InsulinImmunoreactivityStabIlIzed In Serum and ... - Clinical Chemistry

1 downloads 0 Views 468KB Size Report
29, No. 8, 1983. 1559. Wesee this effect most commonly in specimens from intensive-care ... were added to 5 mL of blood, the result .... Conclusionsfrom a Pilot.
We see this effect most commonly in specimens from intensive-care units, where syringes, needles, and cannulas often are heparinized before specimens are collected for blood-gas analysis. The minimum volume of anticoagulant that would be left in a heparimzed syringe is the dead space of the nozzle, about 100 tL. If this volume of heparin solution, containing 3 g of chlorocresol, were added to 5 mL of blood, the result for glucose would be about 1 mmollL too high, a difference of clinical importance

that probably

is

only in hypoglycemia or in serial sampling. Any more of the preservative, however, would produce a proportionately greater effect. This is a recognized chemical interference (the instrument handbook states that “the YSI model 23 AM is not suitable for use with samples conmining thymol or other phenol preservatives; these are interfering substances for the model 23 AM and are contraindicated”), but is easily overlookedbecause of an unsuspected technique of specimen collection. We advise users of this instrument to check the techniques and anticoagulants used by requesting clinicians. D. Peel Dept. of din.

Biochem. Queen Elizabeth Hosp. Gateshead, NE9 6SX

UI’. J. D. S. Kay F. Taylor Dept. of Clin. Biochem. Hosp. for Sick Children Great Ormond St. London,

UJC.

Immunoreactive insulIn concn (±SD) remaining after five days at 30 or 37#{176}C (% of untreated -20#{176}C value) 30#{176}C

Treatment Serum

None NaF

NPM

± 6** (n 4) 39 ± 17#{176}’ (n = 9) 97±11(n=9)

37

Temperatures

To the Editor: Finding that insulin immunoreactivity is unstable in plasma at 20 and 37#{176}C (1), we then attempted to find substances capable of slowing the rate of loss sufficiently to allow postal transport of specimens for insulin assay. In a report showing that insulin is similarly unstable in serum (2), Oliver presented evidence that sodium fluoride, 5 g/L, is an effective preservative. We could not confirm this, but have found buffered N-phenylmaleimide (NPM) to have a useful stabilizing effect. NPM tubes contained 40 L of a 50 mmolJL solution of NPM in ethanol and 40 pL of aqueous buffer [0.8 mol/L

NH4H2PO4,

0.2 molJL (NH4)2HP04]

24

=

±

7** (n =

4)

30 ± 12#{176}#{176} (n = 9) 94±16(n=9)

Plasma

None NaF NPM

77 78

±

5#{176} (n = 4)

±

15#{176} (n = 10)

54 35

108±33(n=10)

± ±

6” (n = 4) 9#{176} (n = 10)

101±28(n=10)

p < 0.01 that mean = 100 (t-test) p < 0.001 thatmean = 100 (t-test)

that had been air-dried together. NaF tubes contained 5 mg of NaF powder. Serum and plasma (--1 mg of Na2EDTA per milliliter) were collected from normal subjects about an hour after lunch (range: 70-330 pmol of insulin per liter) and 1-mL aliquots were added to capped polypropylene NaF, NPM, and control (no additives) tubes. After mixing and dissolving the additives we incubated the tubes at 30 or 37#{176}C. Comparison aliquots were stored at -20 #{176}C. After five days all aliquots were frozen and stored at -20 #{176}C until the immunoreactive insulin concentration was determined by the method of Albano et al. (3), with use of an antiserum raised in this laboratory. For analysis of the results, insulin concentrations for each aliquot were expressed as a percentage of the concentration in the corresponding aliquot stored at -20 #{176}C. The results tabulated above confirm that significant insulin immunoreactivity is lost from serum and plasma stored five days at

In conclusion, we could not confirm the effectiveness of fluoride as an inhibitor of insulin degradation in serum but we find that, as expected from the properties of known insulin-degrading enzymes, a maleimide is an effective inhibitor of immunoreactive insulin loss in serum and plasma.

References 1. Livesey JH, Hodgkinson SC, Roud HK, Donald BA. Effect of time, temperature and

freezing on the stability of immunoreactive LH, FSH, TSH, growth hormone, prolactin and insulin in plasma. Clin Biochem 13, 151-155 (1980). 2. Oliver LK. Sodium fluoride stabilizes insulin at room temperature. Clin Chem 23,

303(1977). 3. Albano, JD, Ekins RP, Maritz, G, Turner

RD. A sensitive, precise radioimmunoassay

of serum insulin relying on charcoal separation of bound and free moieties. Acts Endocrinol 70, 487-509 (1972) 4. Yokono K, Yoshimichi I, Shii K, et al. Purification and characterization of insulin degrading enzyme from pig skeletal muscle. Endocrinology 108, 1527-1532 (1981). 30#{176}C. Losses are even greater at 37#{176}C.5 Varandani Fr. Insulin degradation. XII. The aminoacid composition, amino terminal However, addition of sodium fluoride and carbohydrate content of beef pancreatic to give a final concentration of 5 g/L glutathione-insulin transhydrogenase. Biohad no useful preservative effect at chim Biophys Acts 371, 577-581 (1974). either temperature of storage, the

losses at 30#{176}C being virtually

InsulinImmunoreactivityStabIlIzed In Serum and Plasmaat Ambient

37#{176}C

identical

to control samples, and at 37 #{176}C over 60% of insulin immunoreactivity was lost with NaF present in both serum and plasma. In contrast, buffered Nphenylmaleimide, 2 mmol/L, largely preserved insulin immunoreactivity in

both serum and plasma at both 30 and 37#{176}C. This effectiveness of NPM was not unexpected, because it is a good inhibitor of insulin-specific protease (4) and glutathione insulin transhydrogenase (5), either or both of which might be responsible for the insulindegrading activity in plasma and sorum. Buffered NPM does not interfere in the insulin assay and is more effecfive than unbuffered NPM for preserving insulin immunoreactivity in plasma (Livesey, unpublished results). It may be that our results with fluoride differ from those of Oliver (2) because of the possibility of differing specificity of our antiserum.

John Livesey Richard A. Donald Helen K. Roud Dept. of Endocrinol. Princess Margaret Cashmere Rd.

Christchurch,

Hosp.

2,

New Zealand

Conclusionsfrom a Pilot ImmunoreactlveTrypainNewborn Screen for CystIcFibrosis To the Editor: We have now completed a two-year pilot screening program in which immunoreactive trypsin (IRT) was measured in blood eluted from ifiter papers on which it was collected as part of a

CLINICAL CHEMISTRY, Vol. 29, No. 8, 1983

1559

universal phenylketonuria/hypothyroidism screen. The project was based on an hypothesis first outlined by Crossley et al. (1), viz., that an abnormally high blood IRT value for a newborn is suggestive of cystic fibrosis (CF). Details of the methods were previously described (2). We tested 25 000 newborns and identified seven infants confirmed to have CF. One of the seven was not detected in the initial assay, but a repeat card, taken one week later because the infant was premature (birth weight, 1086 g), revealed an increased value for IRT. Values for all affected infants were within the top 1% of each assay run of 200 specimens. During the project period, eight infants were diagnosed with CF. The one not detected by the IRT program did not have a filter paper card taken at birth. From our experience, we conclude that the IRT test is a reliable indicator of infants with CF if the reagents used are of good quality and the specimens are fresh. We would use only reagents comparable to those outlined by Crossley et al. (3) and would assay the specimens within a few days of their arrival at the screening laboratory.

the immune function of the body because of its association with histocompatibility antigens (2) and also because its amino acid sequence closely resembles that of the constant part of the immunoglobulin chain (3). This peptide not only is present on the surface membrane of both B and T cells (4) but also in the membranes of all nucleated cells (5). Data on its concentration have been used as a reliable indicator of glomerular and tubular functions (6). Abnormally high concentrations of it have been reported in patients with impaired renal functions (6), rheumatoid arthritis and Sjogren’s syndrome (7), and malignant tumors (8), and more striking increases were shown in lymphoproliferative disorders (9). More recently, increased concentrations of 32-microglobulin were shown in an immunodeficient homosexual man (10). In the present preliminary study, we have determined /32-microglobulin in the sera of patients with ms, as tabulated below. Note that 29 of 31 patients with AIDS had supranormal /32-microglobulin concentrations, whereas five of 11 normal homosexual men had increased values as compared with normal

controls.

References

PrMlcroglobuiin, mg/L

‘1.Croesley JR. Elliott RB, Smith PA. Dried blood-spotscreening for cystic fibrosis in the newborn. Lancet 1, 472-474 (1979). 2 Kirby LT, Applegarth DA, Davidson, AGF, et al. Use of a dried blood spot immunoreactive-trypsin assay for detection of cystic fibrosis in infants. Clin Chem 27, 678-680 (1981). 3. Crossley JR. Smith PA, Edgar BW, et al. Neonatal screening for cystic fibrosis, using immunoreactive trypsin assay in dried blood spots. Clin C/urn Acts 113, 111-121

(1981). L T. Kirby A. G. F Davidson D A. Applegarth L T. K. Wong D. F. Hardwick Depts. of Pathol. and Pediatrics Univ. of British Columbia and Children’s Hosp. Vancouver, B.C. V6T 1W5, Canada

No. 15

31 11

Range Mean Normal controls 1.6-2.4 1.9 AIDS patients

2.0-16.0

To the Editor: $2-Microglobulin, a small peptide 11 600), is normally present in trace amounts in both serum and urine (1). It may play an important role in

(M

1560

anti-13-microglobulin sera. Immunol Cornman 2, 403-414 (1973). 5. Nilssen K, Evrin PE, BerggArd I, Panten J. Involvement of lymphoid and non-lym-

phoid cells in the production of 13.2.microglobulin-a homologue of the constant domain of IgG. Nature New Biol 244, 44-45

(1973). 6. Ervin PE, Wibell L. Serum 132-microglobWin in various disorders. Clin Chim Acts 43, 183-187 (1973). 7. Talal N, Grey HM, Zvaifier N, et al. Elevated salivary and synovial fluid 132microglobulin in Sjogren syndrome and rheumatoid arthritis. Science 188, 11961198 (1975). 8. Kithier K, Cejka J, Belarnaric J, et al. 13-Microglobulin: Occurrence in fetal life and malignancy. Clin Chim Acts 52, 293299 (1974). 9. Poulik MD, Farrah D, Smithies 0. Association of urinary 32-nucroglobulin with myeloproliferativediseases.Fed Proc Fed Am Soc Exp Biol 31, 741 (1972). 10. Francioli P, Clement F. Beta2-microglobulin and immunodeficiency in a homosexual man. N Engi J Med 307,1402-1403

(1982). 11. Gottlied MS, SchroffR, Schanker H, et al. Pneumocystis carinii pneumonia and mucosal candidiasis in previously healthy homosexual men. N Engi J Med 305, 14251431 (1981). 12. Beta2-microglobulin in hematological malignancies. Br Med J 282, 1013 (1981). B. Bhalla Bijan Safai Roland Mertelsmann Morton K. Schwartz Ravi

0.24

3.50

Normal homosexual men 1.7-4.5 3.0

a Different from normal homosexual men at p