| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Endocrinology and Metabolism |
1 VA San Diego Healthcare System, San Diego, CA.
Departments of 2
Pathology and 4
Medicine, University of California, San Diego, CA.
3 Lawrence Livermore National Laboratories, Livermore, CA.
5 Department of Nutrition, University of California, Davis, CA.
aAddress correspondence to this author at: VAMC-113, 3350 La Jolla Village, San Diego, CA 92161. Fax 858-642-6304; e-mail rlfitzgerald{at}vapop.ucsd.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: 41Ca was formulated and aliquoted into individual vials, and to the extent possible, the 41Ca doses were tested according to US Pharmacopeia (USP) guidelines. A 10 nCi dose of 41Ca was administered intravenously to 4 end stage renal disease (ESRD) patients on hemodialysis and 4 healthy control individuals. Distribution kinetics were determined over 168 days. Calcium was isolated with 3 precipitation steps and a cation-exchange column, and 41Ca/Ca ratios in serum were then measured by accelerator mass spectrometry.
Results: The dosing solution was chemically and radiologically pure, contained <0.1 endotoxin unit/mL, and passed USP sterility tests. Quantification of 41Ca/Ca ratios was linear from 6 x 10–14 to 9.1 x 10–10. The run-to-run imprecision (as CV) of the method was 4% at 4.6 x 10–11 and 6% at 9.1 x 10–10. The area under the curve of 41Ca in the central compartment vs time was significantly less for ESRD patients than for controls (P <0.005).
Conclusions: Isotope ratios spanning 5 orders of magnitude can be measured by accelerator mass spectrometry with excellent precision in the range observed in samples collected from patients who have received 10 nCi of 41Ca. The 41Ca at this dose caused no adverse effects in 8 volunteers. This is the first report of the use of 41Ca to monitor differences in bone turnover between healthy individuals and ESRD patients.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
41Ca has a half-life of 104 000 years and decays by electron capture to the naturally occurring and stable isotope 41K, emitting x-rays and electrons of very low energy (<3.6 keV) in the process (13)(14). For biological experiments, 41Ca is typically prepared by irradiating calcium samples with neutrons in a nuclear reactor. Other radioisotopes of calcium, 45Ca and 47Ca, are also formed during this process, but these have relatively short half-lives (163 and 4.5 days, respectively). After sufficient time is allowed for the 45Ca to decay, the only radioactive material remaining is 41Ca. The long lifetime of 41Ca prevents its routine analysis by decay counting, but AMS is particularly well suited for long-term monitoring of bone resorption by direct quantification of 41Ca concentrations. The long half-life of 41Ca, combined with the sensitivity of AMS, allows isotope ratios to be measured for a period of years, a time span that is not feasible with other isotopes of calcium. Potential clinical uses of 41Ca include monitoring of bone loss in patients with end stage renal disease (ESRD), diagnosis of osteoporosis, monitoring of the effectiveness of antiresorptive drugs, and early detection of bone metastasis.
Biochemical markers of bone disease have been in use for more than 30 years, but although beneficial in certain settings, all have drawbacks, primarily related to analytical and biological variability (15)(16)(17)(18)(19). One method for interpreting bone markers involves looking for the "least significant change", which incorporates both the analytical and biological variability. The least significant change amounts for bone alkaline phosphatase (AP), osteocalcin, pyridinoline, deoxypyridinoline, and N-telopeptide are in the range of 15% to 40% (20). Bone mineral density, as measured by dual energy x-ray absorptiometry, provides a quantitative measure of the mineralization of bone but is relatively insensitive to change, generally taking 12 to 18 months before changes can be measured. A widely used diagnostic method for identifying bone disease is biopsy and histomorphometry; however, this procedure has morbidity and, in the absence of severe bone disease, is not clinically indicated (21). We report a novel technique for directly assessing the long-term turnover of the mineral phase of bone by measuring 41Ca/Ca ratios in serum after an intravenous dose of 10 nCi of 41Ca.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
preparation of the dosing material
Initial purification and isotopic analysis of the 41Ca material used for the dosing solution have been described (22). We used an aliquot of the 41Ca master stock solution (described as "C1" in the original publication) as the starting material for the dosing solution (22). Radiologic purity is virtually guaranteed from the lack of significant elemental contaminants and the long decay period (the original dose material was purchased in 1984). However, to ensure that we had no radioactive contaminants, we measured the emitted radiation with a gamma/beta survey meter and a high-purity germanium detector.
We received 2.060 g of a solution of CaCO3 in 4 mol/L HNO3 containing 0.7034 mg of calcium per gram of solution. By weight, the 41Ca abundance in the solution was 1.232% (22); consequently, this solution contained 18 µg of 41Ca or 1500 nCi. The target for the dosing solution was 10 nCi of 41Ca in each vial as a 9-mL solution. In a sterile hood, using sterile technique, we added 
500 mL of 0.9% sodium chloride injection (9 g/L) USP (Baxter Healthcare) and 3.75 mL of 10% USP CaCl2 (0.9 mol/L; American Reagent Labs) to a 2-L sterile vessel. The 2.060 g of 41Ca solution was quantitatively transferred to this vessel, and the pH was adjusted to 5–7 by addition of 7.5% sodium bicarbonate USP (0.9 mol/L; American Reagent Labs). The remainder of the solution was diluted to 1350 mL by weight. In the sterile hood, the dosing solution was filtered through a 0.2 µm filter (Corning). With the solution still in the sterile hood, we volumetrically pipetted 9-mL portions into 10-mL clear vials (Wheaton Scientific). A gray straight plug (Wheaton Scientific) and aluminum seal (Wheaton Scientific) were placed loosely on the vial and held in place with a piece of autoclave indicating paper. The doses were autoclaved at 121 °C and 20 psi (140 kPa) for 30 min. In addition to the autoclave indicating tape, the autoclave process was monitored with a biological indicator (3M Attest) to determine that sterilization was successful. Immediately after autoclaving, the aluminum seals were crimped in the sterile hood. Representative samples were analyzed for endotoxins and sterility as described below. Throughout the process, plastic containers and pipetting devices were used to decrease the chance that trace metals would be introduced into the dosing solution. Pyrogenicity (endotoxin testing), sterility, bacterial, and fungistasis testing were then performed according to United States Pharmacopeia (USP) guidelines (23)(24). Elemental composition (purity) was checked by inductively coupled plasma/mass spectrometry using good manufacturing procedures (GMP). Inductively coupled plasma/mass spectrometry was chosen for purity testing because this method has excellent sensitivity (<1 µg/L for most elements), is broad spectrum (able to detect over 60 elements), and is the type of analysis best suited to identify potential impurities in the 41Ca dosing solution. There are no USP guidelines for acceptable limits of impurities for 41Ca dosing solutions; we therefore chose a conservative limit of <0.00001% for any heavy metal contaminant [i.e., the maximum amount of heavy metals to which individuals would be exposed would be 0.00009 g (90 µg)].
Radiologic purity of the 41Ca master stock solution (1.5 µCi) was determined with a gamma/beta survey meter and with a high-purity germanium detector. To obtain an energy spectrum of the electron capture decay of 41Ca to 41K, the entire 1.5 µCi of 41Ca was placed in a high-purity germanium detector for 1 week.
patients
The human subject protocol was approved by Institutional Review Boards at the University of California’s Human Research Protections Program and Lawrence Livermore National Laboratory. ESRD patients were selected as the study group because 75%–100% of these patients have bone disease (25), and this study was designed as a proof-of-principle investigation. Our aim was to determine whether 41Ca/Ca ratios changed more rapidly in patients with known bone disease than in controls without known bone disease. To further enhance the differences between our control and study groups, we chose patients with high-turnover bone disease as evidenced by increased concentrations of parathyroid hormone (PTH). Because the absorption of ingested calcium in ESRD patients is known to vary widely, we chose to administer the dose intravenously to ensure that all controls and patients received the same initial dose of 41Ca.
Eight human volunteers participated in the study, 4 controls and 4 ESRD patients. All volunteers gave written informed consent. Vital signs and pain rating (0–10) of the volunteers were obtained before administration of 10 nCi of 41Ca via a heparin lock. To further ensure sterility, the entire 9 mL of solution to be administered was aspirated through a 0.22 µm pore-size sterile filter into a 10-mL syringe. The sterile filter was removed, and the dose was administered by slow intravenous push through a heparin lock. The heparin lock was then flushed with 10 mL of normal saline. The heparin lock was removed, and vital signs and pain scores were obtained after administration. The ESRD patients received hemodialysis 3 times weekly with 3.5–4 h per treatment. Their dialysis was performed with high-flux membranes, and their calcium management followed K-DOQI guidelines (26). 41Ca was administered immediately after dialysis treatment.
sample analysis
Samples were analyzed at the Center for AMS at Lawrence Livermore National Laboratory, which performs high-throughput, routine analyses of multiple nuclides, including 41Ca (4)(10). The automated AMS control system and streamlined sample preparation allow preparation and measurement of nuclides in more than 100 samples per day by a single operator.
Total AP, urea nitrogen, creatinine, potassium, sodium, chloride, carbon dioxide, phosphate, calcium, albumin, and glucose concentrations were measured with Beckman Synchron reagents on an LX20Pro analyzer. The PTH midmolecule was measured with a previously described assay (27)(28). 25-Hydroxyvitamin D, 1,25-dihydroxyvitamin D, PTH (intact; detects amino acids 7–84), N-telopeptide, and bone-specific AP were measured by ARUP Laboratories (Salt Lake City, UT).
Analysis of 41Ca in plasma by AMS was adapted from methods published by Lin et al. (29) and Freeman et al. (30). Elimination of the microwave digestion step simplified the analysis. 41Ca/Ca ratios were measured by AMS after calcium was isolated and purified using the steps described below.
We added 130 µL of 30 g/L (0.75 mol/L) calcium as CaCO3 to 1 mL of plasma samples to ensure that there was sufficient calcium for AMS (4 mg of Ca, or 0.1 mmol). Samples were then acid-digested for 30 min at room temperature with 2 mL of concentrated HNO3. After digestion, 7 mL of deionized H2O was added to the specimens; a yellow precipitant formed and was discarded after centrifugation (750g for 5 min), and the supernatant was retained. The calcium in the supernatant was then precipitated overnight with 1 mL of saturated ammonium oxalate and 3 mL of concentrated NH4OH. Samples were centrifuged (750g for 10 min), the supernatant was discarded, and the resulting calcium oxalate was dissolved in 0.4 mL of 5 mol/L HNO3. The samples were then diluted with deionized H2O to an acid strength of 0.08 mol/L and chromatographed on a cation-exchange column as described below.
Cation-exchange columns (1.6 g of Bio-Rad AG 50W-X8, 200–400 mesh, hydrogen form) were prepared by washing with 5 mL of 5 mol/L HNO3, 10 mL of deionized H2O, and 5 mL of 0.08 mol/L HNO3. The acid-digested plasma extracts were applied to the columns, which were then sequentially washed with two 4.5-mL portions of 0.08 mol/L HNO3. The calcium was then eluted with two 4.25-mL portions of 5 mol/L HNO3. The pH of the resulting solution was increased with 1.5 mL of deionized H2O and 1.5 mL of concentrated ammonium hydroxide, and samples were allowed to cool for 30 min. After the samples were cool, 3 mL of concentrated HF was added, and CaF2 was allowed to precipitate overnight. (Note: concentrated HF is extremely lethal, and all users must be properly trained before using this acid.) Samples were centrifuged, and the resulting CaF2 was washed 3 times with 1 mL of deionized H2O. After the H2O was decanted, samples were placed in a 100 °C oven to dry for at least 12 h. Recovery of the added calcium was determined gravimetrically. Samples were then mixed with silver powder (2–4 parts CaF2 to silver by weight) to improve conductivity in the AMS source and loaded into aluminum targets for AMS analysis.
As outlined in Fig. 1
, AMS 41Ca measurements were accomplished via (a) ionization of CaF2/silver mixture, (b) selection of 97 or 98 m/z ions (target ions are 40CaF3– and 41CaF3–, respectively), (c) acceleration of ions through 9 x 101
V, (d) removal of the 3 fluorine atoms and 9 electrons from the calcium atom through charge exchange in a thin carbon foil at the high-voltage terminal to destroy any molecular interferences, (e) a second stage of acceleration to ground potential, (f) measurement of 40Ca8+ in an offset Faraday cup, and (g) ion identification and single-particle counting of 41Ca8+ in a multianode gas ionization detector. The details of the AMS analysis can be found elsewhere (4). Our analytical 41Ca/Ca background value was 6 x 10–14. Measurement precision was monitored by use of secondary standards similar in isotope ratio to unknowns and was <3%, assuming sufficient sample 41Ca content to identify at least 1000 41Ca ions.
|
The method was validated at 41Ca/Ca isotope ratios varying from blanks (6 x 10–14) to concentrations in the range expected 2 h post dose (9.08 x 10–10). We added 41Ca to these sera to achieve the target isotope ratios and processed them as described above. Within-run variability was determined at 6 isotope ratios (from blanks to 9.08 x 10–10) with n = 3 at each concentration. Run-to-run variability was determined at 3 isotope ratios (from blanks to 9.08 x 10–10) with each batch of samples. Run-to-run variability represented a minimum of 7 different batches over a 6-month period.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
); both had been added as part of the formulation process.
|
The radiologic purity of the 41Ca master stock solution (1.5 µCi) was determined for the concentrated stock solution, rather than the diluted 41Ca solution that was given to volunteers, to ensure maximized sensitivity and radiologic purity. In the master stock solution, neither 
nor ß radiation above background was detected (Table 1
). In fact, according to US Nuclear Regulatory Commission guidelines, the measured results of the prepared dosing solution fell below the limit of what is considered radioactive (2 nCi/g). As can be seen in Fig. 3
, purity testing gave the expected low-energy internal bremsstrahlung emissions from the electron capture decay of 41Ca and no measurable ß or activity, thus demonstrating that the 41Ca stock solution was radiologically pure. The maximum energy of the sample occurred at 421 keV, a result consistent with the electron capture decay of 41Ca to 41K (31).
|
|
Results of USP sterility tests showed no growth after 14 days, and USP endotoxin testing demonstrated <0.1 endotoxin units/mL. Taken together, these results demonstrate that the 41Ca dose was radiologically and chemically pure and, importantly, also sterile and free of endotoxin.
The difference of results from expected values and the within-run imprecision of the method for 41Ca/Ca ratios in serum samples with added 41Ca over 4 orders of magnitude of isotope ratios expected when administering 10-nCi doses of 41Ca to human volunteers are shown in Table 2
. Run-to-run imprecision was in the 5% range with errors of 1%–3% (Table 3
). As can also be seen in Table 3
, with isotope ratios near background (6 x 10–14), it is difficult to routinely obtain good accuracy and precision because the background amount of 41Ca is so small that it is difficult to detect, an ideal situation for labeling studies. However, after a single 10-nCi dose, volunteers had 41Ca/Ca ratios several orders of magnitude greater than background; therefore, the poor precision and accuracy at background isotope ratios was not relevant to our studies. The mean recovery of calcium through the purification process was 87%.
|
|
There were no postdose changes in vital signs or pain ratings, compared with predose values, for any of the 8 volunteers. The mean baseline clinical chemistry values for the controls and the ESRD patients are shown in Table 4
. The patients were hyperparathyroid, and the increased concentrations of N-telopeptide, bone-specific AP, and total AP provide evidence of increased bone turnover.
|
The initial distribution and elimination kinetics of the 10-nCi 41Ca dose in the plasma from the controls (mean age, 54 years; n = 4) and the ESRD patients [mean age, 59 years; mean (SD) time on dialysis, 2.0 (1.1) years; n = 4] are shown in Fig. 4
. As can be seen in Fig. 4
, over the initial 28 days, the 41Ca/Ca isotope ratio in plasma decreased more rapidly in the ESRD patients than in the controls, giving a lower area under the curve for ESRD patients compared with the controls (P <0.005).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1000-fold less than that used in studies involving other calcium radioisotopes (35). In addition to the perceived risk of radioactivity, other limitations of using 41Ca/Ca ratios to monitor bone health include the high cost of instrumentation and the extensive sample preparation required for analysis. The transition of AMS from the research laboratory to a clinical laboratory may not occur in the near future, but it is worth remembering that the first mass spectrometers, with very limited capabilities, began as large complicated instruments relegated to sophisticated research laboratories. Today, many clinical laboratories have several mass spectrometers with a variety of ionization and analyzer configurations. At present, there is sufficient AMS beam time to meet current needs. If the methods we are developing become essential for monitoring bone health, additional resources will need to be developed. Alternative techniques for measuring 41Ca include resonance ionization mass spectrometry (36) and atom trap trace analysis (37), but these methods have not yet matched the throughput or sensitivity of AMS.
Before 41Ca is used clinically, it must be demonstrated that measurement of the isotope ratio after a 41Ca dose has clinical relevance. Our study is the first to show that 41Ca/Ca ratios in a diseased group are different from those in a control group. Other reports involving persons who received 41Ca describe data from single individuals (12)(13)(32). Using a 10-nCi 41Ca dose, Johnson et al. (32) were able to measure 41Ca/Ca ratios for more than 800 days and attributed small fluctuations in 41Ca/Ca to menstrual cycle phases. Freeman et al. (12) compared calcium kinetics determined with dual stable isotopes with 41Ca measurements and showed good agreement. They showed that a surgically menopausal female was losing 100 mg of calcium from bone per day (12). A separate publication by Freeman et al. (13) showed results suggesting that 41Ca/Ca measurements had less variability than N-telopeptide and that 41Ca/Ca ratios appear to decrease when bisphosphonates are administered (13). Although these earlier studies were limited in size, they suggest intriguing possibilities for applications of this technology.
It is difficult to prove that the 41Ca we administered ended up in the bone of the study participants, but several lines of evidence support this conclusion. It is well known that 99% of total body calcium is stored in bone. In addition, animal experiments have demonstrated that 24 h after administration of 45Ca, an average of 85% of the absorbed calcium is found in the skeleton (38). Using 41Ca, we have shown that 4 weeks after mice received 0.55 nCi of 41Ca, the concentration of 41Ca was 5000 times higher in bone than other soft tissue (39). Bronner et al.(40) showed that 5 days after adolescent boys received 45Ca intravenously, the total quantity of 45Ca excreted did not exceed 15% of the dose. These authors contended that nearly all of the calcium that enters the body at a given time is first retained, presumably in the skeleton (40). It should be noted that initially after an intravenous dose of labeled calcium, most of the label is found in quiescent surface that is in dynamic equilibrium with blood (41).
The rapid initial decrease in 41Ca/Ca that we observed in plasma of ESRD patients indicates a faster disappearance of 41Ca from the central compartment in the patients than in the controls. We hypothesize that the increased bone turnover in ESRD patients increases the available sites of mineralization and hence incorporation of the tracer into the bone formation process. This produces a more rapid initial clearance of 41Ca from the serum compartment. One could also argue that the 41Ca we administered to the ESRD patients was being cleared during hemodialysis, thus explaining the more rapid decrease of the tracer in these patients. Using 47Ca, however, Cochran et al. (35) demonstrated that ESRD patients with hyperparathyroidism retained 94%–98% of the 47Ca and that the dialysate and fecal radioactivity were minor components (35). Clearly, more studies are needed with 41Ca, but this technology offers the potential for monitoring calcium kinetics over a time period of years, using doses of tracer that pose a radiation risk that is too small to measure.
Management of renal osteodystrophy remains a difficult challenge faced by nephrologists. Current treatment guidelines suggest that in ESRD patients, nephrologists should aim to keep the circulating concentration of intact PTH in a range from 2 to 4 times the upper limit of the reference interval, although the form of PTH to be measured and the target concentration range remain controversial (25)(42). The need for improved tools for managing renal patients is clear because the clinical methods currently used to measure renal osteodystrophy are indirect. The ability to directly monitor calcium metabolism may improve our understanding of bone health, which is increasingly recognized as a predictor of long-term outcomes in dialysis patients. In a recent review of the impact of mineral metabolism on mortality and morbidity in >40 000 hemodialysis patients, the attributable risk associated with disorders of mineral metabolism (18%) was greater than that for inefficient dialysis (5%) or anemia (11%) (43).
In summary, we have characterized a dosing material in terms of chemical and radiologic purity in addition to sterility and pyrogenicity. This dose was administered to 8 human volunteers, and no adverse effects were observed. Using the methodology described in this report, we were able to measure 41Ca/Ca isotope ratios for more than 170 days in volunteers who received 10 nCi of 41Ca. The data presented show that this method can be used to monitor differences in calcium kinetics between healthy individuals and patients with ESRD and hyperparathyroidism.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  J. Skulan, T. Bullen, A. D. Anbar, J. E. Puzas, L. Shackelford, A. LeBlanc, and S. M. Smith Natural Calcium Isotopic Composition of Urine as a Marker of Bone Mineral Balance Clin. Chem., June 1, 2007; 53(6): 1155 - 1158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. K. Cheong, B. R. Martin, G. S. Jackson, D. Elmore, G. P. McCabe, J. R. Nolan, S. Barnes, M. Peacock, and C. M. Weaver Soy Isoflavones Do Not Affect Bone Resorption in Postmenopausal Women: A Dose-Response Study Using a Novel Approach with 41Ca J. Clin. Endocrinol. Metab., February 1, 2007; 92(2): 577 - 582. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. P. Heaney Measuring What Isn't There Clin. Chem., November 1, 2005; 51(11): 2003 - 2004. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, ß, or 
and 
represent the mean (SD) values for 41Ca/Ca in the controls (n = 4) and the ESRD patients on hemodialysis (n = 4), respectively. The area under the curve for the ESRD patients was significantly less than that for the controls (P < 0.005).