Jun 30, 1983 - presence ofhuman luteinizing hormone. Am J Obstet Gynecol 113, ... of the glycoprotein tropic hormones.-choriogonadotropin, lutropin,.
measures human chorionic gonadotropin in the presence of human luteinizing hormone. Am J Obstet Gynecol 113, which specifically 751-758
(1972).
5. Fowler JE, Taylor G, Blom J, Stutzman RE. Experience with serum alpha-fetoprotein and human chorionic gonadotropin in nonseminomatous testicular tumors. J Umi 124, 365-368 (1980). 6. Donaldson ES, VanNagell JR. Pursell S, et al. Multiple biochemical markers in patients with gynecologic malignancies. Cancer 45, 948-953 (1980). 7. Broder LE, Weintraub BD, Rosen SW, et al. Placental proteins and their subunits as tumor markers in prostatic carcinoma. Cancer 40, 211-216
(1977).
8. Fowler JE, Platoff GE, Kubrock CA, Stutaman RE. Commercial radioimmunoassay for beta subunit of human chorionic gonadotropin: Falsely positive determinations due to elevated serum luteinizing hormone. Cancer 49, 136-139 (1982).
CLIN. CHEM.
29/11,
1966-1968
9. Hamada K, Tomoda S, Sugawa T, Takahashi KP. Simultaneous competitive enzyme immunoassay for human chorionic gonadotropin. Endocrinol Jpn 25, 515-517 (1978). 10. Iwasa S. Kitada C, Yoshida I, et al. Highly specific enzyme immunoassay for human chorionic gonadotropin. J Biochem 89, 1091-1099 (1981). Ii. VanWeeman BK, Schuurs AHWM. Immunoassay using antibody-enzyme conjugates. FEBS Leit 43, 215-218 (1974). 12. Wads HG, Danisch RJ, Baxter SR. et al. Enzyme immunoassay
of the glycoprotein tropic hormones.-choriogonadotropin, lutropin, thyrotropin-with solid phase monoclonal antibody for the asubunit and enzyme-coupled monoclonal antibody specific for the /3subunit. Clin Chem 28, 1862-1866 (1982). 13. Vaitukaitis JL. Immunologic and physical characterization of human chorionic gonadotropin (hCG) secreted by tumors. J Clin Endocrinol Metab 37, 505-514 (1973).
(1983)
Quality-Control Sera for Routine Determination of Aluminum by Electrothermal Atomic Absorption Spectroscopy Fred
V. Leung and A. Raiph Henderson
Several commercially available quality-controlsera were
analyzed for aluminum content by atomic absorption spectroscopy with a stabilized-temperature graphite furnace. The values obtained ranged between 4 and 1250 zg/L (0.148 to 46.235 mol/L). No significant difference was detected for
between-vial variation for four lots of quality-control sera (p> 0.05). Control sera stored in 1-mL polypropylene vials and frozen at -20 #{176}C for up to six months showed no significant variation in aluminum content (p>
0.05), but those stored in
their original glass containers had significantly increased aluminum content (p < 0.001) over a four-week period.
plied quality-control sera and determined their aluminum content. The suitability of such material would circumvent the need to prepare “in-house” sera.
Materials and Methods We determined the concentration of aluminum in serum with use of a stabilized temperature platform furnace as previously described (7). Precautions such as pre-rinsing all containers in a saturated solution (about 0.5 mol/L) of Na2EDTA and rinsing several times with purified water (Milli-RO
and Milli-Q
purification
systems;
Millipore
Ltd.,
The accumulation of aluminum in patients with chronic renal failure reportedly contributes to dialysis encephalopathy and to osteomalacia that is resistant to vitamin D therapy (1-3). The therapeutic removal of aluminum with desferrioxammne and dialysis improves these patients’ wellbeing significantly (4, 5). The method of choice for monitoring concentrations of aluminum in serum for the diagnosis and treatment of these patients is atomic absorption spectrometry with use of a graphite furnace (6, 7). The lack of reference material and an assayed qualitycontrol sera to ensure the reliable determination of serum aluminum by atomic absorption spectrometry has prompted us to develop and to search for suitable sera to ifil this need. We selected without conscious bias 13 commercially sup-
Mississauga, Ontario MV 1L2 Canada) were followed to avoid container contamination of the samples. The commercial quality-control sera we studied are listed in Table 1. Each product was prepared in the original vial as described by the supplier except that purified water was used to reconstitute the lyophilized material. We tested as many as 10 replicates of each serum to establish its mean value for aluminum. A portion of each serum was aliquoted into 1-mL fractions and stored frozen at -20 #{176}C. The protocol for the preparation of “in-house” qualitycontrol sera is similar to that reported earlier (7). Sera obtained from patients admitted to this hospital and determined to have a low aluminum content ( 0.05) for the aluminum values over a six-month period. At 4#{176}C, the “in-house” sera show a marginal increase in concentration for aluminum after five days. Liquid quality-control sera prepared from fresh-frozen human plasma stabilized with ethylene glycol, however, gave reproducible results for aluminum on storage at 4#{176}C over a three-week period. These sera were stored in their original plastic containers and remain liquid at -20 #{176}C.
Discussion
To monitor the precision of routine determinations of in serum by atomic absorption spectroscopy, one should use control materials as similar as possible to human serum. Owing to the unavailability of analyzed commercial Standard statistical methods were used for the single reference sera, we initially prepared our own pooled qualityclassification analysis of variance (ovA) (8). control sera. As shown in Table 2, the three levels of our inhouse pool gave satisfactory results for within-run and Results between-run variation. To meet the needs of a busy routine Assayed values. The concentration of aluminum deterclinical laboratory, we investigated the use of commercial mined in the quality-control sera spanned a wide range from quality-control material. Representative commercial prodnormal to >1000 ,ugfL (>37.0 Lmol/L). Table 2 lists the ucts analyzed for aluminum content gave reproducible remean values and CV for within-run and between-run assays sults for each lot of material analyzed. Because about 80% of for each quality-control serum. The CV for within-run assay aluminum in serum is bound to protein constituents (9), controls are less than 9%, with the lowest CV (0.7%) being particularly to albumin (10), this trace element may possiobtained for controls containing the highest aluminum bly be introduced into the quality-control material during concentrations. Between-run CVs were as much as 11.7% at processing. Most of the quality-control material contained a the lower concentrations of aluminum. human serum base, but bovine-based product showed no Vial variations. Between-vial variation of the lyophilized substantial differences, i.e., did not have lower aluminum control sera was tested on lots containing four reprecontent or greater stability on storage. sentative concentrations of aluminum: 18, 77, 124, and 276 The lyophilized quality-control products were all in pow,ugIL (0.667, 2.854, 4.595, and 10.228 moI/L). Six vials of der form except for two lots of a product manufactured by a each lot of sera were reconstituted and the aluminum new beaded-form technique. These later products contained concentrations determined. Analysis of variance gave Fthe highest concentrations of aluminum. Purified water ratios of 1.1429 to 1.6302, which indicated no significant (aluminum-free) should be used to reconstitute the lyophidifference-between vials (p > 0.05). lized sera. The diluent supplied by the manufacturer with Storage effects. The effects of storage at 4#{176}C and -20 #{176}C on the quality-control sera should be analyzed for aluminum the concentration of aluminum are illustrated for two reprecontent before use. The aluminum content of the diluents sentative control sera (Figure 1). Sera stored at 4#{176}C in their supplied with four of the products we tested ranged from original glass vials after reconstitution show a significant undetectable to >7.0 mol/L. The main sources of variation in lyophilized quality300 control products include the reproducibility of the diluent a volume, vial-to-vial constituent (filling) variation, within.A day variation of the measurement system, and between-day A run variations. As evaluated by the ANOVA process, the A aggregate of these individual component variances showed 1200 that the greatest contribution to the variation was the vial_0 ....___a...___a_____&_a_____6.*__& to-vial component (11). We assessed this factor and showed z that vial-to-vial variation does contribute to the error but 2 B was within acceptable limits (p> 0.05) for the overall assay -J variance. I00 The contamination of samples with aluminum has been 0 described as being related to the type of container used for its collection and storage (12). We tested the effects of storage on the lyophilized quality-control products at 4#{176}C in I I I I I I I I their original glass vials and found a gradual increase in 0 16 2 4 8 20 24 28 32 aluminum content after 48 h, which becomes markedly TIME (dl greater in four weeks. Fig. 1. Effects of storage at 4#{176}C (dashed lines) and at -20#{176}C (so/id This contamination is probably due to leaching of the lines) on aluminum concentration of two representative quality-control element from the surface of the glass. When we used sera (A WPOAS Iand B Sera Chem N) aluminum
-
=
=
CLINICAL CHEMISTRY, Vol.29,No. 11,1983
1967
Table 2. AlumInum Content of Control Sera WIthIn-run
Between-run CV, %
n
x ± soa
0.3
7.5
4
4 ±
0.5
0.4
6.7
0.8
5.3
5 5
6 ± 15 ±
0.7 1.5
6 1.5
1.4 1.2
9 8
1.9
3.2
3.5
3.7
6 7
59 ± 5.5 98 ± 10
10.2
8 8
69 ± 16±
7.4 11.2
Serum and vial
n
I ± so
Decision 1 2
8 8
4 ± 6 ±
8
15 ±
5 8
381 ± 121 ±
5 6
59 ± 94 ±
9 9
72 ± 17±
3 1.5
4.2
3
CV, % 11 11.7 10
Monitrol I Ii Versatol Lo Hi Sera-Chem N A
8.8
383 ± 12 119 ± 8
5 1.8
3.1 6.7 9.3
Omega
I
8
Il
8
1248 ± 9 1179±21
0.7
7
1251 ± 44
3.5
1.8
6
1149±35
3.0
2.3 1.9
7 8
WPQAS
123 ± 271 ±
7 8
I II In-house pool 1
2.8 5.2
25 ± 1.2 4.8 54 ± 2.0 3.7 3 10 126 ± 4.2 3.3 a Eachsample measured in triplicate; units are in zg/L (to convert to nmol/L, multiply by 37.06).
2
10 10
polypropylene containers, previously EDTA-washed and rinsed with purified water, there was no noticeable change in aluminum content (7). We routinely store our “in-house”
quality-control sera in snap-capped 1.5-mL polypropylene washed as described. Similar treatment of the commercial quality-control sera in 1-mL aliquots resulted in stable results for aluminum on storage at -20 #{176}C (Figure 1). The ethylene glycol-based materials we tested are suitable for use in our analytical procedure for aluminum determination, their aluminum content being closer than the other materials to the normal reference range, which may be useful for measuring low concentrations of aluminum. The presence of about 30% ethylene glycol did not appear to bias the analytical detection of aluminum, but the accordingly decreased proportion of human-based matrix in the sample may make it less representative of a serum specimen. The reproducibility of results with this set of materials is acceptable for routine monitoring at the lower aluminum value (CV
