| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Endocrinology and Metabolism (L.G., F.T., F.C., A.Pia, P.R., G.M., G.B.), S. Croce and Carle Hospital, 12100 Cuneo, Italy; and Division of Endocrinology and Oncology (A.Pio.) S. G. Battista Hospital, 10126 Turin, Italy
Address all correspondence and requests for reprints to: Giorgio Borretta, M.D., Division of Endocrinology and Metabolism, Department of Internal Medicine, A.S.O. S. Croce and Carle, Via M. Coppino, 26, 12100 Cuneo, Italy. E-mail: borretta.g{at}ospedale.cuneo.it.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsDiscussionReferences |
|---|
Aim: The aim of the study was to assess the differences in bone status in a series of consecutive patients affected by PHPT without overt renal failure at diagnosis grouped according to creatinine clearance (Ccr).
Methods: A total of 161 consecutive patients with PHPT were studied. They were divided into two groups based on Ccr. Group A had Ccr 70 ml/min or less (n = 49), and group B had Ccr greater than 70 ml/min (n = 112). PTH, total and ionized serum calcium; urinary calcium and phosphate; serum 25-hydroxyvitamin D3; serum and urinary bone markers; lumbar, forearm, and femoral bone mineral density (BMD) were evaluated.
Results: Patients in group A were older than those in group B (P < 0.0001). PTH levels did not differ in the two groups, whereas both urinary calcium and phosphorus were lower in group A than group B (P < 0.01). Lower BMD was evident in group A at lumbar spine (P < 0.002), forearm (P < 0.0001), and femur (P < 0.01). In asymptomatic PHPT, those with Ccr 70 ml/min or less had lower forearm BMD than patients with higher Ccr (P < 0.00001). When adjusting for age and body mass index in PHPT, BMD at each site persisted being lower (P < 0.05) in group A than group B. In all PHPT subjects, Ccr (beta = 0.29, P < 0.0005), age (beta = –0.27, P < 0.00001), and PTH levels (beta = –0.27, P < 0.0005) were all independently associated with forearm BMD.
Conclusions: In PHPT a slight decrease in renal function is associated with more severe BMD decrease, independent of age, body mass index, and PTH levels. This association is also present in asymptomatic PHPT and strengthens the National Institutes of Health recommendations for surgery in patients with mild PHPT.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
The impact of renal function on the biochemical expression of PHPT has been well documented (6, 7, 8, 9, 10). In mild PHPT, glomerular filtration rate (GFR) has been reported to be inversely correlated to the preoperative serum levels of intact PTH, ionized calcium, and osteocalcin (6). More recent data confirm the association between a slight reduction in GFR measured as Ccr and increased bone turnover reflected by higher levels of PTH, calcium, and osteocalcin in both symptomatic and asymptomatic disease (7). Other findings (7, 8, 9) strengthened the impact of reduced renal function, on the clinical expression of PHPT. In particular, preoperative renal function was associated with increased prevalence of hypertension and diabetes (7) and was linked to the risk of death at long-term follow-up in PHPT (8, 10). The recovery from PHPT is associated with improvement of renal function, and surgery generally leads to a substantial increase of renal concentration capacity (9).
Although all these studies indicate a relationship between renal function and clinical features of PHPT, including bone impairment, to the best of our knowledge, no data on the impact of slight renal failure on bone mineral density (BMD) values are currently available. BMD levels, measured by dual-energy x-ray absorption (DEXA), represent the present-day gold standard to define osteopenia or osteoporosis and predict the risk of fracture (11, 12). They also represent one of the key elements to consider as surgical indications in asymptomatic PHPT (4, 5). Therefore, we thought it would be of interest to study the impact of slight renal failure on BMD and bone turnover in this condition. In particular, we defined the effect of a mild decline of Ccr on bone status, measured in terms of BMD as well as on the basis of serum and urinary bone markers in a large series of consecutive patients with PHPT without overt renal failure at the time of the diagnosis.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
One hundred seventy-three consecutive patients affected by PHPT were studied from 1993 through 2003.
PHPT was diagnosed by the concomitant elevation of total and ionized serum calcium coupled with elevated or inappropriately elevated levels of intact PTH. None of the patients had hepatic or overt renal failure or active thyroid disorder (TSH levels < 0.2 or > 4 µU/ml represented exclusion criteria). To exclude overt renal failure, we evaluated that none of the patients had a previous diagnosis of chronic renal failure or had creatinine levels greater than 1.4 or greater than 1.5 mg/dl (depending on whether female or male) according to the guidelines of the European Society of Hypertension 2003, which indicated these values as a sign of early renal damage (13). For the latter reason, four patients with creatinine levels higher than 1.4–1.5 mg/dl were excluded. Ccr was available for 161 of the remaining 169 patients. Thus, 161 patients represented the study group.
Ccr was assessed using the Cockroft and Gault formulae (14) as follows: Ccr (milliliters per minute) in males = (140 – age) x weight (kilograms)/72 x creatinine (milligrams per deciliter); in females = (140 – age) x weight (kilograms)/72 x creatinine (milligrams per deciliter) x 0.85.
None of the patients had been taking medications that are known to interfere with calcium and vitamin D metabolism in the 6 months before the study.
Eighty-three of 161 patients (51%) were classified as having mild/asymptomatic PHPT (aPHPT). The criteria used to define this group were chosen according to the summary statement by Bilezikian et al. (5), based on the NIH Consensus conference on asymptomatic PHPT of 1991 (4).
A history of kidney stone disease was present in 30% of the patients, whereas classical bone manifestations, such as osteitis fibrosa, cystic lesions, subperiostal resorptions of phalanges, or brown tumors, were recorded in 19% of the patients. Osteoporosis was diagnosed according to the World Health Organization classification (11) in 67.6% of all PHPT patients, whereas 26.6% of the patients showed osteopenia.
All biochemical and hormonal data as well as the results of neck ultrasound, parathyroid scintigraphy, and DEXA were progressively reported in a database.
Study
The patients were studied at the Division of Endocrinology of S. Croce Hospital, Cuneo, Italy, from 1993 through 2003. All the evaluations were performed during a single hospitalization at the time of the diagnosis.
The patients underwent a screening that included the following: 1) complete family and personal medical history; 2) physical examination including body weight, body mass index (BMI) and blood pressure; 3) biochemical evaluation of general blood and urinary parameters, including: total and ionized serum calcium, serum phosphate, serum creatinine, plasma fibrinogen, serum immunoreactive intact PTH, total alkaline phosphate (ALP) and bone isoenzyme serum ALP, serum osteocalcin, urinary cross-links, serum 25-hydroxyvitamin D3 [25(OH)D3] 24 h-urinary calcium (Cau), and urinary phosphate; 4) neck ultrasound, parathyroid scintigraphy; 5) x-ray evaluation of the phalanxes and the skull; and 6) densitometric evaluation of lumbar spine, total proximal femur, and forearm.
Patients consented to these investigations as part of their normal medical care.
The patients were divided into two groups according to Ccr values: group A with Ccr of 70 ml/min or less and group B with Ccr greater than 70 ml/min. This value was chosen because it was already assumed by other authors to represent the cut-off value between slight renal failure and normal renal function in PHPT (7). Moreover, it is known that in chronic renal failure (CRF), secondary hyperparathyroidism may develop in early stages of GFR reduction (15). We thus hypothesized that this Ccr level could make for an easier identification of the possible negative effect of secondary hyperparathyroidism on bone metabolism in PHPT.
Methods and measurements
Biochemical analysis. Total serum calcium, serum phosphate, and serum creatinine as well as Cau were analyzed by a standard autoanalyzer using colorimetric and enzymatic methods, whereas ionized serum calcium was analyzed by a specific probe after correction for pH.
Total serum ALP was measured by a colorimetric assay and serum osteocalcin by a fluorometric assay (TRACE; Kryptur, Brahms, Germany), whereas urinary cross-links were measured by an immunochemiluminometric assay (Diagnostic Products Corp., Los Angeles, CA).
Serum immunoreactive intact PTH was measured by an immunochemiluminometric assay (Diagnostic Products Corp.) with intra- and interassay coefficients of variation of 4.2–5.7 and 6.3–8.8%, respectively.
Serum 25(OH)D3 levels were measured by RIA (Biosource Europe, Nivelles, Belgium) with intra- and interassay coefficients of variation of 6.1–7.9 and 7.1–8.2%, respectively.
Densitometric measurements. he BMD of the lumbar spine (L2-L4), proximal femur, and distal third of the nondominat radius was measured by DEXA QDR-4500 (Hologic, Bedford, MA). The data on bone density are reported as absolute measurements of BMD (grams per square centimeter), T score (SD from mean BMD of young adult population), or Z score (SD from mean BMD of age-matched population).
Statistical analysis
Data are expressed as mean ± SD.
All analyses were performed using Statistica (version 7.0 for Windows; StatSoft Inc., Tulsa, OK).
Statistical differences in mean levels of the variables were assessed by Mann Whitney U test and ANCOVA (with age and BMI as covariates) followed by LSD post hoc test.
Spearman rank correlation test was calculated between Ccr and BMD or PTH or serum/Cau or bone markers. Multiple linear regression analysis (all variables entered model) was used to determine whether Ccr or PTH or age were independently related to forearm BMD.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
. PHPT patients were mostly asymptomatic (83 of 161) and female (120 of 161). Their mean Ccr was 86.0 ± 29.4 ml/min with a range from 30 to 168 ml/min and a median of 83 ml/min.
|
). PTH levels did not differ significantly in the two groups (197.4 ± 148.4 and 184.4 ± 132.1 pg/ml, in groups A and B, respectively) with the following details: group A, range 58–630, median level 148, lower quartile 96, higher quartile 241 pg/ml; group B, range 42–723, median level 140, lower quartile 90, higher quartile 226 pg/ml.
No significant differences were present in total and ionized serum calcium in the two groups, whereas Cau and urinary phosphate were lower in group A than B (P < 0.01). Serum or urinary bone markers as well as serum 25(OH)D3 levels were not significantly different in the two groups, although a trend of higher osteocalcin levels was observed in group A (P = 0.09). Among biochemical parameters evaluated, plasma fibrinogen was higher in group A than group B (P < 0.03).
BMD as well as T score at lumbar spine (P < 0.002), femur (P < 0.01), and forearm (P < 0.0001) were lower in group A than B (Fig. 1). Even Z score at forearm was lower in group A than group B (P < 0,01). All biochemical, hormonal, and BMD data are reported in Table 2.
|
|
Adjusting for age or BMI in all PHPT patients, both BMD and T score at each bone site persisted being lower in group A than group B. Similarly, in aPHPT forearm BMD persisted, being lower in patients with lower Ccr levels, independently of age or BMI (0.38 ± 0.1 vs. 0.49 ± 0.1 g/cm2, P < 0.0001).
In PHPT, there was a positive correlation between Ccr and forearm BMD (r = 0.54, P < 0.00005) (Fig. 2) as well as lumbar BMD (r = 0.32, P < 0.05). Moreover, we noticed a positive association between Ccr and Cau (r = 0.3, P < 0.02), whereas we did not find any significant association either between Ccr and PTH or between Ccr and bone markers. A negative correlation between Ccr and plasma fibrinogen was found (r = –0.2, P < 0.05) in the whole group of patients.
|
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
The link among renal function, PTH, and bone metabolism is well known in both physiological and pathological conditions (6, 7, 15, 16, 17).
In CRF the progressive reduction of GFR is commonly associated with an increase in PTH, leading to progressive bone damage and finally renal osteodystrophy (15, 16, 17). However, skeletal changes begin even at an early stage of CRF when reduced BMD, increased levels of PTH, and biochemical markers of bone turnover can be found (15, 16, 17). Nevertheless, recent analysis from the Third National Health and Nutrition Examination Survey (18) demonstrated that although adult subjects with worse renal function have significantly lower femoral BMD, this association can be explained by confounding factors, namely sex, age, and weight. Thus, the negative effects of reduced renal function on BMD could be minimal or subclinical, at least among persons with mild CRF (18).
In mild PHPT even a mild decrease of renal function represents a further indication for surgery (4, 5). This indication is based on the fact that a decline of renal function can influence the clinical expression of PHPT and worsen bone status. However, to this day clear evidence of the relationship between renal function and bone impairment, in terms of BMD decrease, is still lacking.
We studied a large group of subjects affected by PHPT with normal or slightly decreased renal function, and we chose the value of Ccr 70 ml/min as the limit between normal renal function and early CRF, in accord with previous studies (7, 15, 16, 17).
A negative influence of renal function on bone metabolism had already been reported in PHPT (6, 7). In fact, serum levels of calcium, PTH, and osteocalcin were reported significantly higher and 24-h Cau and phosphate significantly lower in patients with concomitant PHPT and renal function slightly decreased in comparison with patients with normal renal function (7). Our results confirmed a decrease in Cau and phosphate in PHPT patients with lower Ccr. The reduced excretion of Cau in patients with decreased renal function suggests that the evaluation of this parameter may be misleading in the clinical assessment of these patients. According to the NIH Consensus Development Conference statement (4, 5), surgery is recommended when Cau excretion is higher than 400 mg in 24 h. Cau excretion, however, can be reduced in patients with slight decrease of renal function, hence limiting the role of hypercalciuria as a marker of more aggressive disease that might require surgery.
We did not observe any significant difference in serum PTH and total and ionized calcium between patients with preserved renal function and patients with reduced function. Neither serum nor urinary bone markers were significantly different in the two groups, although a trend of higher osteocalcin levels was present in patients with reduced Ccr. Furthermore, no significant association between Ccr and PTH or serum calcium was found in the whole group of patients.
These findings are different from previous reports (7) and can be due to the very strict criteria of inclusion of our study. In fact, we considered patients with mostly normal creatinine levels, whereas in other studies (7), these criteria were not clearly defined and patients with greater variability of serum creatinine could have been studied. Accordingly, most patients in our study had Ccr values higher than 60 ml/min and more homogeneously distributed levels of calcium and PTH.
The lack of difference in PTH in the two groups may be related to the similarly low serum phosphate levels because the increase of serum phosphate is one of the most important and precocious stimuli of PTH secretion in the early stages of CRF (19, 20).
The trend toward higher osteocalcin levels in the group with reduced renal function could reflect an increased bone turnover. However, this finding could also be explained by a prolongation of osteocalcin half-life due to the decrease of GFR, which leads to an increase of osteocalcin levels as detected by immunometric assay (21). Therefore, the slightly higher osteocalcin levels in patients with reduced Ccr might simply reflect a reduced excretion rate or an interference of fragments of degradation on the assay (17, 21) rather than an increase of bone turnover. Thus, the biochemical determination of bone markers may be limited in the presence of reduced GFR and in some cases may be unreliable.
The analysis of the relationships between renal function and BMD in PHPT indicated a negative impact of Ccr decrease on BMD that, to the best of our knowledge, has never been described before in this condition. In particular, a more severe decrease in mineral density at each bone site was recorded in patients with lower Ccr, independently of PTH levels. Also in asymptomatic patients, forearm BMD was lower in those with reduced renal function, and again this relationship was independent of PTH.
To rule out the impact of age or BMI, both of which independently influence BMD levels (11, 12, 18, 22), we evaluated the same results, considering age or BMI as covariates and the analysis confirmed the findings. Furthermore, considering all PHPT patients, we observed a positive association between Ccr and BMD at lumbar spine and forearm. The latter is almost entirely made of cortical bone and showed the strongest association with Ccr level (lower BMD at forearm in patients with lower Ccr levels). Thus, a slight decrease of renal function can represent a further risk for bone disease and fracture in PHPT, mainly at the cortical level. This is an interesting finding if we consider that in this condition cortical bone is commonly more impaired than trabecular bone (23, 24).
It is hard to explain the lower BMD in the group with reduced renal function, PTH, and bone markers being not significantly different. Indeed, even in overt renal failure, many manifestations of bone damage cannot be explained exclusively by PTH excess (25). Other factors, such as immune factors, endothelial dysfunction and inflammation, or decreased calcitriol activity, can exert major influences on bone metabolism and can be involved in the pathogenesis of bone damage in CRF (25, 26, 27, 28, 29).
The increased levels of plasma fibrinogen in patients with reduced Ccr could suggest the presence of subclinical inflammation, which has been implicated in the pathophysiology of bone damage during early renal failure (28, 29). The mechanism of inflammation-induced bone loss could involve cytokines and growth factors system. For instance, the imbalance in the signaling system consisting of receptor activator of nuclear factor-
B ligand (RANKL), its receptor RANKL, and its decoy receptor osteoprotegerin (OPG) has been reported in both PHPT and CRF (25, 30, 31). However, recent studies indicate that the measurement of circulating OPG may not reflect the local activity of the OPG/RANKL system at bone cells (30, 31). Thus, further studies are needed to clarify the mechanisms by which a mild CRF amplifies the PHPT-induced bone damage.
Whatever the pathogenetic mechanism, the importance of our findings is remarkable, considering the primary role that BMD measurement has in both the clinical assessment and treatment of the modern form of PHPT, in which typical bone disease is rare, whereas osteoporosis is common (1, 2, 3, 4, 5, 23, 24). In fact, a BMD decrease may influence the therapeutical approach to mild PHPT (4, 5).
In conclusion, our findings indicate that in PHPT a slight decrease in renal function is associated with more severe bone damage, in particular by a more severe decrease in BMD at each bone site and especially at cortical bone sites. This association is independent of age or PTH or BMI and is also present in patients with mild PHPT. Some mechanisms, in addition to PTH excess, such as subclinical inflammation, may be claimed to explain the relationship between renal and BMD impairment in this condition.
As a whole these data strengthen the NIH recommendations for surgery in patients with mild PHPT who have a slight decrease of renal function.
|
Footnotes |
|---|
Abbreviations: ALP, Alkaline phosphate; aPHPT, asymptomatic PHPT; BMD, bone mineral density; BMI, body mass index; Cau, urinary calcium; Ccr, creatinine clearance; CRF, chronic renal failure; DEXA, dual-energy x-ray absorption; GFR, glomerular filtration rate; 25(OH)D3, 25-hydroxyvitamin D3; OPG, osteoprotegerin; PHPT, primary hyperparathyroidism; RANKL, receptor activator of nuclear factor-B ligand.
Received January 12, 2006.
Accepted May 22, 2006.
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsDiscussionReferences |
|---|
B ligand mRNA levels in primary hyperparathyroidism: effect of parathyroidectomy and association with bone metabolism. Bone 35:256–265[Medline]This article has been cited by other articles:
 |
|
 |
|
  F. Tassone, L. Gianotti, I. Emmolo, M. Ghio, and G. Borretta Glomerular Filtration Rate and Parathyroid Hormone Secretion in Primary Hyperparathyroidism J. Clin. Endocrinol. Metab., November 1, 2009; 94(11): 4458 - 4461. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
