| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Diagnostic Systems Laboratories (Canada), Inc. (A.D., M.J.K.), and the Department of Laboratory Medicine and Pathobiology, Faculty of Medicine, University of Toronto (M.J.K.), Toronto, Ontario, Canada; Diagnostic Systems Laboratories, Inc. (J.M.), Webster, Texas 77598; and the Institute of Endocrinology, Metabolism, and Reproduction (V.M., J.G.-A.), Quito, Ecuador
Address all correspondence and requests for reprints to: M. J. Khosravi, Ph.D., Diagnostic Systems Laboratories (Canada), Inc., Mount Sinai Hospital, Room 653, 600 University Avenue, Toronto, Ontario, Canada M5G 1X5.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The clinical assessment of GH status has been controversial, primarily due to the episodic nature of GH secretion, its relatively short circulating half-life, and the considerable variability in GH measurement by different assays (11, 12). Currently, evaluation of GH sufficiency may involve multiple venous blood samplings for determination of pituitary GH secretion in response to a number of physiological or pharmacological stimuli. Because of the reported limitations of the provocative GH testing, alternative screening procedures have been sought (11, 12). As blood IGF-I and IGFBP-3 are highly dependent on GH secretion (4, 5, 6, 7, 8, 9, 13), determination of their levels has recently been recognized as the most effective means in the evaluation of GH-IGF axis status, particularly in children with short stature (12).
In children, blood sampling by way of venipuncture has been problematic. Alternative procedures involve collection of blood by capillary puncture from the heel, finger, or earlobe. Capillary blood dried on filter paper is well established and has found world-wide applications in a number of large scale infant and population screening programs (14, 15, 16, 17, 18). The procedure requires a small volume of blood, is less traumatic than venipuncture, and has been tightly regulated (19) to support a reliable and safe specimen collection and delivery system at a significantly low cost. The latter considerations may assume greater importance for applications requiring sample collection and transportation to distant laboratories and when transportation of liquid specimens may not be feasible and/or practical.
We here describe application of whole blood collected on filter paper in the analysis of IGF axis components. We demonstrate distribution of IGF-I and IGFBP-3 in the plasma fraction of whole blood and provide evidence that blood dried on filter paper is an ideal specimen for the determination of circulating IGF-I and IGFBP-3 levels.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Paired plasma and capillary dried blood spots from subjects with untreated GH receptor deficiency [GHRD; five males, aged 1–32 yr (mean age ± SD, 17 ± 14.5); and six females, aged 2.3–65 yr (mean age, 29.9 ± 25 yr)] and age-matched normal controls [8 males, aged 1–27 yr (mean age, 13.3 ± 10.4 yr); and 8 females, aged 2–67 yr (mean age, 36.6 ± 23.7)] were collected from the patient population at the Institute of Endocrinology, Metabolism, and Reproduction (Quito, Ecuador). Ecuadorian GHRD patients were selected for this study because they have extremely low levels of IGF-I and IGFBP-3 (20). The sample collection protocol was approved by the ethics committee of Institute of Endocrinology, Metabolism, and Reproduction in compliance with the laws and regulations of the United States and Ecuador. All subjects and/or their guardians signed the approved informed consent form. The samples were collected as described below, transferred to Diagnostic Systems Laboratories (Webster, TX), and stored until use.
Random whole blood samples (n = 45), collected by venipuncture in the presence of ethylenediamine tetraacetate (EDTA), were obtained from the clinical laboratory at Mount Sinai Hospital (Toronto, Canada). The samples were from an adult population and were processed within 48 h for preparation of matched whole blood, plasma, and dried blood spots as described below.
Blood collection filter paper cards (no. 903) were obtained from Schleicher and Schuell (Keene, NH). The paper is manufactured according to the specifications set by the national committee for clinical laboratory standards for blood collection (19). All other chemical reagents were of highest quality and were obtained from Sigma Chemical Co. (St. Louis, MO) or Amresco (Solon, OH). Forty-eight-well cell culture plates were products of Costar (Cambridge, MA).
Sample preparation procedures
One drop of capillary blood, obtained from the fingertip, or EDTA-whole blood in a disposable pipette was applied to the filter paper to form a discrete spot and was allowed to air-dry at room temperature for 3 h. The dried blood spots were stored in plastic bags at room temperature unless otherwise indicated. For collection of plasma, the EDTA-whole blood samples were allowed to stand at room temperature for 2–4 h, and plasma was transferred to sample tubes. Plasma samples not assayed within 48 h were stored at -20 or -70 C and used within 1–4 weeks.
Extraction of IGF-I and IGFBP-3 from dried blood spots
All analysis were performed using a single 8-mm diameter dried blood filter paper disc with an impregnation whole blood volume of approximately 13 µL. The blood spot disc was punched out by a manual paper puncher from an area completely impregnated with blood. The disc was placed into a well of a 48-well cell culture plate and extracted for IGF-I or IGFBP-3 analysis. The extraction methods were optimized for extraction buffers, protocol, and incubation time.
For optimal extraction of IGF-I, 200 µL extraction buffer (0.005 mol/L Tris, pH 7.0, and 0.5 mL/L Tween-20) were added per well containing a single dried blood spot disc, and the wells were shaken (500–700 rpm) at room temperature for 1 h. To dissociate IGF-I from IGFBPs before analysis, the extract was treated to stepwise addition of acidification and neutralization buffers developed by Diagnostic Systems Laboratories for nonextraction IGF-I enzyme-linked immunosorbent assay (ELISA), which yield results similar to those of extraction assays (21). The treatment involved addition of 200 µL/well of the IGF-I acidification buffer (0.4 mol/L glycine-HCl, pH 2.0), followed by 30-min incubation as described above, and addition of 400 µL/well of the IGF-I neutralization buffer (0.85 mol/L Tris and 0.3% 8-anilino-1-naphthalenesulfonic acid). The acid-neutralized extract (20 µL) was analyzed for IGF-I by Diagnostic Systems Laboratories ACTIVE IGF-I ELISA.
The IGFBP-3 extraction protocol involved addition of 0.5 mL of the IGFBP-3 extraction buffer (0.05 mol/L sodium borate, pH 8.5; 9 g/L NaCl; 10 g/L BSA; and 0.1 g/L thimerosal) per single dried blood spot disc/well. The wells were shaken for 1 h as described above, and the extract (25 µL) was analyzed for IGFBP-3 using a Diagnostic Systems Laboratories ACTIVE IGFBP-3 ELISA.
IGF-I and IGFBP-3 ELISA
IGF-I and IGFBP-3 in serum, plasma, and whole blood were also analyzed by Diagnostic Systems Laboratories ACTIVE nonextraction IGF-I ELISA and IGFBP-3 ELISA. Both assays are noncompetitive (22, 23) and involve horseradish peroxidase-labeled detection antibody as previously described (24). The IGF-I ELISA incorporates a 101-fold sample pretreatment (acid neutralization) dilution factor and a total incubation time of less than 3 h, and demonstrates an overall precision of less than 10% (22). Similarly, the IGFBP-3 ELISA incorporates a 101-fold sample predilution, has a total incubation time of about 3 h, and has an overall precision of less than 10% (23). Absorbance of ELISAs were measured with the Labsystems Multiskan Multisoft microplate reader (Labsystems, Helsinki, Finland).
Calibration of dried blood spot IGF-I and IGFBP-3 assays
As IGF-I and IGFBP-3 ELISAs required about 100-fold sample predilution (22, 23), a similar dilution factor was incorporated in the dried blood spot extraction procedures to permit employment of the current kits’ calibrators. With this approach, the IGF-I or IGFBP-3 measured in plasma and in the dried blood spot extract would be expected to correlate with a slope representing their respective dilution factors. Fine adjustment of the extraction buffer volume or the standards could be then used to obtain closely comparable plasma and dried blood spot values.
Dried blood spot IGF-I and IGFBP-3 assay validation procedures
To evaluate IGF-I and IGFBP-3 extraction efficiency, levels in whole blood and corresponding dried blood spots were compared. Ten microliters of six different EDTA-blood samples were spotted onto the filter paper as described, and an extract of the entire 10 µL dried blood spot was analyzed for IGF-I and IGFBP-3 along with 10 µL of the corresponding whole blood sample. The IGF-I dried blood spot intraassay CVs, which include between-spot extraction variations, were evaluated by replicate analysis (n = 12) of extracts obtained from three to six separate dried blood spots; interassay CVs were determined by duplicate analysis of seven separate dried blood spot extracts assayed in seven separate runs. Similarly, the IGFBP-3 intraassay CVs were determined by replicate analysis (n = 24) of extracts obtained from 12 separate spots assayed in duplicate; the interassay CVs were determined by duplicate analysis of five separate blood spot extracts in five separate runs. Stability was assessed by storing replicate dried blood spots of three representative specimens at 37 C, room temperature, 4 C, and -20 C. The dried blood spot IGF-I and IGFBP-3 were then measured after 0, 8, 27, and 40 days of storage, and values were calculated as a percentage of the corresponding day 0 level.
IGF-I and IGFBP-3 in whole blood
Distribution of IGF-I and IGFBP-3 in whole blood. A freshly collected EDTA-whole blood sample (hematocrit, 0.33) was aliquoted into several samples of equal volume and centrifuged, except for one sample representing the originally drawn whole blood specimen. The relative plasma volume of the centrifuged samples was then changed to create a set of whole blood samples with hematocrit ranging from 0.20–0.70. After mixing, all samples, including the original 0.33 hematocrit sample as well as the plasma sample, were analyzed by IGF-I and IGFBP-3 ELISA. The measured IGF-I or IGFBP-3 in each blood sample was also compared to plasma IGF-I or IGFBP-3 concentration, and whole blood to plasma IGF-I and IGFBP-3 concentration ratios were established. These ratios were compared to the fractional plasma volume of the whole blood samples, calculated as 1 minus the hematocrit value. Directly comparable results indicate the analyte presence within the plasma portion of whole blood as recently reported for prostate-specific antigen (25).
Effect of hematocrit on whole blood and dried blood spot IGF-I and IGFBP-3. The above whole blood samples with hematocrits of 0 (plasma) to 0.70 were spotted onto the filter paper, dried, and extracted for IGF-I and IGFBP-3 analysis. Differences in the effects of hematocrit on the analyte concentrations measured in whole blood, or extracts of the dried blood spots were compared.
Data reduction
ELISA data were analyzed using data reduction packages included in the Labsystems instrument based on cubic spline (smoothed) curve fit. Other statistical analyses were performed using the Microsoft Excel 97 statistical software package (Microsoft Corp., Redmond, WA). Descriptive data are presented as the mean and SD unless otherwise specified. Linear regression analysis was performed by the least squares method, and correlation coefficients were determined by the Pearson method.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The effects of various parameters on extraction efficiency of dried blood spot IGF-I and IGFBP-3 were investigated. Both IGF-I and IGFBP-3 were completely released within 60 min of extraction, and a variety of extraction reagents, including deionized H2O, appeared equally effective (data not shown).
The efficiency of the optimized extraction methods was evaluated by
comparing whole blood concentrations of IGF-I and IGFBP-3 with those
measured in extracts of the corresponding dried blood spots as
described in Materials and Methods. As shown in Table 1
, the recovery of dried blood spot IGF-I
and IGFBP-3 ranged from 87–107% and 91–109%, respectively, with
mean values of 97% and 101%.
|
).
|
IGF-I and IGFBP-3 in whole blood
As circulating molecules may be differentially distributed among
the various whole blood subfractions (plasma and blood cell
subfractions), IGF-I and IGFBP-3 levels in plasma and in a series of
corresponding whole blood samples with hematocrit ranging from 0
(plasma) to 0.70 were measured. As shown in Table 3, the whole blood IGF-I and IGFBP-3
levels varied inversely as a function of variations in hematocrit,
suggesting distribution of IGF-I and IGFBP-3 in the plasma fraction.
This was further supported by the comparison of whole blood/plasma
IGF-I and IGFBP-3 concentration ratios with the corresponding
fractional plasma volume of each whole blood sample (calculated as
1 - hematocrit), which showed the expected (25) closely
comparable relationship (Table 4).
|
|
). In the
hematocrit range of 0.2–0.7, the whole blood and dried blood spot
IGF-I levels (mean ± SD) were 198 ± 73 and
196 ± 33 µg/L, and IGFBP-3 levels were 2.3 ± 1.0 and
4.1 ± 0.8 mg/L, respectively. When the whole blood and dried
blood spot values were calculated as the percent change from the
corresponding concentration measured in the original 0.33 hematocrit
sample, the whole blood IGF-I and IGFBP-3 appeared to change by about
8–10% for every 0.05 U change in hematocrit. In contrast, the dried
blood spot IGF-I and IGFBP-3 levels changed by about
3–4%/0.05 U change in hematocrit. Only at abnormally high hematocrit
(>0.6) was the change in dried blood extract levels relative to that
measured in the corresponding 0.33 hematocrit specimen more than 10%
(data not shown). Comparative analysis
IGF-I and IGFBP-3 levels in plasma, whole blood, and dried blood
spots of freshly drawn random samples (n = 45) were compared.
Regression analysis of data showed an acceptable linear relationship
between plasma levels and those of whole blood and dried blood spots
(Figs. 1 and 2). The obvious deviation of slopes from
unity is primarily due to the differences in the volume of the plasma
fraction of whole blood and dried blood spots relative to the volume of
plasma used.
|
|
Paired venous blood EDTA-plasma and capillary dried blood spot
IGF-I and IGFBP-3 from subjects with untreated GHRD (n = 11) and
age-matched normal subjects (n = 16) were simultaneously assayed.
Similar to the above, there were excellent correlations between plasma
and corresponding capillary dried blood spot levels, with slope values
of near unity (Fig. 3). The capillary
dried blood spot assays were capable of identical sensitivity as the
conventional plasma sampling methods in distinguishing between GHRD and
normal IGF-I and IGFBP-3 levels. In these samples, the mean
(±SD) plasma and capillary dried blood spot IGF-I levels
were 98.3 ± 100 and 92.9 ± 90 µg/L, respectively. The
mean values for IGFBP-3 were 1.95 ± 1.35 and 2.3 ± 1.4
mg/L.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although immunoassays capable of reliable determinations of GH-IGF axis components are now commercially available (22, 23, 26), the disadvantages of blood sampling by venipuncture and the practicality and safety concerns of liquid sample transportation have not been adequately addressed. The ability to measure components of the IGF peptide family in a few microliters of dried blood may circumvent these limitations and allow a broader application of their measurements. Blood sampling on filter paper is well established (19) and has found extensive applications in large scale infant and population screening programs to detect inborn error of metabolism and congenital defects (14, 15), infectious disease (16, 17), and nutritional deficiency (18). We here demonstrated that the same approach is ideal for blood IGF-I and IGFBP-3 analysis, as these analytes are highly stable in dried blood spot and are readily transferable into a liquid phase without any apparent loss of immunoreactivity.
Variations in hematocrit could invalidate whole blood analysis of analytes that are primarily distributed in the plasma fraction unless corrective measures are implemented. The effect of changes in hematocrit may not, however, be a major concern when blood dried on filter paper is used as a specimen. As it has been recently reported (25), application of blood to filter paper appears to involve a built-in corrective mechanism that would significantly minimize the effect of hematocrit. We similarly observed that blood samples with a high hematocrit impregnated a proportionally smaller area than the same volume of blood that had a lower hematocrit. The effect of high hematocrit on reducing the blood fractional plasma volume applied, therefore, was opposed by its effect on increasing the amount of blood dried per unit area on the filter paper. Because both IGF-I and IGFBP-3 are present in plasma fraction, and a constant area of dried blood spot is used for their analysis, dried blood IGF-I and IGFBP-3 levels were significantly less affected by changes in hematocrit than their measurements in whole blood. The variations in IGF-I and IGFBP-3 levels (mean ± percent CV) measured in dried blood extracts of samples with an adjusted hematocrit range of 0.2–0.62 were within acceptable limits (204 µg/L ± 14% and 4.4 mg/L ± 11%, respectively).
Calibration of dried blood spot assays with liquid calibrators was initially instituted because of the unavailability of human blood lacking endogenous IGFs and IGFBPs. However, as changes in hematocrit appeared to have minimal effect on dried blood spot determinations, the use of liquid calibrators was found to be an ideal alternative. By implementing a similar dilution factor in the extraction protocols used for serum analysis, the dried blood spot IGF-I and IGFBP-3 levels were within 5–15% of those measured in the corresponding plasma samples. The use of liquid calibrators had additional advantages of simplified formulation and significantly improved consistency. The potential variations in extraction efficiency could be monitored by inclusion of a set of dried blood quality control spots in each assay-run. A similar approach to calibration of the dried blood spot prostate-specific antigen assay has been recently described (25).
The dried blood spot methods showed acceptable analytical performance characteristics and demonstrated excellent agreements compared with the conventional plasma IGF-I and IGFBP-3 assays. Despite the expected variations in whole blood IGF-I and IGFBP-3 measurements due to the hematocrit effect, their comparisons with plasma levels also showed relatively minimal scattering along the regression line. This is perhaps due to the narrow range of hematocrit that is normally encountered even in a randomly selected patient population. The close agreement between dried blood spot and conventional serum assays strongly supports the validity of dried blood spot IGF-I and IGFBP-3 analysis. In this context, analysis of IGF-I and IGFBP-3 in paired capillary dried blood spots and venous blood plasma samples demonstrated similar sensitivities in differentiating between Ecuadorian GHRD subjects, who reportedly express very low IGF-I and IGFBP-3 levels (20), and their age-matched normal controls. The obvious advantages of filter paper capillary blood sampling combined with the acceptable analytical performance characteristics of the dried blood spot assays makes the approach ideal for pediatric applications. In addition, the remarkable stability of IGF-I and IGFBP-3 in dried blood spots is highly advantageous when the stability, cost, and safety of liquid sample collection and transportation to distant laboratories may be limiting factors.
In summary, we described measurement of IGF-related components from dried blood filter paper spots. The IGF/IGFBP-3 dried blood spot assays are capable of comparable sensitivity as plasma measurements and are highly insensitive to changes in hematocrit that might be problematic for whole blood analysis. Because they allow for a small sample volume drawn by capillary puncture and they have a high stability after drying, the IGF/IGFBP-3 filter paper assays are of particular value in pediatric population and/or applications requiring reliable and cost-effective sample collection and transportation.
Received November 6, 1997.
Revised March 10, 1998.
Accepted March 27, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. Khosravi, A. Diamandi, J. Mistry, and R. G. Krishna The High Molecular Weight Insulin-Like Growth Factor-Binding Protein Complex: Epitope Mapping, Immunoassay, and Preliminary Clinical Evaluation J. Clin. Endocrinol. Metab., August 1, 1999; 84(8): 2826 - 2833. [Abstract] [Full Text]
|
|
|
|
|
|
M. L. Mitchell and Hermos; Comment on Dried Blood Spot Assay of IGF-I and IGFBP-3 J. Clin. Endocrinol. Metab., February 1, 1999; 84(2): 822 - 822. [Full Text]
|
|
|
|
|
|
J. Khosravi and A. Diamandi Dried Blood Spot Assay of Insulin-Like Growth Factor (IGF)-I and IGF Binding Protein-3 (IGFBP-3)--Authors' Response J. Clin. Endocrinol. Metab., February 1, 1999; 84(2): 822a - 823. [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
; n = 11) and age-matched normal controls (•; n = 16).
Values represent the mean of duplicate measurements.