| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters |
Laboratorio Analisi Chimico Cliniche 1, Azienda Ospedaliera ‘Spedali Civili’, 25125 Brescia, Italy.
a Author for correspondence. Fax 39-0303995369; e-mail panteghi{at}osp.unibs.it
To the Editor:
Roche Diagnostics recently automated (1) (on the Elecsys® analyzer) the ß-CrossLapsTM assay (2) for measurement of bone-derived degradation products of type I collagen carboxy-telopeptide in serum (s-ßCTX). However, information on the biological variation of s-ßCTX is lacking, a fact that is limiting because the clinical utility of bone markers can be significantly affected by their physiological variations (3).
To investigate the biological variation of s-ßCTX, we took four blood specimens from each of 10 healthy volunteers (5 men and 5 premenopausal women; ages 25–50 years) on the same day once a week for 4 weeks. "Healthy" subjects were studied to ensure that any fluctuation in s-ßCTX concentrations could truly be attributed to biological variation and were not caused by pathological variations in bone homeostasis. In accordance with the Helsinki II Declaration, the design and execution of the experiment were explained thoroughly to the subjects, and informed consent was obtained. The inclusion criteria were that the subjects be within 20% of ideal body weight, have no bone or connective tissue disorder, and for women, have regular menstrual cycles and no use of hormonal contraceptives. In addition, none of the subjects took any medication or consumed substantial quantities of alcohol.
Venous blood was obtained between 0800 and 0900 from subjects who had fasted for 12 h and had not smoked or exercised that morning. Samples were collected by the same phlebotomist using vacuum collection tubes with minimal stasis while subjects were in the sitting position. Serum specimens, separated by centrifugation, were stored at -25 °C until analysis. When all of the specimens were available, they were thawed, mixed, centrifuged, and analyzed in a single run in duplicate in random order. s-ßCTX was measured using the ß-CrossLaps/Serum assay on the Elecsys 2010 analyzer.
The analytical (CVA), within-subject
(CVI), and between-subject
(CVG) components of variation were calculated by
nested ANOVA from replicate analyses (4). In
particular, CVA was estimated from the duplicate
results for each specimen, CVI from the serial
results for each subject, and CVG from the total
variance of the data minus the analytical and intraindividual
components. The desired analytical imprecision; the index of
individuality (CVI/CVG),
which yields information about the utility of conventional
population-based reference intervals; the critical difference [2.77
(CVA2 +
CVI2)1/2],
i.e., the minimal significant difference (P <0.05) between
two consecutive measurements of the marker in the same patient; and the
number of specimens that should be collected to estimate (P
<0.05) the homeostatic set point of an individual within ± 10%
[1.962
(CVA2 +
CVI2)/100] were also estimated
(Table 1
) (4).
|
Although the mean s-ßCTX values showed no sex-related difference, women had a lower within-subject variability and a higher interindividual fluctuation. This marked individuality of s-ßCTX in women, reflected by the low index of individuality, suggests that individuals may have values that are highly unusual for them but still within a population-derived reference interval, making the use of conventionally calculated reference limits inadequate for interpretation of s-ßCTX values in females. The results of recent analyses for an individual may best be used as a guide to possible pathology in that individual. Conversely, given the lower CVG and the high index of individuality in men, the use of reference limits may be of value in assessing unusual s-ßCTX results in male patients.
The desirable analytical imprecision for s-ßCTX, taken to be 
0.5
CVI, was 8.5%. Thus, in this limited
assessment, the intrabatch imprecision (CVA) of
the Elecsys CrossLaps assay met this goal. The critical difference was
51% for all subjects and 40% for women. Because preliminary reports
have shown mean changes of 60% or more in s-ßCTX concentrations in
females post treatment (2), our results suggest that
s-ßCTX may be of use in monitoring response to therapy in individual
patients. Thirteen samples are required to estimate an individual’s
s-ßCTX value to within 10% of the true mean value. Clearly, it is
impossible to obtain in clinical practice enough samples from a patient
to estimate that individual’s true s-ßCTX value, but at least two
samples should be obtained in the baseline evaluation of a subject to
significantly reduce the effect of biological variability on the
estimation of s-ßCTX concentrations.
Acknowledgments
We thank Francesca Stefini for skillful technical assistance and Roche Diagnostics for the gift of reagents to carry out the study.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
