Clinical Chemistry Link to Randox Laboratories Web Site
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 45: 1501-1509, 1999;
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sarno, M.
Right arrow Articles by Warnick, G. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sarno, M.
Right arrow Articles by Warnick, G. R.
(Clinical Chemistry. 1999;45:1501-1509.)
© 1999 American Association for Clinical Chemistry, Inc.

Articles

A Collection Method and High-Sensitivity Enzyme Immunoassay for Sweat Pyridinoline and Deoxypyridinoline Cross-Links

Mark Sarno1,a, Helen Powell2, Gayle Tjersland2, Donald Schoendorfer3, Holden Harris2, Kimberly Adams2, Peggy Ogata2 and G. Russell Warnick2

1 Vision Biotechnology Consulting, 306-N W El Norte Pkwy., PMB 311, Escondido, CA 92026.

2 Pacific Biometrics, Inc., 220 West Harrison, Seattle, WA 98119.

3 Sudormed, Inc., 12341 Newport Ave., Suite D-200, Santa Ana, CA 92705.
a Address for correspondence. Fax 760-634-3233; e-mail mjsarno{at}aol.com


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: Collagen cross-link molecules such as pyridinoline (PYD), deoxypyridinoline (DPD), and N-terminal cross-linked peptides (NTX) have been measured in urine as indices of bone resorption. However, very little is known regarding the excretion of pyridinolines into other biological fluids. We report a collection device, normalizing analyte, and high-sensitivity immunoassay for quantitative analysis of free pyridinoline cross-links in sweat.

Methods: Flame atomic emission and ion-selective electrode techniques were used to measure potassium as a sweat volume marker. The Pyrilinks immunoassay for urine free pyridinolines was optimized to increase sensitivity for measurements in sweat. The precision, accuracy, and detection limit of this assay were characterized. To assess values and variability of sweat pyridinolines in human subjects, a nonocclusive skin patch was used to collect sweat samples from a reference group and from a mixed group experiencing accelerated bone resorption, postmenopausal women and men receiving gonadotropin-releasing hormone for prostate cancer.

Results: The immunoassay intra- and interassay variations were <=10% and <16%, respectively, with a detection limit of 309 pmol/L. Linearity upon dilution and analytical recovery ranged from 93% to 109% and 85% to 122%, respectively. Sweat PYD values normalized to potassium output yielded a weekly intraindividual biological variability of 14.7%. The mean increase in the population experiencing increased bone resorption vs the reference group was 36% (P <0.05) for sweat PYD/K vs 23–40% (P <0.05) for urinary PYD/Cr, DPD/Cr, and NTX/Cr.

Conclusion: We conclude that this new platform sweat collection technology and PYD immunoassay show potential as an indicator of bone resorption.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
In recent years, quantitative methods for markers of bone metabolism have been developed. These markers encompass indicators of bone formation and bone resorption (1)(2)(3). In the case of resorption, the markers have generally been analyzed in urine. For instance, the pyridinoline (PYD)1 cross-links, PYD and deoxypyridinoline (DPD), which are the tripeptide cross-links of collagen type I, are continuously excreted into urine as a result of the process of bone resorption. Although these cross-linking molecules have been detected in aorta, dentine, ligament, and cartilage (PYD), the far greater mass and metabolic turnover rate in bone suggest that their presence in bodily fluids is essentially bone derived. To measure these markers, HPLC and immunoassay methods have been developed. Using these methods, the clinical utility of the PYD cross-links has been investigated extensively. Studies have shown significant correlations with bone turnover as assessed by histomorphometry (4) and radiotracer kinetics (5); increases in menopause because of estrogen deficiency, followed by restoration to normal concentrations after hormone replacement therapy (6)(7)(8); response to antiresorptive therapy using bisphosphonates (9)(10); and, associations with increased risk of fracture (11)(12). However, although the utility of these methods has been validated, significant diurnal (35–70%) (13)(14)(15) and day-to-day (10–27%) (16)(17)(18) variations have been observed. Additionally, 9–30% weekly (19) and 19–29% long-term (9)(20)(21) biological variabilities have been reported.

To reduce the observed diurnal and day-to-day variation, several groups have attempted to measure the PYD cross-links in other bodily fluids, including serum (22)(23) and plasma (23). However, single-point collections of these fluids are still subject to discernible diurnal and day-to-day variation. We have reported previously on detection of free PYD in sweat (24) and have recently developed a platform technology that uses a collection of sweat on a nonocclusive skin patch consisting of a transparent, hypoallergenic, gas-permeable membrane and a cellulose fiber absorbent pad. This device, known as the OsteopatchTM, is worn for 5 consecutive days, during which the nonvolatile components of sweat are deposited on the absorbent pad while the volatile components evaporate through the semipermeable membrane.

The 5-day collection period provides a time-integrated sample that attenuates the diurnal and day-to-day components of biological variation. This platform technology has been combined with a highly sensitive immunoassay we have developed for measurement of free PYD cross-links in human sweat. This assay uses a high-affinity monoclonal antibody that recognizes PYD and DPD cross-links with equal specificity. Additionally, in a manner analogous to urinary creatinine (Cr) correction, a marker of output coexcreted in sweat is required to normalize the PYD values for variations in sweat volume. This marker would ideally be one that is accurately measured with standardized methods in the clinical chemistry laboratory and is highly correlated with sweat output. Our studies suggest that sweat potassium meets these criteria. It is readily measured either by flame atomic emission spectroscopy (FAES) (25) or ion selective electrode (ISE) (26), and previous studies have demonstrated stable concentrations relatively independent of sweat rate (27)(28), thus indicating potential for high correlation to sweat output.

The present study describes the validation of potassium as a sweat volume marker and the development and characterization of the high-sensitivity assay for sweat PYDs. Additionally, to assess clinical feasibility, the assay was used in conjunction with the Osteopatch collection device to determine values of sweat PYD cross-links in populations of individuals with normal and accelerated bone turnover because of gonadotropin suppression. The reference group consisted of premenopausal women and healthy men, whereas the population with expected accelerated turnover consisted of postmenopausal women receiving no antiresorptive therapy and men on gonadotropin-releasing hormone (GnRH) therapy for prostate cancer (a therapy expected to increase bone resorption) (29). Lastly, comparisons to established urinary analytes collected under tightly controlled conditions were performed.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Materials
The absorbent pad, release liner, and polyurethane adhesive liner components of the Osteopatch were manufactured by 3M. PYD immunoassay reagents, including PYD-coated microplates, monoclonal anti-PYD alkaline phosphatase conjugate, PYD calibrators, controls, wash buffer, substrate, diethanolamine buffer, and stop solution, were obtained from Metra Biosystems, Inc. using process modifications developed by our group (as described in Results). Potassium analysis reagents, including standards and controls, were purchased from Beckman Instruments, Inc. Acetic acid, potassium chloride, sodium chloride, calcium chloride dihydrate, glycine, urea, sodium azide, 300 g/L bovine serum albumin solution, Tween 20, and mono and dibasic sodium phosphate were obtained from Sigma Chemical Co. Isopropanol swabs were obtained from Triad Medical, Inc.

procedures
Sweat collection on the Osteopatch.
Patches were applied to the abdomen or lower back (lateral to the midline) of each subject after cleansing of the sites with an isopropanol swab. Subjects were allowed to perform routine daily activities, including showering and exercise, during the specified period of wear.

Extraction of collected analyte.
After the specified period of patch wear, which was dependent on the experimental design (typically 5 days, but less in some experiments, as indicated), each absorbent pad was removed and shipped to the centralized testing site in a sealed pouch. Upon receipt at the laboratory, the pad was rolled tightly and placed in a microfuge tube. One mL of a 10 mmol/L acetic acid, 0.1 mL/L Tween 20 (pH 4.50) buffer was added to each tube. The tube was then agitated for 60 min at room temperature. The extracted material was then centrifuged at 3000g for a period of 10 min to isolate the liquid extract. Each sample was then dried by rotary evaporation on a Labconco Centrivap instrument and reconstituted in 0.4 mL of a 100 mmol/L sodium phosphate (pH 8.0) buffer. Samples were then recentrifuged at 1400g for 5 min to remove any pad fibers before analysis.

Potassium measurement.
Sweat potassium concentration was determined both by FAES as described previously and by ion-selective electrode on the Beckman Instruments Synchron CX®3 analyzer. The ISE chemistry uses an electrode containing a valinomycin membrane in conjunction with a reference electrode to determine potassium activity via indirect potentiometry. Correlation between the FAES and ISE methods was performed by analysis of Osteopatch extracts from 18 healthy subjects.

Recovery of potassium.
Potassium (1.3 µmol) was added to 32 patches and allowed to dry. Potassium was extracted as described previously, lyophilized, and reconstituted in 0.4 mL of 100 mmol/L sodium phosphate (pH 8.0). Potassium was measured by FAES, and the recovery was determined vs the added amount.

Potassium vs sweat volume.
Sixteen Osteopatches were modified to contain 20 layers of absorbent pad material. These patches were then covered with occlusive material [1509 double-coated medical tape (3M Corp.) laminated to Saran® plastic film (Dow Brands)] to prevent escape of volatile constituents. The patches were then weighed and applied to the abdomen of two subjects (four patches each). The patches were worn for a period of 7.5 h, and the experiment was repeated on a second day by the same subjects. On the second day of patch wear, one patch on subject 2 had become detached and was judged to be compromised. After removal of the remaining 15 patches, the patches were reweighed, and the mass was converted to sweat volume using specific gravity. Potassium was then determined by FAES. The results were expressed as µmol potassium/patch vs µL of sweat collected.

Urinary resorption marker assays.
First morning void (FMV) and 24-h urine collections were performed under strict sampling conditions. Specifically, on the first day of collection, the first morning urine was discarded. After this, the next 24 h of urine was collected. After this collection, the FMV from the following day was collected. The FMV collection time was the same period during which the first morning urine was discarded on the previous day. After collecting a small aliquot of the FMV for urinary resorption marker analysis, the remainder of the FMV was combined with the 24-h collection. The PyrilinksTM and Pyrilinks-D assays for free urinary PYD and DPD (Metra Biosystems) were performed as described previously (30)(31). The OsteomarkTM assay (Ostex International) for the N-telopeptides of type I collagen was also performed as described previously (32).

High-sensitivity sweat pyridinoline immunoassay.
PYD calibrators and controls were diluted 1:10 with 100 mmol/L sodium phosphate buffer (pH 8.0). One hundred microliters of calibrators, controls, and samples were then added to duplicate wells of a PYD-coated microplate. Fifty microliters of reconstituted anti-PYD-alkaline phosphatase conjugate was then added to each well, and the plate was incubated overnight at 2–8 °C without agitation. Each well was washed three times with wash buffer, and 150 µL of p-nitrophenyl phosphate in diethanolamine was added afterward. The substrate was allowed to incubate for a period of 5 h at room temperature. One hundred microliters of Stop Solution (1 mol/L NaOH) was added to each well. The absorbance of each well was then read with a SpectraMax 340 microplate reader (Molecular Devices, Inc.) set at 405-nm wavelength, and the recoveries of controls and unknowns were calculated from a four-parameter fit of the calibrators. The recoveries of unknowns were then normalized for sweat volume via calculation of the PYD/K ratio (nmol PYD/mol K).

Clinical study design.
After Institutional Review Board approval, 40 subjects were recruited. Informed consent was acquired from all subjects. The reference group included 13 premenopausal women and 5 apparently healthy men. The postmenopausal/GnRH (PM/GnRH) population included 19 postmenopausal females who were receiving no antiresorptive therapy and 3 males on GnRH therapy (Lupron) for prostate cancer. The PM/GnRH mixed population was chosen because both male and female subjects were experiencing gonadotropin suppression, which would be expected to induce accelerated bone resorption, and because it was relatively easy to recruit patients. Each subject wore four Osteopatches for 5 days on three successive occasions during a period of 4 weeks. During each 5-day period, each subject also gave a FMV and 24-h urine sample under strict sampling conditions as described previously. No patches were found to be detached or otherwise compromised at the end of the wear period, and no subject reported an adverse event during or after the study.

data analysis
Addition and recovery.
To assess recovery of exogenous PYD added to human sweat, three pools were prepared from sweat extracts from 10 subjects, each of whom wore eight patches. Two concentrations of PYD were added to each extract pool. The percentage of recovery was calculated using the formula:

Biological variation.
The week-to-week intraindividual biological variation of sweat and urinary resorption markers was calculated using the following equation:

where SI = biological variation, ST = total variation, and SA = analytical variation. The total variation was calculated by computing the CV (%) of the corrected marker values from all replicates and all runs of the samples from the three successive periods of sample collection. Therefore, biological and analytical variation are captured in this computation. The analytical variation was calculated by computing the mean CV of corrected marker values from all replicates and all analytical runs of each period of sample collection individually. In this experimental design, the biological variability as calculated for the sweat PYD/K system included the component from variability of manual pad processing because the contributions of the extraction, evaporation, and centrifugation steps were not isolated.

Statistical analysis of clinical study results.
The distribution of the reference and PM/GnRH populations was tested using the Wilk-Shapiro normality test. For this test, the null hypothesis was that the populations were not normally distributed. Therefore, P <0.05 is considered an indication of non-normality. Discrimination between the reference and PM/GnRH populations was assessed using the Student t-test in the case of normally distributed populations and the Mann–Whitney test for nonnormally distributed populations. Additionally, T-scores were computed by subtracting the mean of the reference group from the mean of the PM/GnRH group and subsequently dividing by the SD of the reference group population. This calculation is equivalent to the calculation of T-scores in bone mineral densitometry. Correlations among sweat and urine markers and with subject age were performed using Spearman's analysis because of non-gaussian distribution of most data sets. In all cases, JMP software (SAS Institute) was used for the statistical investigation.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
correlation of FAES and ISE
Extracts from Osteopatches worn by 18 healthy subjects were used to determine the correlation between the two potassium determination methodologies. The slope of the regression line (Fig. 1 ) was 1.01, and the Pearson correlation (r) of 0.9987 is significant at P <0.0001. On the basis of these results, FAES and ISE methods were used interchangeably throughout the rest of the studies performed.



View larger version (15K):
[in this window]
[in a new window]
  		Figure 1. Sweat potassium determined by FAES and ISE.

r = 0.9987; P <0.0001; n = 18; y = -0.09; slope = 1.01.

sweat potassium as a normalizing marker
To assess variation in the extraction procedure, known amounts (1.3 µmol) of potassium were added to patches (n = 32). After extraction, lyophilization, and reconstitution, potassium was determined by FAES. The recovery was found to be 95.4% ± 6.6%. The Pearson correlation (r) between potassium excreted and collected on the patch during a 7.5-h period and sweat volume was found to be 0.9223, which was significant at P <0.0001 (Fig. 2 ). However, although this short-term collection experimental design was efficient for determining the correlation, it did not assess the uniformity of potassium output over a 5-day period, which is the intended collection period of the Osteopatch. To test this, a study was performed on three subjects, each of whom wore patches for a period of five days. During that period, each subject also wore five 1-day patches (removed after 24 h). The 1-day and multiday patches were extracted as described previously and then analyzed for potassium by ISE. The results were determined as the sum of the 1-day electrolyte determinations vs the cumulative 5-day collection. The percentage of recoveries of the sum single day values ranged from 88.7% to 102.7% of the cumulative 5-day patch values.



View larger version (11K):
[in this window]
[in a new window]
  		Figure 2. Total sweat potassium from patches worn by five healthy subjects vs total volume of sweat collected.

r = 0.9223; P <0.0001; n = 15; y = -0.06.

high-sensitivity sweat PYD immunoassay
The sensitivity of the Metra Pyrilinks kit was optimized for the concentrations of PYD detected in urine. However, we have found the concentrations in sweat to be less than one-tenth of those in urine (24). To optimize the sensitivity of the system, factorial experiments were performed. Optimization of the amount of PYD coupled to the solid phase, sample volume, and conjugate dilution resulted in an 11.4-fold increase in sensitivity as measured by the estimated dose at 80% (ED80) of the binding in the absence of antigen (B0). Competitive inhibition curves for the standard Pyrilinks assay in comparison to the high-sensitivity assay are shown in Fig. 3 . The ED80 for the Pyrilinks assay was 2840 pmol/L PYD vs 250 pmol/L for the high-sensitivity sweat assay.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 3. Competitive inhibition curves for the urinary Pyrilinks test (•), r = 0.9999, and the high-sensitivity sweat test ({blacksquare}), r = 0.9998.

Intra- and interassay reproducibility were tested using solutions of PYD added at various concentrations into an artificial sweat matrix of 7 mmol/L KCl, 4 mmol/L NaCl, 0.25 mmol/L CaCl2, 1.67 mmol/L glycine, 1.5 mmol/L NaN3, and 0.1 g/L bovine serum albumin. For the intraassay study, 20 replicates of each solution were tested in a single assay; for the interassay investigation, duplicates of each solution were run in 16 total assays on a single day of testing. Table 1 shows the mean recovery (pmol/L PYD), SD, and CV. Intra- and interassay variation ranged from 3.9% to 10.0% and 7.1% to 15.2%, respectively.


View this table:
[in this window]
[in a new window]
  		Table 1. Intra- and interassay reproducibility for five solutions with added PYD.1

To assess patient sample dilution and recovery, three solutions of PYD added at various concentrations to the artificial sweat matrix were serially diluted with the assay zero diluent/calibrator. The recovery of the undiluted sample was used as the reference value from which expected values were calculated. The percentage of recovery was calculated as the observed value divided by the expected value. The results (Table 2 ) show percentage of recoveries ranging from 93% to 109%.


View this table:
[in this window]
[in a new window]
  		Table 2. Linearity upon dilution.

To assess recovery of exogenous PYD added into human sweat, three pools were prepared from sweat extracts from 10 subjects, each of whom wore eight patches. Two concentrations of PYD were added into each extract pool. The PYD concentrations were measured in triplicate before and after addition of exogenous PYD. The results (Table 3 ) show recoveries ranging from 85% to 122% with a mean analytical recovery of 102.8%.


View this table:
[in this window]
[in a new window]
  		Table 3. Recovery.

The detection limit was tested by assay of the zero diluent/calibrator (0 pmol/L PYD) in replicates of 20. The minimum detectable concentration (MDC) was defined as the concentration of PYD corresponding to the response, in milliabsorbance (mA) units, that was 2 SDs less than the mean absorbance of the zero diluent/calibrator. The experiment was repeated in 16 individual analytical runs, and the MDC was set at the upper limit of the 95% confidence interval. The MDC was 309 pmol/L PYD.

The total, analytical, and weekly intraindividual biological variabilities were 16.2%, 6.8%, and 14.7%, respectively, for sweat PYD/K. These values were equivalent to the results for the 24-h urinary markers, which displayed slightly lower values than FMV urinary PYD/Cr, DPD/Cr, and NTX/Cr (Table 4 ). To determine whether the menstrual cycle had an effect on biological variation in the sweat system, the data were reanalyzed separately for premenopausal and postmenopausal subjects. The intraindividual biological variation in premenopausal subjects was found to be 15.3%, which was not significantly different from the postmenopausal subjects (Student's t-test, P >0.05); the biological variability was 14.5%.


View this table:
[in this window]
[in a new window]
  		Table 4. Total, analytical, and weekly intraindividual biological variation.1

normality analysis and population discriminating ability
Comparisons of the Osteopatch results were made with the results obtained by FMV and 24-h urinary PYD/Cr, DPD/Cr, and NTX/Cr. For the reference group of 5 healthy males and 13 premenopausal females, the populations were normally distributed for sweat PYD/K, FMV, and 24-h urinary PYD/Cr, FMV DPD/Cr, and 24-h NTX/Cr (P >0.05). FMV NTX/Cr and 24-h DPD/Cr were nonnormally distributed (P <0.05). For the PM/GnRH group, the data were normally distributed for FMV and 24-h PYD/Cr, and FMV and 24-h DPD/Cr (P >0.05). All other markers were nonnormally distributed (P <0.05).

Table 5 shows the Mann–Whitney and Student t-test results and the means and SDs for the reference and PM/GnRH groups for all markers. The mean PYD/K excretion in sweat of the reference group was 119.7 ± 31.2 nmol/mol (mean ± SD). A preliminary reference interval was determined as a nonparametric 10th–90th percentile and encompassed the range of values between 78.6 and 175.9 nmol/mol. The mean PYD/K excretion in sweat for the PM/GnRH group was 163.3 nmol/mol, which was significantly higher (P < 0.05) than the reference group (Fig. 4 ). The ratio of the PM/GnRH group mean to the reference group mean was 1.36, representing a 36% increase. This discriminatory power was not changed if the three male GnRH subjects were removed from the analysis.


View this table:
[in this window]
[in a new window]
  		Table 5. Discrimination of subject populations.



View larger version (11K):
[in this window]
[in a new window]
  		Figure 4. Levels of sweat PYD/K in the reference and PM/GnRH groups.

Discrimination of the population is significant at P <0.05 by the Mann–Whitney test.

The mean FMV and 24-h PYD/Cr, DPD/Cr, and NTX/Cr values for the PM/GnRH population were similarly increased. Only the 24-h NTX/Cr data showed a higher mean increase than PYD/K of the PM/GnRH population values to the reference group values (1.40). The FMV PYD/Cr showed the lowest mean increase (1.23).

To account for the impact of variation on discrimination of the populations, a T-score analysis was also performed (Table 5Up ). The T-score for sweat PYD/K is 1.40 which with one exception, is comparable with the scores for the urinary markers. The T-score for FMV NTX/Cr was 0.60, which is substantially lower than the scores for sweat PYD/K and the other urinary markers. The low T-score for FMV NTX/Cr is primarily because of the high SD of NTX/Cr values in the reference group.

Correlations between sweat and urine markers and with subject age were performed by Spearman's analysis. The results (Table 6 ) show significant correlations (P < 0.05) between sweat PYD/K and urinary PYD/Cr and NTX/Cr but not urinary DPD/Cr. Correlations with subject age were significant for all markers except FMV and 24-h NTX/Cr measurements.


View this table:
[in this window]
[in a new window]
  		Table 6. Correlations (Spearman rho) between sweat and urine markers and subject age.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The present work describes a new quantitative diagnostic platform technology that can be applied to measuring many analytes in sweat. This technology consists of a combination of a sweat collection device, a marker of sweat output, and a quantitative immunoassay. The sweat collection device used has been applied previously to qualitative measurement of drugs of abuse (33). Its first application to a quantitative system is the high-sensitivity immunoassay specific for free sweat PYD and DPD cross-links reported in this study. This method is an alternative to their measurement in urine.

Potassium has been identified as the marker of sweat output and has been shown to be reliably measured by standard laboratory methods. This marker is consistently recovered from the patch and is highly correlated with sweat volume. Furthermore, Dill et al. (34) have shown that sweat potassium is insensitive to subject age, whereas sweat sodium and chloride concentrations were positively correlated, making them unsuitable as volume markers. To determine whether hydration might influence sweat potassium output, Costill et al. (35) dehydrated subjects to -3% of their body weight through controlled exercise. During certain days, lost fluid was replaced with water only, and on other days, it was replaced with a glucose-electrolyte solution. In this study, there was no significant difference in sweat potassium concentrations as a result of the replacement regimen. To assess whether dietary variation might compromise sweat potassium measurements, Lane et al. (36) studied the effects of diet on sweat composition. One study group consumed a National Aeronautics and Space Administration space diet containing 2.8 g/day of potassium. The other group consumed a laboratory diet containing 3.3 g/day of potassium. Again, there was no significant difference in sweat potassium concentration, whereas sweat sodium and chloride concentrations varied with dietary intake concentrations.

The immunoassay for PYD reported in this study has been optimized to increase sensitivity and has been shown to be accurate for measurement of PYDs in sweat. The assay is based on a high-affinity monoclonal antibody that has been shown to be specific for free PYD cross-links. The antibody recognizes free PYD and DPD with equal specificity, whereas the cross-reactivity of the antibody with amino acids and PYD-containing peptides of >1000 Da is <5% and 2.5%, respectively (30).

Although the sweat collection technology does not allow the analysis of diurnal variation because of the low concentrations of PYD excreted within 24 h, the weekly intraindividual biological variation of the sweat measurement has been shown to be minimal. The results demonstrate variability equivalent to resorption markers measured in 24-h urine collections and lower than urinary markers in FMV collections. Additionally, it should be noted that our calculation of biological variation for the sweat PYD/K measurements incorporated the contribution of variation in the manual pad processing procedure (extraction, evaporation, and centrifugation). Therefore, the true biological variation of sweat PYD measurements is likely to be lower than the 14.71% reported in this study. This may be further reduced once the pad processing procedure is automated.

There is no significant difference in intraindividual biological variation between pre- and postmenopausal females in our study. This is in contrast to a recent report by Gorai et al. (37), which demonstrated significant variation during the menstrual cycle.

With regard to population discrimination, statistical assessments of the distribution of the PM/GnRH population are suggestive of the ability of the sweat PYD system to identify subjects with increased bone resorption. When expressed as T-scores, the sweat system displays performance second only to 24-h urinary DPD/Cr measurements in the population of subjects with gonadotropin suppression evaluated in this study. Correlations between sweat PYD/K and the urinary markers were generally lower than the correlations between the urinary analytes themselves. This may be because of differences in excretion pathways or to renal processing of peptide-bound PYDs to free PYDs, a finding reported recently by Colwell and Eastell (38).

In conclusion, in combination with the platform collection technology, this new marker shows potential as an indicator of bone resorption. The findings in this study suggest that the sweat system is a noninvasive alternative to urine collections. Furthermore, this technology may eventually be suitable for prescription self-use, in a fashion similar to fecal occult blood testing. However, additional cross-sectional and longitudinal studies with disorders such as osteoporosis, Paget disease of bone, hyperthyroidism, and hyperparathyroidism are indicated and will be useful in further defining the clinical utility of the system.

conflict of interest
All authors are affiliated with Pacific Biometrics, Inc. as either direct employees, consultants (Mark Sarno), or as collaborators under contract (Donald Schoendorfer).


   Acknowledgments
 
We thank Emory Wright of Metra Biosystems for technical assistance in developing the plate-coating conditions for the high-sensitivity sweat PYD immunoassay.


   Footnotes
 
1 Nonstandard abbreviations: PYD, pyridinoline; DPD, deoxypyridinoline; Cr, creatinine; FAES, flame atomic emission spectrometry; ISE, ion-selective electrode; PM/GnRH, mixed population of postmenopausal women and men on gonadotropin-releasing hormone therapy for prostate cancer; FMV, first morning void; ED80, estimated dose at 80% of the B0; B0, binding signal in a competitive immunoassay in the absence of antigen; and MDC, minimal detectable concentration.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Delmas PD. Biochemical markers of bone turnover. J Bone Miner Res 1993;8(Suppl 2):S549-S555.
  2. Kleerekoper M. Biochemical markers of bone remodeling. Am J Med Sci 1996;8:270-277.
  3. van Daele PL, Birkenhager JC, Pols HA. Biochemical markers of bone turnover: an update. Neth J Med 1994;44 2:65-72.[ISI][Medline] [Order article via Infotrieve]
  4. Delmas PD, Schlemmer A, Gineyts E, Riis B, Christiansen C. Urinary excretion of pyridinoline crosslinks correlates with bone turnover measured on iliac crest biopsy in patients with vertebral osteoporosis. J Bone Miner Res 1991;6:639-644. [ISI][Medline] [Order article via Infotrieve]
  5. Eastell R, Colwell A, Hampton L, Reeve J. Biochemical markers of bone resorption compared with estimates of bone resorption from radiotracer kinetic studies in osteoporosis. J Bone Miner Res 1997;12:59-65. [ISI][Medline] [Order article via Infotrieve]
  6. Hassager C, Colwell A, Assiri AM, Eastell R, Russell RG, Christiansen C. Effect of menopause and hormone replacement therapy on urinary excretion of pyridinium cross-links: a longitudinal and cross-sectional study. Clin Endocrinol 1992;37:45-50. [Medline] [Order article via Infotrieve]
  7. Uebelhart D, Schlemmer A, Johansen JS, Gineyts E, Christiansen C, Delmas PD. Effect of menopause and hormone replacement therapy on the urinary excretion of pyridinium cross-links. J Clin Endocrinol Metab 1991;72:367-373. [Abstract]
  8. Melton LJ, Khosla S, Atkinson EJ, O'Fallon WM, Riggs BL. Relationship of bone turnover to bone density and fractures. J Bone Miner Res 1997;12:1083-1091. [ISI][Medline] [Order article via Infotrieve]
  9. Garnero P, Shih WJ, Gineyts E, Karpf DB, Delmas PD. Comparison of new biochemical markers of bone turnover in late postmenopausal osteoporotic women in response to alendronate treatment. J Clin Endocrinol Metab 1994;79:1693-1700. [Abstract]
  10. Harris S, Gertz B, Genant H, Eyre D, Survill T, Ventura J, et al. The effect of short term treatment with alendronate on vertebral density and biochemical markers of bone remodeling in early postmenopausal women. J Clin Endocrinol Metab 1993;76:1399-1406. [Abstract]
  11. van Daele PL, Seibel MJ, Burger H, Hofman A, Grobbee DE, van Leeuwen JP, et al. Case-control analysis of bone resorption markers, disability, and hip fracture risk: the Rotterdam study. Br Med J 1996;312:482-483. [Free Full Text]
  12. Garnero P, Hausherr E, Chapuy MC, Marcelli C, Grandjean H, Muller C. Markers of bone resorption predict hip fracture in elderly women: the EPIDOS prospective study. J Bone Miner Res 1996;11:1531-1538. [ISI][Medline] [Order article via Infotrieve]
  13. Blumsohn A, Herrington K, Hannon RA, Shao P, Eyre DR, Eastell R. The effect of calcium supplementation on the circadian rhythm of bone resorption. J Clin Endocrinol Metab 1994;79:730-735. [Abstract]
  14. Eastell R, Calvo MS, Burritt MF, Offord KP, Russell RGG, Riggs BL. Abnormalities in circadian patterns of bone resorption and renal calcium conservation in type 1 osteoporosis. J Clin Endocrinol Metab 1992;74:487-494. [Abstract]
  15. Pagani F, Panteghini M. Diurnal rhythm in urinary excretion of pyridinium crosslinks. Clin Chem 1994;40:952-953. [Free Full Text]
  16. DePriest SG, Cerelli MJ, To WL, Freeman KD, Merrell L, Hesley RP, et al. The effect of day to day and diurnal variation on two immunoassays for pyridinium crosslinks in urine. J Bone Miner Res 1995;10(Suppl 1):S473.
  17. Blumsohn A, Hannon RA, Al-Dehaimi AW, Eastell R. Short-term intraindividual variability of markers of bone turnover in healthy adults. J Bone Miner Res 1994;9(Suppl 1):S153.
  18. McLaren AM, Isdale AH, Whiting PH, Bird HA, Robins SP. Physiological variations in the urinary excretion of pyridinium crosslinks of collagen. Br J Rheumatol 1993;32:307-312. [Abstract/Free Full Text]
  19. Jensen JEB, Sorenson HA, Kollerup G, Jensen LB, Sorenson OH. Biological variation of biochemical bone markers. Scand J Clin Lab Investig 1994;54(Suppl 219):36-39.
  20. Kleerekoper M, Wilson PS, Simpson P. Within subject variability of biochemical markers of bone remodeling in normal older women. J Bone Miner Res 1994;9(Suppl 1):S394.
  21. Morris HA, Wishart JM, Horowitz M, Need AG, Nordin BEC. The reproducibility of bone-related biochemical variables in post-menopausal women. Ann Clin Biochem 1990;27:562-568.
  22. James I, Crowley C, Perrett D. Assay of pyridinium crosslinks in serum using narrow-bore ion-paired reversed-phase high-performance liquid chromatography. J Chromatogr 1993;612:41-48. [ISI][Medline] [Order article via Infotrieve]
  23. Abbiati G, Rigoldi M, Frignani S, Colombo L, Mussini E. Determination of pyridinium crosslinks in plasma and serum by high-performance liquid chromatography. J Chromatogr B 1994;656:303-310. [ISI][Medline] [Order article via Infotrieve]
  24. Warnick GR, Harris HH, Doneen BD. Identification of bone-loss markers, pyridinoline (PYD) and deoxypyridinoline (dPYD) in sweat by HPLC [Abstract]. Clin Chem 1997;43:S177.
  25. Velapoldi RA, Paule RC, Schaffer R, Mandel J, Moody JR. A reference method for the determination of potassium in serum. Special Publication 260-263 Gaithersburg: National Bureau of Standards, 1979..
  26. Mercer D, Huczko L, Ingram K. Precision and linearity studies on the Beckman Synchron CX®3 Delta [Abstract]. Clin Chem 1995;41:S186.
  27. Sato K, Kang WH, Saga K, Sato KT. Biology of sweat glands and their disorders. I. Normal sweat gland function. J Am Acad Dermatol 1989;20:537-563. [ISI][Medline] [Order article via Infotrieve]
  28. Sato K, Sato F. Na+, K+, H+, Cl-, Ca2+ concentrations in cystic fibrosis eccrine sweat in vivo and in vitro. J Lab Clin Med 1990;115:504-511. [ISI][Medline] [Order article via Infotrieve]
  29. Campodarve I, Chesnut CH, Dugowson C, Gottesman J, Kiviat MD, Wonderly R, et al. Do men treated with leuprolide for prostate cancer lose bone?. J Bone Miner Res 1995;10(Suppl 1):S343.
  30. Gomez B, Ardakani S, Evans BJ, Merrell LD, Jenkins DK, Kung VT. Monoclonal antibody assay for free urinary pyridinium cross-links. Clin Chem 1996;42:1168-1175. [Abstract/Free Full Text]
  31. Robins SP, Woitge H, Hesley RP, Ju J, Seyedin S, Seibel MJ. Direct, enzyme-linked immunoassay for urinary deoxypyridinoline as a specific marker for measuring bone resorption. J Bone Miner Res 1994;9:1643-1649. [ISI][Medline] [Order article via Infotrieve]
  32. Hanson D, Weis M, Bollen A, Maslan S, Singer F, Eyre D. A specific immunoassay for monitoring human bone resorption: quantitation of type I collagen cross-linked n-telopeptides in urine. J Bone Miner Res 1992;7 11:1251-1258.[ISI][Medline] [Order article via Infotrieve]
  33. Spiehler V, Fay J, Fogerson R, Schoendorfer D, Niedbala R. Enzyme immunoassay for qualitative detection of cocaine in sweat. Clin Chem 1996;42:34-38. [Abstract/Free Full Text]
  34. Dill BD, Horvath SM, Van Beaumont W, Gehlsen G, Burrus K. Sweat electrolytes in desert walks. J Appl Phys 1967;23:746-751. [Free Full Text]
  35. Costill D, Coté R, Miller E, Miller T, Wynder S. Water and electrolyte replacement during repeated days of work in the heat. Aviat Space Environ Med 1975;46:795-800. [Medline] [Order article via Infotrieve]
  36. Lane HW, Roessler G, Nelson E, Cerda J. Effect of physical activity on human potassium metabolism in a hot and humid environment. Am J Clin Nutr 1978;31:838-843. [Abstract/Free Full Text]
  37. Gorai I, Taguchi Y, Chaki O, Kikuchi R, Nakayama M, Yang BC, et al. Serum soluble interleukin-6 receptor and biochemical markers of bone metabolism show significant variations during the menstrual cycle. J Clin Endocrinol Metab 1998;83:326-332. [Abstract/Free Full Text]
  38. Colwell A, Eastell R. The renal clearance of free and conjugated pyridinium crosslinks of collagen. J Bone Miner Res 1996;11:1976-1980. [ISI][Medline] [Order article via Infotrieve]



The following articles in journals at HighWire Press have cited this article:


Home page
 Clin. Chem.Home page
H. W. Vesper, L. M. Demers, R. Eastell, P. Garnero, M. Kleerekoper, S. P. Robins, A. K. Srivastava, G. R. Warnick, N. B. Watts, and G. L. Myers
Assessment and Recommendations on Factors Contributing to Preanalytical Variability of Urinary Pyridinoline and Deoxypyridinoline
Clin. Chem., February 1, 2002; 48(2): 220 - 235.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sarno, M.
Right arrow Articles by Warnick, G. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sarno, M.
Right arrow Articles by Warnick, G. R.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS