormone secretion in normal volunteers and patients with secondary hyperparathyroidism. J Clin Endocrinol Metab 76:1489 1494, 1993 12. SALUSICY IB, FINE RN, KANGARLOO H, GOLD R, PAUNIER L, GOODMAN WG, BRILL JE, GILLI 0, SLATOPOLSEY E, COBURN JW: "High-dose" calcitriol for control of renal osteodystrophy in children on CAPD. Kidney mt 32:89 95, 1987 13. PARFITr AM, DREZNER MK, GLORIEUX FR, KANIS JA, MAL- LUCHE HH, MEUNIER PJ, Orr SM, RECKER RR: Bone histomorphometry: Standardization of nomenclature, symbols, and units. I Bone Miner Res 2:595 610, 1987 14. NUSSBAUM SR, ZAHRADNII( Ri, LAVIGNE JR. BRENNAN GL, NOZAWA-UNO C, KIM LY, KEUTMANN T, WANG CA, POTTS it Ja, SnORE GV: Highly sensitive two-site immunoradiometric assay of parathyrin, and its clinical utility in evaluating patients with hypercalcemia. Gun Chem 33:1364 1367, 1987 15. Sox HC JR: Probability theory in the use of diagnostic tests. Ann Intern Med 104:60 66, 1986 16. HonsoN EM, EVANS RA, DUNSTAN CR, HILLS EE, SHAW PF: Quantitative bone histology in children with chronic renal failure. Kidney mt 21:833 839, 1982 17. NOR1SIAN ME, MAZUR AT, BORDEN S, GRUSKIN A, ANAST C, BARON R, RASMUSSEN H: Early diagnosis of juvenile renal osteodystrophy. I Pediatr 97:226 232, 1980 18. Hsu AC, KooR SW, FRASER D, CUMMING WA, FORNASIER VL: Renal osteodystrophy in children with chronic renal failure: An unexpectedly common and incapactating complication. Pediatrics 70:742 750, 1982 19. WITMER G, MARGOLIS A, FONTAINE 0, FRITSCLI J, LENOIR 0, BROYER M, BALSAN S: Effects of 25-hydroxycholecalciferol on bone lesions of children with terminal renal failure. Kidney mt 10:395 408, 1976 20. ANDRESS DL, ENDRES DB, MALONEY NA, KonP JB, COBURN JW, SHERRARD DJ: Comparison of parathyroid hormone assays with bone histomorphometry in renal osteodystrophy. I Cliii Endocrinol Metab 63:1163 1 169, 1986 21. PITTS TO, PIRAINO BH, MITRO R, CHEN TC, SEGRE GV, GREEN- BERG A, PUSCHETT JB: Hyperparathyroidism and 1,25-dihydroxyvitamin D deficiency in mild, moderate and severe renal failure. J Clin Endocrinol Metab 67:876 881, 1988 22. PAREII-r AM, MATHEWS C, RAo D, FRAME B, KLEEREKOPER M, VILLANUEVA AR: Impaired osteoblast function in metabolic bone disease, in Osteoporosis: Recent Advances in Patho genesis and Treatment, edited by DELUCA HF, FROST H, JEE W, JOHNSTON C, PARFITT AM, Baltimore, University Park Press, 1981, p. 321 23. NORRIS KC, CRooKs PW, NEBEKER HG, HERCZ 0, MILLINER DS, GERSZI K, SLATOPOLSKY E, ANDRESS DL, SlinaRAiw Di, COBURN JW: Clinical and laboratory features of aluminum-related bone disease: Differences between sporadic and "epidemic" forms of the syndrome. Am I Kidney Dis 6:342 347, 1985 24. ANDRnOLI SP, BERGSTEIN JM, SHERRARD DJ: Aluminum intoxication from aluminum-containing phosphate binders in children with azotemia not undergoing dialysis. N Eng! J Med 310:1079 1084, 1984 25. SALUSKY IB, COBURN JW, PAUNIER L, SHERRARD Di, FINE RN: Role of aluminum hydroxide in raising serum aluminum levels in children undergoing continuous ambulatory peritoneal dialysis. I Pediatr 105:717 720, 1984 26. SALUSKY lb. C0BURN JW, FOLEY J, NELSON P. FINE RN: Effects of oral calcium carbonate on control of serum phosphorus and changes in plasma aluminum levels after discontinuation of aluminum-containing gels in children receiving dialysis. I Pediatr 108: 767 770, 1986 27. F0URNIER A, MORINIERE PH, SEBERT JL, DKHISSI H, ATIK A, LEFLON P. RENAUD H, GUER1S J, GREGOIRE I, IDRISSI A, GARA- BEDIAN M: Calcium carbonate, an aluminum-free agent for control of hyperphosphatemia, hypocalcemia and hyperparathyroidism in uremia. Kidney mt 29:S-l15 S-119, 1986 28. SLATOPOLSKY E, WEERTS C, LOPEZ-HILKER S, NoRw000 K, ZINK M, WINDUS M, DELMEZ J: Calcium carbonate is an effective phosphate binder in patients with chronic renal failure undergoing dialysis. N Eng! I Med 315: 157 161, 1986