Serum Bone Resorption Markers after Parathyroidectomy for Renal Hyperparathyroidism: Correlation Analyses for the Cross-Linked N-telopeptide of Collagen I and Tartrate-Resistant Acid Phosphatase

Patients on long-term dialysis may develop secondary hyperparathyroidism (SHPT) with increased serum concentrations of bone resorption markers such as the cross-linked N-telopeptide of type I collagen (NTX) and type-5b tartrate-resistant acid phosphatase (TRAP). When SHPT proves refractory to treatment, parathyroidectomy (PTX) may be needed. Renal patients on maintenance HD who received PTX for refractory SHPT (n = 23) or who did not develop refractory SHPT (control subjects; n = 25) were followed prospectively for 4 weeks. Serum intact parathyroid hormone (iPTH), NTX, TRAP, and bone alkaline phosphatase (BAP) concentrations were measured serially and correlation analyses were performed. iPTH values decreased rapidly and dramatically. BAP values increased progressively with peak increases observed at 2 weeks after surgery. NTX and TRAP values decreased concurrently and progressively through 4 weeks following PTX. A significant correlation between TRAP and NTX values was observed before PTX but not at 4 weeks after PTX. Additionally, the fractional changes in serum TRAP were larger than those in serum NTX at all times examined after PTX. Serum iPTH, TRAP, and NTX values declined rapidly following PTX for SHPT. Serum TRAP values declined to greater degrees than serum NTX values throughout the 4-week period following PTX.


Introduction
Secondary hyperparathyroidism (SHPT), a common complication of chronic renal disease and of end-stage renal disease under dialysis, results from vitamin D deficiency, impaired mineral metabolism, and decreased serum erythropoietin values [1][2][3]. Serum parathyroid hormone (PTH) concentrations are elevated such that bone demineralization and bone turnover are enhanced, a condition characterized by the predominance of bone resorption over bone formation [4,5]. Patients with SHPT are therefore typically treated with vitamin D analogues or calcimimetics. When SHPT is resistant to such treatments, parathyroidectomy (PTX) is indicated; substantial increases in bone mineral density are observed following this surgery [6,7].
Bone remodeling normally begins with the creation of resorption pits by osteoclasts, followed by osteoclast apoptosis, osteoblast formation, and mineralization [8,9]. However, 2 The Scientific World Journal this form of remodeling is not observed in patients who have undergone PTX. The pathophysiology of the "hungry bone syndrome" observed after PTX is similar to the bone marrow ablation model used to investigate osteogenic capacity [10,11]. Accordingly, PTX induces an initial rapid and prominent activation of bone synthetic activity, followed by a suppression of osteoclast activity which persists for at least one year [12].
Several enzymes and type I collagen fragments released during bone formation or resorption have recently been proposed as markers of bone remodeling. Type-5b tartrateresistant acid phosphatase (TRAP) is an osteoclast-specific enzyme released during the process of bone resorption, and serum concentrations of the enzyme have been found to correlate directly with the number of osteoclasts in bone biopsies. TRAP is therefore proposed to serve as a marker of bone resorption in renal disease [13]. The cross-linked N-telopeptide of type I collagen (NTX), another marker of osteoclastic bone resorption, is measurable in urine or serum [14]. Although renal disease precludes urinary measurements of NTX for hemodialysis (HD) patients, measurement of serum NTX should serve as a valid marker of the bone resorption state for such patients [15].
Persistent suppression of osteoclast function is wellestablished to follow PTX. However, the relative dispositions of TRAP and NTX and their relationships to other markers of bone resorption have not been established following the surgery. The following study was undertaken to determine serum concentrations of iPTH, NTX, and TRAP at various times following PTX. The main objective was to ascertain whether the post-surgical concentrations of TRAP and NTX are interrelated.

Patients.
A total of 48 patients receiving long-term hemodialysis therapy were enrolled in this study, which included 23 subjects with refractory SHPT and 25 subjects who did not develop refractory SHPT (control subjects). Informed consent was secured from all subjects. Patients deemed eligible were between 32 and 72 years of age. Control subjects had serum iPTH values <300 pg/mL; gender and age distributions for the control group were similar to those of the study group. Patients with refractory SHPT were known to have end-stage renal disease of at least 3 months duration and were receiving maintenance HD three times weekly. None of the patients in this study exhibited signs of malnutrition and all had serum aluminum values below 1.0 µmol/L. The study was approved by the Institutional Review Board of the Cardinal Tien and Tri-Service General Hospitals (Taipei, Taiwan).
PTX was considered necessary when (a) optimal medical and dietary treatments for hyperparathyroidism were unsuccessful, (b) high serum iPTH values persisted, and (c) drugresistant hyperphosphatemia, hypercalcemia, severe osteopathy, vascular calcification, and calciphylaxis were observed. Exclusion criteria included the presence of adynamic bone disease, recent infection, chronic obstructive pulmonary disease, malignant disease, chronic alcoholism, gastrointestinal disease, coronary artery disease, or use of mineralocorticoids, immunosuppressants, or anabolic agents [16,17]. Patients who had received a renal transplant were also excluded.
The known causes of renal failure were diabetic nephropathy (n = 6), hypertensive nephrosclerosis (n = 2), chronic glomerulonephritis (n = 8), polycystic kidney disease (n = 1), and analgesic nephropathy (n = 1). Five subjects had renal failure of unknown cause. No subject had received an aluminum-containing phosphate binder for at least one year prior to surgery.

Parathyroidectomy (PTX).
After performance of total PTX, resected parathyroid tissue was divided into pieces of approximately 1 mm in diameter, and a single piece (100 mg of tissue) was auto-transplanted into the subcutaneous fat of one forearm. Following PTX and transplantation of parathyroid tissue, serum iPTH values were <100 pg/mL and serum calcium values were maintained at approximately 8 to 9 mg/dL. If iPTH subsequently rose to values >100 pg/mL, serum calcium was maintained at approximately 9 to 10 mg/dL by administration of low-dose active vitamin D and calcium salts to prevent recurrent HPT. When iPTH values fell below 70 pg/mL, the vitamin D treatment was interrupted to avoid adynamic bone disease [18].
No supplementation with vitamin D was provided to any patient for at least one month preceding surgery. Postoperatively, all subjects received oral calcitriol (1,25(OH) 2 vitamin D 3 , 2.0 µg/d) [19]. Otherwise, intravenous calcium gluconate was administered postoperatively as needed. Some patients received oral calcium supplements at established daily doses. No cases of permanent hypoparathyroidism were observed.

Biochemical Parameters and Serum Bone Resorption
Markers. To obtain baseline values, fasting (10 h) blood samples were collected between 8 : 00 and 9 : 00 AM preoperatively on the day of PTX (D0). Post-PTX blood samples were collected at 24 and 72 h (D2 and D4, resp.) and at 1, 2, and 4 weeks (W1, W2 and W4, resp.) after the surgery. To harvest serum, blood samples were subjected to centrifugation for 30 min. Serum was obtained within 1 h and stored at −30 • C until use for measurements of iPTH, calcium, phosphate, and other bone metabolism parameters.
Serum iPTH concentrations were measured using a two-site immunoradiometric assay (Nichols Institute diagnostics, San Juan Capistrano, CA, USA) which detects the biologically intact 84-amino-acid chain of PTH. Total calcium (TCa), serum phosphate (Pi), alkaline phosphatase, and albumin were determined with an AU5000 automated chemistry analyzer (Olympus, Tokyo, Japan). Serum aluminum was measured by atomic absorption spectrometry in a graphite oven. Serum NTX concentrations were measured using an ELIZA kit (OSTEOMARK; Mochida pharmaceutical Co., Tokyo, Japan); this marker is measured spectrophotometrically and its concentration is determined from a standard calibration curve. Normal NTX values ranged between 6.2 and 19.0 nM bone collagen equivalents/L (BCE/L) for premenopausal women and between 5.4 and 24.2 nM BCE/L for men. Serum TRAP activity was assayed in multi-well plates in the presence of monoclonal antibody

Correlation Analyses.
To examine potential relationships between NTX and other parameters, measurements obtained at various times were pooled and statistical analysis was performed ( Table 2). NTX values correlated significantly with those for iPTH, TRAP, BAP, TCa, and Pi (r = 0.450, P < 0.001; r = 0.690, P < 0.001; r = 0.448, P < 0.001; r = −0.161, P = 0.042; and r = 0.197, P = 0.012, resp.). We calculated univariate Spearman's correlation coefficients to determine the relationships between serum NTX and iPTH, BAP, and TRAP levels at different times after PTX (Table 3). Figure 2 presents correlations between serum NTX values and those for other bone biomarkers before (D0) and at 4 weeks after PTX (W4). Before PTX, iPTH concentrations  activity as reflected by elevated NTX values. However, this correlation was lost at 4 weeks after PTX (P = 0.276). TRAP and NTX values (panel (b)) were also found to correlate significantly before (r = 0.874, P < 0.001), but not 4 weeks after (r = 0.293, P = 0.174), PTX. Correlations between NTX and BAP are presented in panel (c). Since BAP concentrations were observed to peak at W2, this time was chosen for analysis. BAP and NTX were significantly correlated both before and 2 weeks after PTX (r = 0.701, P < 0.001 versus r = 0.777, P < 0.001, resp.).

Discussion
The present study is the first to examine a variety of serum markers of bone resorption with respect to their disposition and interrelationships following PTX for SHPT. iPTH concentrations were found to decrease profoundly within 6 The Scientific World Journal Post-PTX blood samples were collected at 24 and 72 h (D2 and D4, resp.) and at 1, 2, and 4 weeks (W1, W2 and W4, resp.) after the surgery. Spearman's correlation coefficients were calculated assuming a nonnormal distribution.
24 h following PTX. Serum NTX and TRAP values declined progressively over the 4-week period following the surgery. By contrast, BAP concentrations rose postoperatively within 24 h and peaked at 2 weeks post-PTX. These findings agree with those of previous studies [20] in which decreases in NTX and TRAP and increases in BAP were observed after PTX for SHPT. Significant correlations of NTX with iPTH, TRAP, BAP, TCa, and Pi were observed in the present study. However, NTX correlated significantly with iPTH and TRAP before (D0) but not at 4 weeks (W4) after surgery.
Additionally, the mean percent decrease in serum NTX values was significantly smaller than that in serum TRAP values at all times following PTX. The finding that TRAP and NTX values, both of which reflect osteoclastic activity, did not correlate directly following PTX was unexpected. NTX, which is formed during the process of bone resorption by osteoclasts, is proposed to serve as a valid marker of the process [21]. In patients without renal disease, NTX values fluctuate greatly in response to changes in bone resorption patterns and may vary as much as 50% on a daily The Scientific World Journal basis [22]. Although impaired urine production precludes urinary measurement of NTX in HD patients, serum NTX provides an excellent marker of the bone resorption state in these patients [15,23,24]. Accordingly, HD patients have high serum NTX concentrations and exhibit rapid rates of bone loss, and NTX values have been found to predict accurately the rate of bone loss over a 2-year follow-up period for these patients [15]. In postmenopausal women, effective bisphosphonate treatment results in a quantifiable reduction in urinary excretion of NTX and improvement in bone mass [25,26]. In addition, serum NTX has been found useful as a monitor of the anti-resorptive efficacies of pharmacologic treatments for tumor bony metastasis [27,28]. Findings of the present study agree well with these observations. A significant and progressive decrease in serum NTX values was observed after PTX, indicative of a significant attenuation of bone collagen breakdown. Furthermore, after PTX for primary hyperparathyroidism, a condition in which increased bone resorption is related to excessive secretion of PTH [29]. Serum PTH and NTX concentrations decline rapidly and in parallel; declines in these markers are similar to those observed in the present study after PTX for SHPT. Serum TRAP concentration is not influenced by renal function and exhibits less biological variation compared with urinary and serum concentrations of NTX and Ctelopeptides of type-I collagen [30]. Increased osteoclastic activity during the process of bone resorption is associated with increased synthesis and secretion of TRAP [31,32]. TRAP activity is increased in patients with chronic renal failure and in uremic patients with SHPT [33]. Decreases in serum TRAP reflect decreases in osteoclast activity, and serum TRAP is independently correlated with decreased bone mineral density [34]. Furthermore, pre-PTX serum TRAP values have been found to be predictive of bone mineral densities for the femoral neck and lumbar spine at 1-year following PTX [35]. In the present study, a significant correlation between iPTH and TRAP was observed before and after PTX, highlighting the role of PTH in bone resorption in patients with SHPT. The observed decline in serum TRAP activity after PTX therapy is therefore proposed to be attributable to the lower serum iPTH values in these patients.
In maintenance HD or pre-dialysis patients, an elevation in serum BAP is related to increased cardiovascular mortality and vascular calcification [36,37]. In the present study, an elevation in serum BAP was observed with peak values obtained at 2 weeks post-surgery. This observation may be related to development during the early post-PTX period of the "hungry bone syndrome" associated with increased calcium deposition in bone [38]. The observed increase in serum BAP concentration at these early times may also promote mineralization by extraction of inhibitory pyrophosphate [39,40].
A significant correlation between BAP and NTX concentrations was observed before and 2 weeks after PTX. This observation is consistent with coupling of osteoblastic bone formation and osteoclastic bone resorption regardless of PTX. By contrast, TRAP and NTX concentrations did not correlate significantly at 4 weeks post-surgery, and the fractional changes in TRAP concentrations did not correlate with those for NTX at any time post-PTX. These findings favor the hypothesis that although osteoclast number is decreased after PTX, those osteoclasts that remain are still capable of degrading collagen. However, renal retention of NTX cannot be excluded. Other limitations of the current study include the small sample size, follow-up is limited to 4 weeks and also the lack of bone histomorphometric data.
In conclusion, serum iPTH, TRAP, and NTX concentrations decline rapidly following PTX for SHPT. The decline in these markers is attributable to suppression of bone resorption. Fractional decreases in serum TRAP are larger than those in serum NTX during the post-PTX period. Whether the latter observation indicates that bone collagen digestive ability declines more slowly than TRAP activity or reflects the renal retention of NTX warrants further study.