Diabetic patients on maintenance dialysis often are characterized by a relative parathyroid hormone (PTH) deficiency and a form of renal osteodystrophy with low bone turnover known as adynamic bone. The goal of the present study was to determine whether a reduction in the dialysate calcium concentration would increase the predialysis (basal) PTH and maximal PTH level. Thirty-three diabetic maintenance hemodialysis patients with basal PTH values less than 300 pg/ml were randomized to be dialyzed with either a regular (3.0 mEq/liter or 3.5 mEq/liter, group I) or low (2.25 mEq/liter or 2.5 mEq/liter, group II) calcium dialysate for 1 year. At baseline and after 6 months and 12 months of study, low (1 mEq/liter) and high (4 mEq/liter) calcium dialysis studies were performed to determine parathyroid function. At baseline, basal (I, 126 ± 20 vs. II, 108 ± 19 pg/ml) and maximal (I, 269 pg/ml ± 40 pg/ml vs. II, 342 pg/ml ± 65 pg/ml) PTH levels were not different. By 6 months, basal (I, 98 ± 18 vs. II, 200 ± 34 pg/ml, p = 0.02) and maximal (I, 276 pg/ml ± 37 pg/ml vs. II, 529 pg/ml ± 115 pg/ml; p = 0.05) PTH levels were greater in group II. Repeated measures analysis of variance (ANOVA) of the 20 patients who completed the entire 12-month study showed that only in group II patients were basal PTH (p = 0.01), maximal PTH (p = 0.01), and the basal/maximal PTH ratio (p = 0.03) different; by post hoc test, each was greater (p < 0.05) at 6 months and 12 months than at baseline. When study values at 0,6, and 12 months in all patients were combined, an inverse correlation was present between basal calcium and both the basal/maximal PTH ratio (r = −0.59; p < 0.001) and the basal PTH (r = −0.60; p < 0.001). In conclusion, in diabetic hemodialysis patients with a relative PTH deficiency (1) the use of a low calcium dialysate increases basal and maximal PTH levels, (2) the increased secretory capacity (maximal PTH) during treatment with a low calcium dialysate suggests the possibility of enhanced parathyroid gland growth, and (3) the inverse correlation between basal calcium and both the basal/maximal PTH ratio and the basal PTH suggests that the steady-state PTH level is largely determined by the prevailing serum calcium concentration. (J Bone Miner Res 2000;15:927–935)
Diabeticpatients on maintenance hemodialysis often are characterized by a relative deficiency of parathyroid hormone (PTH) and a form of renal osteodystrophy with low bone turnover known as adynamic bone.(1–3) Because PTH levels, although low for azotemic patients are greater than in normal individuals, the adynamic bone would appear to be at least partially a result of skeletal resistance to PTH. Indeed, in studies of renal osteodystrophy in dialysis patients, it has been reported that PTH levels 2–4 times the upper limits of normal are needed to maintain a normal osteoblast surface and bone formation rate.(4,5)
Until recently, the goal during hemodialysis was to suppress PTH secretion by using a 3-mEq/liter or 3.5-mEq/liter calcium dialysate. However, in hemodialysis patients with a relative PTH deficiency and adynamic bone, dialysis with a 3-mEq/liter or 3.5-mEq/liter calcium dialysate designed to suppress PTH secretion may not be appropriate. In two studies in nondiabetic maintenance hemodialysis patients with a relative PTH deficiency, the use of a 2.5-mEq/liter calcium dialysate for 1 year resulted in an increase in predialysis PTH levels.(6,7) However, in neither study were diabetic patients specifically evaluated nor was parathyroid function determined. Thus, it is possible that the increased predialysis PTH levels were simply a secretory response to the calcium-lowering effects of the low calcium dialysate rather than an enhanced capacity to maximally secrete PTH.
The goal in the present study was to determine whether in diabetic maintenance hemodialysis patients with a relative PTH deficiency, hemodialysis for 1 year with a low calcium dialysate would increase the predialysis (basal) PTH level and the maximal secretory capacity of the parathyroid gland (maximal PTH).
MATERIALS AND METHODS
The criteria for inclusion in this study of maintenance hemodialysis patients were diabetes and a predialysis intact PTH level less than 300 pg/ml. Thirty-six patients were enrolled in the study. Before randomization 1 patient died, 1 patient received a kidney transplant, and 1 patient was determined not to be diabetic and was removed from the study. The mean age of the remaining 33 patients was 57.2 years ± 2.1 years; 19 patients were female and 14 patients were male. The mean duration of hemodialysis before entry into the study was 31.7 months ± 5.3 months. The primary phosphate binder was calcium carbonate in all patients and the goal was to maintain the serum phosphorus at less than 6 mg/dl. The mean PTH value for these 33 patients was 117 pg/ml ± 14 pg/ml. PTH values were between 9 and 100 pg/ml in 16 patients, between 100 and 200 pg/ml in 11 patients, and between 200 and 293 pg/ml in 6 patients. None of the patients received calcitriol during the study or in the remote past.
Patients were studied at five hemodialysis centers, which included Dialysis Services of Florida, Fort Walton Beach, FL, U.S.A. (n = 10); St. Francis Dialysis Facility, Wichita, KS, U.S.A. (n = 8); and hemodialysis units affiliated with the University of California at Irvine, Orange, CA, U.S.A. (n = 3); the University of Chicago, Chicago, IL, U.S.A. (n = 2); and the Hospital Universitario Reina Sofia, Cordoba, Spain (n = 10). At each location, the institutional review board approved the study. Of the 33 patients entered in the study, 16 patients were randomized to be dialyzed with a calcium dialysate of 2.25 mEq/liter or 2.5 mEq/liter (low calcium dialysate) and 17 patients were randomized to be dialyzed with a calcium dialysate of 3.0 mEq/liter or 3.5 mEq/liter (regular calcium dialysate). The five dialysis centers had their own low (2.25 mEq/liter or 2.5 mEq/liter) and regular (3 mEq/liter or 3.5 mEq/liter) calcium dialysates and the decision was made to allow each to use its low and regular calcium dialysates. The respective age, duration of dialysis, and gender distribution for the patients in group I (regular calcium dialysate) and group II (low calcium dialysate) were similar 57.5 years ± 3.2 years versus 56.9 years ± 2.9 years, 32.0 months ± 7.2 months versus 31.4 months ± 7.6 months, and 8 females and 8 males versus 11 females and 6 males. Of the 16 patients who received a regular calcium dialysate (group I), 10 completed the 12-month study; 3 patients died, 2 before study at 6 months and 1 before study at 12 months; 2 patients received a kidney transplant before study at 6 months; and 1 patient was changed to continuous ambulatory peritoneal dialysis (CAPD) before study at 6 months because of problems with vascular access. Of the 17 patients who received a low calcium dialysate (group II), 11 patients completed the 12-month study but 1 of these patients refused study at 6 months. Six patients did not complete the study: 1 patient received a kidney transplant before study at 6 months and 5 patients died, 1 before study at 6 months and 4 before study at 12 months. In addition, in group II, 1 patient did not have a high calcium dialysis at 6 months and another patient did not have a low calcium dialysis at 12 months.
As we have done previously, to determine maximal PTH secretion and suppression, a low calcium (1 mEq/liter) hemodialysis and a high calcium (4 mEq/liter) hemodialysis were performed on separate dialysis days within 1 week; these studies were performed at baseline and after 6 months and 12 months of treatment with the low or regular calcium dialysate.(8–11) From the data obtained during dialysis-induced hypo- and hypercalcemia, the following terms were defined:
(1)Basal PTH was the predialysis PTH level.
(2)Maximal PTH was the highest PTH level observed in response to hypocalcemia and that an additional reduction of the serum calcium did not further increase the PTH value.
(3)Minimal PTH was the lowest PTH level during suppression by hypercalcemia and that a further increase in the serum calcium did not result in any additional decrease in PTH.
(4)The ratio of basal to maximal PTH was the basal PTH divided by the maximal PTH and this fraction was multiplied by 100 to provide a percentage; in normal volunteers, this ratio is 20–25%.(12) By correcting the actual PTH for the overall capacity to produce PTH (maximal PTH), a measure of the relative degree of PTH stimulation is obtained. When the basal calcium is low, the basal to maximal PTH ratio should be high, indicating that the parathyroid gland is using more of its overall capacity to correct the low calcium; conversely, a high basal calcium should provide information on whether PTH secretion is regulated by serum calcium.
(5)The set point of calcium was defined as we have done previously as the serum calcium concentration at which maximal PTH secretion was reduced by 50%.(8–11) Moreover, as was done in other studies, the set point of calcium also was calculated by the method of Brown in which the set point of calcium is the serum calcium at the midrange between the minimal and maximal PTH.
(6)The basal serum calcium was the serum calcium concentration at the basal (predialysis) PTH.(13,14)
The initial intent was to obtain iliac bone biopsy specimens in all patients at the start and completion of the study. Before randomization, iliac bone biopsy specimens were obtained in 13 patients. However, because the high attrition rate resulted in the need to enroll additional patients, iliac biopsies were eliminated as a requirement for participation in the study; moreover, repeat bone biopsies were not performed because many patients declined to have a second bone biopsy. After the iliac biopsy was obtained, it was processed for bone histomorphometric analysis as described previously(15) and sections were stained by the aurine-tricarboxylic acid method for the histological detection of aluminum. Quantification was performed at a magnification × 200 with a Merz-Schenk reticle and 100 fields were counted. Recommendations of the American Society for Bone and Mineral Research for bone histomorphometric nomenclature were followed.(16)
Except for the low and high calcium studies, serum calcium as well as phosphorus, glucose, alkaline phosphatase, albumin, and aluminum were measured by standard laboratory techniques. An immunoradiometric assay was used to measure intact PTH (Allegro, Nichols Institute, San Juan Capistrano, CA, U.S.A.); normal values are 10–65 pg/ml. In all hemodialysis units except in Cordoba, Spain, serum calcium was measured at bedside with an automated calcium analyzer during the low and high calcium studies (Calcette, Precision Systems, Inc., Natick, MA, U.S.A.). In the hemodialysis unit in Cordoba in which 10 patients were studied, ionized calcium (Ciba-Corning, Madrid, Spain) was measured during the low and high calcium studies. To have serum calcium values that could be compared for all patients, the ionized calcium in millimoles was multiplied by 8 to provide a total serum calcium in milligrams per deciliter. Justification for this transformation is supported by the ratio of the measurements of total calcium to ionized calcium; the mean total calcium (n = 23) was 9.28 mg/dl ± 0.21 mg/dl and the mean ionized calcium (n = 10) was 1.15 mM ± 0.03 mM, a ratio of approximately 8. Moreover, because an equal number of the 10 patients was in each group, the effect on total serum calcium should be weighted equally. Of these 10 patients, only 1 patient died and that was after 6 months. Finally, because the total group was a composite of patients in whom ionized (n = 10) or total (n = 23) calcium was measured, we attempted to validate our merging of the two groups by determining whether the correlations between basal calcium and both basal PTH and basal/maximal PTH ratio were similar. In the 10 patients in whom ionized calcium was measured, and in the 11 patients in group I and the 12 patients in group II in whom total calcium was measured, the respective r values for the comparison between serum calcium and basal PTH were −0.56 (ionized calcium), −0.52 (group I), and −0.53 (group II). For the comparison between serum calcium and the basal/maximal PTH ratio, respective r values were −0.48 (ionized calcium), −0.53 (group I), and −0.52 (group II). Thus, these results serve to support our transformation of ionized calcium values to total serum calcium.
The unpaired t-test was used to compare values between groups I and II at each time interval. For the comparison of values in the same patient at the three time intervals (0, 6, and 12 months), the repeated measures analysis of variance (ANOVA) was used followed by a post hoc test, the Fisher's least significant difference (LSD), for intergroup comparisons. For the correlation between two variables, the Pearson's correlation was used. Stepwise multiple regression was used to correlate a dependent variable with two independent variables. A p value < 0.05 was considered significant and the results are expressed as the mean ± SE.
Analysis of all patients entered into the study
As shown in Table 1,16 patients were entered in group I (regular calcium dialysate) and 17 patients were entered in group II (low calcium dialysate). The two groups were similar with respect to the predialysis (basal) PTH, maximal PTH, minimal PTH, predialysis (basal) serum calcium, basal/maximal PTH ratio, the minimal/maximal PTH ratio, and the set point of calcium. As shown in Table 2, the predialysis serum phosphorus, alkaline phosphatase, albumin, glucose, and aluminum levels also were similar between the two groups at baseline. Except for the serum glucose value at 6 months, subsequent values obtained during the same month as the restudy at 6 months and 12 months remained similar. The only biochemical parameter that changed from baseline during the study was the serum alkaline phosphatase in group II, which was greater at 6 months (p = 0.005) and tended to be greater at 12 months (p = 0.06). Finally, serum aluminum measurements were less than 40 μg/liter in all patients during the study and as shown in Table 2, mean serum aluminum levels at baseline, 6 months, and 12 months were much less.
Prestudy bone biopsy specimens were obtained in 13 patients, 6 patients in group I and 7 patients in group II. For the patient with end-stage renal failure, these biopsy specimens were characterized by a low osteoblast and osteoclast surface, a reduced osteoclast number, a normal osteoid volume, and the absence of or minimal fibrosis (Table 3); moreover, these biopsy specimens were characterized by the virtual absence of double tetracycline labels and a marked reduction in the number of single tetracycline labels. Thus, these results indicate the presence of adynamic bone.
Low and high calcium dialysis studies were performed to evaluate parathyroid function at baseline (0 months) and at 6 months and 12 months (Table 1). At 6 months, the number of patients in each group had decreased, but the group dialyzed with the low calcium dialysate (group II) had a greater basal and maximal PTH level. At 12 months, the basal PTH was greater in group II and the maximal PTH and basal/maximal PTH ratio tended to be greater in group II.
The basal/maximal PTH ratio reflects the relative degree of parathyroid gland stimulation and suppression in response to the existing serum calcium concentration. For the basal/maximal PTH ratio to decrease basal PTH must decrease more than maximal PTH but it is not necessary for the absolute basal PTH level to decrease. However, the presence of an inverse correlation between basal calcium and basal PTH would show that basal PTH values did decrease. When patients from the three study intervals (0, 6, and 12 months) were combined (Fig. 1), an inverse correlation was observed between the basal calcium and both the basal/maximal PTH ratio (Fig. 1 A) and the basal PTH (Fig. 1B). Moreover, at each study interval the correlation between basal calcium and the basal/maximal PTH ratio was significant: r = −0.53 and p = 0.001 (0 months); r = −0.58 and p = 0.002 (6 months); and r = −0.72 and p < 0.001 (12 months). Similar results were observed for the correlation between basal calcium and basal PTH: r = −0.59 and p < 0.001 (0 months); r = −0.58 and p = 0.002 (6 months); and r = −0.69 and p < 0.001 (12 months).
Analysis of the patients who completed the entire study
Although 21 patients completed the 12-month study, only 20 patients could be used for the paired analysis because 1 patient refused to be studied at 6 months. When repeated measures ANOVA was used in these 20 patients to compare values at 0, 6, and 12 months, the p values in group II for basal PTH and maximal PTH were both 0.01. As shown in Fig. 2, the post hoc test showed that in group II, values for basal PTH and maximal PTH at 6 months and 12 months were greater than baseline (0 months). For basal calcium, the p value approached significance (ANOVA, p = 0.06), suggesting that serum calcium tended to decrease during the study. In group I, only basal calcium was significant (ANOVA, p = 0.04) and by post hoc test, the basal calcium value at 12 months was greater than 0 months (Fig. 2).
Shown in Fig. 3 is the basal/maximal PTH ratio at 0, 6, and 12 months for the patients who completed the entire study. In group II, the basal/maximal PTH ratio increased (ANOVA, p = 0.03) as the serum calcium concentration tended to decrease during the 12 months. The basal/maximal PTH ratio at both 6 months and 12 months was greater (p < 0.05) than at 0 months. In group I, the serum calcium increased during the study, but although the basal/maximal PTH ratio was less at 6 months and 12 months, it was not significantly different (ANOVA, p = 0.33). The basal/maximal PTH ratio at 0 months was not different between groups I and II. However, at 6 months and 12 months, the basal/maximal PTH ratio was greater (p < 0.05) in group II (Fig. 3).
Ob.S/BS, %, osteoblast surface, the percentage of trabecular surface covered by osteoblasts; OS/BS, %, osteoid surface, the percentage of trabecular surface covered by osteoid; Oc.S/BS, %, osteoclast surface, the percentage of trabecular surface covered by osteoclasts; N.Oc./T.A., osteoclast number/mm2, the number of osteoclasts per total area (square millimeter) of cancellous bone; OV/BV, %, osteoid volume, the percentage of trabecular bone volume occupied by osteoid; BV/TV, %, bone volume, the percentage of cancellous bone occupied by trabecular bone; Fib./TV, %, fibrosis volume, the percentage of cancellous bone occupied by fibrosis.
a Normal values are from Ref 33.
b Although the listed values for the study patients with adynamic bone are not too dissimilar from normal values, it should be emphasized that the study patients had low bone formation rates by tetracycline labeliing and their static values listed in the table are much lower than those in most hemodialysis patients.
Osteoblast surface (Ob.S/BS, %)
0.5 ± 0.3
0.3 ± 0.3
0.6 ± 0.5
1.0 ± 0.4
Osteoid surface (OS/BS, %)
18.5 ± 2.8
16.2 ± 4.9
20.5 ± 3.4
12.6 ± 3.5
Osteoclast surface (Oc.S/BS, %)
1.3 ± 0.5
0.7 ± 0.4
1.8 ± 0.9
0.16 ± 0.06
0.39 ± 0.17
0.21 ± 0.12
0.54 ± 0.31
0.11 ± 0.03
Osteoid volume (OV/BV, %)
3.0 ± 0.5
2.6 ± 0.7
3.3 ± 0.8
2.0 ± 0.6
Bone volume (BV/TV, %)
20.8 ± 1.9
19.8 ± 2.2
21.8 ± 3.4
21.9 ± 2.1
Fibrosis (Fib./TV, %)
0.04 ± 0.02
0.07 ± 0.05
0.01 ± 0.02
Aluminum surface (Al/BS, %)
1.3 ± 0.8
0.6 ± 0.4
2.0 ± 1.6
Our assumption was that in group II the low calcium dialysate would make basal PTH more calcium dependent and in group I the higher calcium dialysate would make the basal PTH less calcium dependent and more dependent on the maximal secretory capacity of the parathyroid gland. To determine the effect that the basal calcium and the maximal PTH (independent variables) had on the basal PTH (dependent variable), a stepwise multiple regression was performed in the patients who completed the 12-month study (Table 4). When all the patients were combined as a single group, both the maximal PTH and the basal calcium (independent variables) had a major effect on the basal PTH (dependent variable); for maximal PTH, the effect was positive and for basal calcium, the effect was negative. During the course of the study, the effect of both basal calcium and maximal PTH increased from baseline (0 months). When the patients were separated into groups I and II, the effect of maximal PTH and basal calcium at baseline was significant but modest in both groups. In group II, at 6 months and 12 months the effect of basal calcium increased and that of maximal PTH remained the same or even decreased. Conversely, in group I the effect of maximal PTH markedly increased by 12 months whereas basal calcium no longer exerted a significant effect at 6 months and 12 months.
The set point of calcium in hemodialysis patients with a relative deficiency of PTH
During the past decade, two methods have been used for calculating the set point of calcium. Previously, we have shown that in hemodialysis patients with basal PTH values >300 pg/ml, both methods provide similar information.(11,17) In the present study in which basal PTH values were <300 pg/ml, the correlation between the set point at 50% and set point at midrange was highly significant (r = 0.98, p < 0.001). A similar correlation was observed between the delta set point at 50% and delta set point at midrange (r = 0.98, p < 0.001); the delta value was defined as the difference in set point between 0 and 6 months, 0 and 12 months, and 6 and 12 months. When a statistical method for determining agreement between two methods of clinical measurement was used, this test showed that a high degree of agreement was present between the two methods for the calculation of the set point.(18)
In diabetic hemodialysis patients with a relative PTH deficiency, the use of a low calcium dialysate for 1 year resulted in a modest but a significant increase in basal and maximal PTH levels. Conversely, in diabetic hemodialysis patients receiving a regular calcium dialysate, basal and maximal PTH levels did not change during the study. The present study shows that as with nondiabetic patients on hemodialysis and on CAPD, the diabetic hemodialysis patient with a relative PTH deficiency is able to increase PTH levels during treatment with a low calcium dialy-sate.(6,7,19,20) Moreover, the present study is the first to show that the increase in basal PTH is accompanied by an increase in maximal PTH. The increase in maximal PTH shows that the increase in basal PTH was not solely a result of a decrease in the basal calcium or a PTH response to maintain the same serum calcium but documents the presence of an increased maximal secretory capacity of the parathyoid gland. However, the increase in maximal PTH was proportionally less than the increase in basal PTH. Finally, our study also showed that the PTH stimulation primarily occurred during the first 6 months.
The low calcium dialysate in group II increased basal and maximal PTH levels. As might be expected with a lower serum calcium, the increase in basal PTH was greater than maximal PTH. This result is best shown by the basal/maximal PTH ratio, which increased at 6 months and 12 months as compared with baseline values. In the patients in group II who completed the study, basal PTH increased from 134 to 252 pg/ml between 0 and 12 months. Although the increase in basal PTH may have a physiologically significant effect on bone, it does not represent a very high PTH value in the hemodialysis patient. It is also of interest to note that minimal if any change in any PTH parameter was observed between 6 and 12 months despite continued dialysis with a low calcium dialysate and a further lowering of serum calcium. Thus, the increase in basal PTH and the enhanced secretory capacity of the parathyroid gland (maximal PTH) were observed primarily during the first 6 months. Whether the enhanced secretory capacity of the parathyroid gland would continue to increase but at a slower rate cannot be determined from our study.
An inverse correlation was observed between the basal calcium and both the basal/maximal PTH ratio and the basal PTH. A similar result between serum calcium and basal PTH has been shown previously by Combe et al. in patients with advanced renal failure and a similar magnitude of hyperparathyroidism.(21) For an inverse correlation to be present between the serum calcium and basal/maximal PTH ratio, an increasing serum calcium must decrease basal PTH relative to maximal PTH. However, it does not necessarily imply that basal PTH must decrease. For example, we have recently reported in hemodialysis patients with more severe hyperparathyroidism (mean basal PTH, 586 pg/ml ± 51 pg/ml) that an inverse correlation was present between the basal/maximal PTH ratio and serum calcium, but the relationship between basal PTH and serum calcium was flat.(22) In these patients, the maximal PTH increased with the serum calcium; an increasing serum calcium decreased basal PTH relative to maximal PTH but did not produce a decrease in basal PTH levels. Conversely, in the patients in the present study with relatively mild hyperparathyroidism (mean basal PTH, 117 pg/ml ± 13 pg/ml, n = 33), the relationship between maximal PTH and serum calcium was flat and not positive, as it was in the patients with more severe hyperparathyroidism. Thus, when basal PTH decreased relative to maximal PTH, an inverse correlation was observed between serum calcium and basal PTH. Although distinct differences are present between the two cited situations, it would seem that in both groups of patients, basal PTH secretion is being regulated by the prevailing serum calcium concentration.
In the patients who completed the entire study, stepwise multiple regression analysis at baseline showed that for the entire group, the basal PTH value had a direct relationship with the maximal PTH and an inverse relationship with the serum calcium. Similar results were observed at baseline for groups I and II, but the P values were less, presumably because of fewer patients in each group. As the study progressed the effect of basal calcium became predominant in group II, probably because the low calcium dialysate reduced the serum calcium and stimulated PTH secretion. Conversely, in group I in which the regular dialysate increased the serum calcium and reduced its range, the effect of maximal PTH became predominant. For the entire group of patients, the effect of maximal PTH and basal calcium both increased during the study presumably because the inclusion of both groups resulted in a wider range of both serum calcium and maximal PTH values.
It is necessary to consider whether in diabetic hemodialysis patients with a relative PTH deficiency it is important to increase PTH levels and to potentially increase bone remodeling. In general, patients with adynamic bone are asymptomatic unless bone aluminum accumulation is present.(23,24) Because of decreased bone remodeling, patients with adynamic bone have a decreased capacity to buffer a calcium load.(25) Although exposure to aluminum may be the only cause of adynamic bone that results in symptomatic bone disease, there appears to be an increased risk of hypercalcemia and hyperphosphatemia, which is most likely related to the decreased buffering capacity of bone.(24–26) In dialysis patients with adynamic bone, it has been recommended that care be taken in the selection of the dialysate calcium and that PTH levels be maintained between 100 and 200 pg/ml.(24) In our study, the mean basal PTH value in the 13 patients with bone biopsies was 130 pg/ml ± 21 pg/ml and still adynamic bone was present. In the absence of aluminum exposure, symptomatic bone disease may not develop. However, the decreased bone buffering capacity and its attendant hypercalcemia and hyperphosphatemia may increase the risk of soft tissue and vascular calcifications, which could conceivably be a factor in the increased reports of a proximal form of calciphylaxis in which PTH values often are not very elevated.(25–31) Moreover, it is possible that the risk of soft tissue and vascular calcifications in these patients could potentially be reduced by increasing the bone buffering capacity by increasing PTH and bone remodeling. Thus, in hemodialysis patients with adynamic bone and especially in those with diabetes in whom the risk of vascular calcifications may be even greater than in the nondiabetic patient, studies should be performed to determine whether the optimal PTH value should be greater than the currently recommended range of 100–200 pg/ml. Such studies also should include bone mineral density measurements to be certain that a reduction in the dialysate calcium does not result in a reduction in bone mineral density.
Table Table 4.. A Stepwise Multiple Regressionto Evaluatethe Effectof Maximal PTH and Basal Calciumon Basal PTH Levelsin Patients Who Completedthe 12-Month Study
Dependent variable – basal PTH
r2 = 0.51
r2 = 0.63
r2 = 0.75
r2 = 0.49
r2 = 0.30
r2 = 0.88
r2 = 0.54
r2 = 0.70
r2 = 0.75
Finally, in previous studies in patients with severe secondary hyperparathyroidism, we and others have shown a strong correlation between the set point at 50% and the set point at midrange.(11,17,32) In the present study in patients with relatively mild hyperparathyroidism, we again show a strong correlation between the two set points. The close correlation between the two methods for calculating the set point is not surprising because both set points are positioned relatively close to each other and track the same PTH-calcium curve.
In conclusion, in diabetic hemodialysis patients with a relative PTH deficiency (1) the use of a low calcium dialy-sate increases basal and maximal PTH levels, (2) the increased secretory capacity (maximal PTH) during treatment with a low calcium dialysate suggests the possibility of enhanced parathyroid gland growth, (3) the inverse correlation between basal calcium and both the basal/maximal PTH ratio and the basal PTH suggests that the prevailing PTH levels are largely dependent on the existing serum calcium concentration, and (4) the basal PTH level is dependent on both the existing serum calcium concentration and the maximal PTH level. Finally, whether PTH stimulation increases bone remodeling and perhaps more importantly, reduces the risk of soft tissue and vascular calcifications by increasing the capacity for bone to buffer calcium, remains to be determined.
This work was supported by grants from the Baxter Extramural Grant Program (A.J.F. and M.R.). During the course of the study, Dr. Aquiles Jara was a recipient of a fellowship from the International Society of Nephrology and a fellowship from the National Kidney Foundation of Southern California. The authors also thank the dialysis staff at each of the dialysis facilities that participated in this study for their help and cooperation. The authors especially wish to acknowledge the contributions of Traci Simpson, Carrie Schindler, and Roma Fuller for their invaluable assistance during the study.