23
Clinica Chimica Acta, 99 (1979) 23-29 @ Elsevier/North-Holland Biomedical Press
CCA 1151
TRACE METAL DETERMINATIONS IN BLOOD OBTAINED EVACUATED COLLECTION TUBES
DARRYL
IN
M. WILLIAMS
Department of Medicine, University of Utah College of Medicine, Salt Lake City, Utah 84132 and the Department of Medicine, Louisiana State University School of Medicine in Shreveport, Shreveport, LA 71130 (U.S.A.) (Received June 7th, 1979)
Summary Using atomic absorption spectrophotometry, we determined levels of iron, copper and zinc in serum or plasma samples collected in acid-washed containers and in commercially available evacuated tubes including one model equipped with rubber stoppers prepared by a process which reduces zinc contamination of the rubber. Serum or plasma from all evacuated tubes yielded iron and copper values which were comparable to those from serum obtained in acidwashed glass. Zinc values were higher from all evacuated tubes than from acidwashed glassware. Values obtained from samples collected in the speciallyprepared tubes were 10% higher than values obtained from samples collected in acid-washed glassware. Thus, the specially-prepared evacuated tubes may only be useful in large-scale clinical screening studies. Traditional blood collection methods for zinc determination should be used when accuracy is required.
Introduction The measurement of serum levels of the trace metals, iron, copper, and zinc has assumed increasing clinical importance and there is a need for reliable screening methods for large-scale clinical surveys. Analytical methods have been developed which improve the accuracy of these metal determinations, but there are numerous sources of contamination which lead to erroneously high values [l]. A common cause of contamination is the use of inadequate blood collection techniques. In particular, samples collected in rubber-stoppered evacuated tubes or with plastic syringes containing rubber plungers may give falsely elevated zinc values [ 21. Butyl rubber, used in such products, is manufactured by Correspondence should be addressed to: Dr. Williams, Department of Medicine, Louisiana sity Medical Center at Shreveport, P.O. Box 33932, Shreveport.. LA 71130 (U.S.A.).
State
Univer-
24
a process which uses zinc salts,including zinc oxide, zinc diethyldithiocarbamate and zinc dimethyldithiocarbamate [3] which result in zinc contamination of the stopper. In order to overcome this drawback, an evacuated bloodcollection tube has been developed which is sealed by a rubber stopper made by a process which does not utilize these salts. The purpose of this study is to compare the results of iron, copper, and zinc determinations in samples collected in these tubes and in other tubes commonly used for the collection of blood. Materials and methods Blood collection Blood samples were obtained from normal volunteers in the fasting state after receiving written, informed consent and with the approval of the Human Use Committee. A single specimen was drawn in an acid-washed, glass syringe mounted with a 20 gauge stainless steel needle (Monoject, Rutherford Medical Industries, Deland, FL) and transferred to an acid-washed glass test tube. Additional samples were drawn in evacuated tubes by an immediate second venipuncture. These tubes were obtained from a commercial source (Vacutainer, Becton Dickson, Inc., Rutherford, NJ) and included plain (red stoppered), leadfree (amber stoppered), heparinized (green stoppered), and special “metalfree” (blue stoppered) tubes. The order of blood drawings in evacuated tubes was varied sequentially from patient to patient. In one half of the subjects, blood was drawn in evacuated tubes in the first venipuncture before a second venipuncture with the glass syringe. Blood obtained with glass syringe or in unheparinized evacuated tubes was allowed to clot at room temperature. Blood samples obtained from cancer patients were drawn at regularly scheduled clinic visits. These samples were obtained with the patients’ signed consent and with the approval of the Human Use Commit~e. Serum or plasma was separated by centrifugation and transferred to acid-washed glass test tubes. If visible hemolysis was present in any specimen, all specimens obtained from that subject were excluded from analysis. In some studies, 2.0 ml of distilled, deionized water was added to collection tubes in place of blood. In additional tubes, 2.0 ml of solution containing 120 pg/dl of each of Fe, Cu, and Zn in 0.1 mol/l HCl were added to representative tubes. These tubes were placed on a rotary mixer for 5 min and then processed in an identical fashion to serum samples. Analytical methods Iron, copper, and zinc determinations were carried out in 2.0 ml serum or plasma samples deproteinized by the addition of trichloroacetic acid at a final concentration of 5.0%. Samples were heated under reflux in a boiling water bath for 5 min and centrifuged [43. The resulting clear supernatant was analyzed for iron, copper, or zinc using a Fer~n-Elmer model 305 atomic absorption spectrophotometer fitted with a 3-slot burner head with wavelength settings of 248.3, 324.8 and 213.9 nm respectively. The instrument was calibrated with standards prepared from metallic salt solutions (Fisher, Fair Lawn, NJ) containing 5% trichloroacetic acid in 0.15 mol/l NaCl. Recorder
25
response was linear over the range measured. Distilled, deionized water and standard metal solutions added to collection tubes were subjected to the same deproteinizing procedure before analysis. Results Serum trace metal determinations The mean values for serum iron, copper, and zinc in 40 normal volunteers are shown in Table I. Values obtained from serum collected in plain, acidwashed glass tubes are in close agreement with values obtained in other series. The values appear to have a normal distribution, and the 95% confidence limits, without regard to sex, are: iron, 67-175 pg/dl; copper, 94-166 pg/dl; and zinc, 77-133 pgldl. Values obtained from samples collected in special rubberstoppered tubes are: iron, 66-174 pg/dl; copper 91-169 pg/dl; and zinc, 88144 pg/dl. Mean values for both iron and copper are comparable regardless of the tubes used for collection. In contrast, mean zinc values are consistently higher when plain evacuated tubes, lead-free evacuated tubes, or heparinized evacuated tubes are used for collection. These values are significantly different (P < 0.001) and there is considerable variability of values obtained from samples collected in the various evacuated tubes. There is also a statistically significant difference between values obtained from samples collected in plain, acid-washed tubes and in special rubber-stoppered evacuated tubes (P < 0.01). Values from the latter are approximately 10% higher than values obtained from samples in plain acidwashed tubes with a range of lOO-124% of these values. The relationship of samples collected by the two methods from 40 normal subjects and 10 cancer patients is shown in Fig. 1. Serum zinc levels are consistently higher in blood collected in the special rubber-stoppered tubes, but there is a good correlation between the two methods (slope = 0.998, r = 0.973). Eight zinc determinations from samples contained in acid-washed tubes are identified as less than the normal range. All matched samples contained in special rubber-stoppered tubes also fall below the normal range determined for samples thus collected.
TABLE I TRACE METAL DETERMINATIONS IN SERUM OR PLASMA UNTEERS IN PREPARED TEST TUBES Tube
FROM NORMAL
Metal concentration (I.rg/dI)
plain, acid washed Plain, evacuated Lead-free. evacuated Heparinized
COLLECTED
, evacuated
Special rubber. evacuated * Refers to mean + S.E.
Iron (n = 40)
(n = 35)
Zinc (n = 40)
121 f 4.3 * 122 f 4.2 120 f 4.5
130 + 3.1 1312 3.4 128 f 2.8
105 f 2.3 194f 6.4 250 f 12.7
copper
121 2 4.4
128 f 2.7
184f
7.2
120 f 4.2
130 ? 3.3
1161
2.3
VOL-
26
IO
20
?bo 40
50
60
SERUM ZINC
7J
80
so
IJO
CONCENTRATION
ACID-WASHED
110 I20
130
140
I30
(j~g/dl)
GLASS TUBES
Fig. 1. Correlation of zinc determinations carried out in human serum obtained from blood drawn in special rubberstoppered evacuated tubes (ordinate) and in glass syringes and acid-washed glassware (abscissa). The dashed line represents hypothetically perfect correlation and the solid line represents the calculated correlation (slope = 0.998. r = 0.973). Values are consistently higher in samples obtained in
evacuated
tubes.
Trace metal determinations in distilled deionized water When distilled, deionized water is agitated within the various collection tubes, it is not remarkably contaminated by iron or copper (Table II) and iron and copper determinations are only minimally elevated above those obtained in water used as a blank. However, water samples agitated in evacuated tubes show mean zinc levels of 9-175 pg/dl greater than the zinc level found in water
TABLE
II
TRACE METAL TEST TUBES
DETERMINATIONS
Tube
Plain, acid washed Plain. evacuated Lead-free, evacuated Heparinized. evacuated Special rubber. evacuated * Refers to mean ? S.E.
IN DISTILLED
Metal concentration
DEIONIZED
WATER
STORED
IN PREPARED
(Mg/dl)
IIOn (n = 6)
Copper (n = 6)
Zinc (n = 6)
0* 0 0 8 f1.8 0.6 + 0.60
3.1 8.4 6.6 5.4 9.0
0.6 9.6 158 175 12
? k 2 k +
0.85 1.75 1.15 1.12 2.3
? 0.60 f 1.12 ? 11.4 f 10.3 f 0.9
27 TABLE III TRACE
METAL
DETERMINATIONS
IN STANDARD
SOLUTIONS
STORED
IN PREPARED
TEST
TUBES Tube
Plain, acid washed Plain, evacuated Lead-free, evacuated Heparinized. evacuated Special rubber, evacuated
Metal concentration (pg/dl) IlWn (n= 6)
Copper (n = 6)
Zinc (n = 6)
122 126 136 175 140
120 f 124 + 120+ 121 + 118?
120+ 0 1572 + 199.4 2604 f 322.3 1608 f 127.9 158 + 6.2
+ f f f +
5.6 * 4.2 4.1 4.4 5.6
1.9 2.4 0 1.2 1.5
* Refers to mean f S.E.
stored in plain acid-washed tubes. Lead-free and heparinized tubes are clearly contaminated with zinc. The special rubber-stoppered tubes yield higher readings than do acid-washed plain tubes. This difference is comparable to the difference observed in clinical samples collected in the two kinds of tubes. Trace metal determinations in standard solutions When trace metal solutions in dilute HCl are agitated within the various collection tubes, there is no remarkable copper contamination from any of the tubes (Table III). However, there is measurable iron contamination in all tubes except the plain acid-washed tubes. There is marked zinc contamination of all tubes except the plain acid-washed tubes. Discussion Within recent years, a number of disorders have been recognized to be associated with disordered trace metal metabolism. In particular, acrodermatitis enteropathica [ 51 and Menkes’ kinky hair disease [6] have been shown to be inherited abnormalities of zinc and copper absorption, respectively. Zinc deficiency has also been recognized as a consequence of poor nutrition [7] and in sickle cell disease [ 81. Disordered copper metabolism has long been recognized in Wilson’s disease [ 91. It has also become apparent that there are important metabolic interactions of the three metals, iron, copper, and zinc. The copper enzyme, ceruloplasmin, appears to play a role in the transport mechanism for iron [lo]. Zinc and iron deficiencies may be observed in the same individual [ 111, and high doses of zinc may result in decreased serum copper levels [ 121. Thus, there is an increasing clinical demand for the determination of trace metal concentrations in the blood and for techniques which permit the rapid, efficient screening of large numbers of patients. However, it is not widely recognized that trace metal8 are ubiquitous and that inadequately prepared laboratory equipment results in the contamination of specimens. In particular, zinc contamination has been traced to aspirating needles [ 131 as well as to the use of applicator sticks for dislodging clot from serum [ 11. Traditionally, trace metal analysis has been carried out in acid-washed glass-
28
ware or in metal-free plastic [14]. These methods do not lend themselves to routine analysis in most clinics since the usual blood collection system is geared to the use of multiple samples gathered in commercially available evacuated tubes. Our studies have shown that collection methods which have been advocated as satisfactory for the measurement of serum iron and copper [15] are inadequate for the measurement of serum zinc. In an attempt to overcome the zinc contamination associated with the rubber used in the stopper; of evacuated tubes, stoppers have been made by a commercial process with low zinc contamination. Tubes equipped with such stoppers appear to be adequate for iron and copper determinations. However, zinc determinations are significantly and generally consistently higher than determinations obtained from serum collected in plain, acid-washed tubes. In clinical samples, the difference of means is 10 pg/dl representing a difference of approximately 10%. This compares with a mean difference of 12 pg/dl observed when deionized distilled water is analyzed after storage in the two different tubes. Values are clearly discordant when stored standard solution is analyzed. The reasons for this discrepancy are not clear, although it is likely due to leaching of metal ions from the stopper by the dilute HCl of the standard solution
[161The discrepancy of serum zinc observed in clinical specimens is sufficient that in studies which require great accuracy, the use of blood samples collected in the special rubber-stoppered tubes cannot be justified. The normal range for serum zinc obtained from such samples (88-144 pg/dl) is higher than the normal range which we obtained from samples collected by traditional methods (76134 pg/dl) and it is higher than other published values (plasma zinc 60-116 pg/dl [l], plasma zinc 80-130pg/dl [2], serum zinc 70-180 pg/dl [17], serum zinc 68-136 pg/dl [18], serum zinc 76-132 pg/dl [19]). However, zinc determinations from blood samples obtained in the special rubber-stoppered tubes may be of sufficient reliability to be useful in large-scale clinical screening studies. This is supported by our observation that patients with hypozincemia detected with conventional collection methods are also detected when the special rubber-stoppered tubes are used for blood collection, providing that the values are compared with normal values obtained from similarly collected samples. Acknowledgements The author wishes to thank Ms. Joan Wiseman of Beckton, Dickson, Inc. for providing special rubber-stoppered Vacutainer tubes. These studies were supported in part by a research grant, AM 04495 of the National Institutes of Health, Bethesda, MD. References 1 Reimold. E.W. and Besch. D.J. (1978) Clin. Chem. 24.675-680 2 Kay, R.G., Tasman-Jones, C.. Pybus, J., Whiting. R. and Black, H. (1976) Ann. Surg. 183, 331-340 3 Shanker. J., Gibaldi. M., Kanig, J.L., Parker. A.P. and Lachman. L. (1967) J. Pharmaceut. Sci. 56, 100-108 4 Williams. D.M.. Atkin, C.L., Frens, D.B. and Bray. P.F. (1977) Pediat. Res. 11. 823-826
29 5 Moynahan.
E.J. (1974)
Lancet ii. 399400
6 Danks. D.M., CampbeII. P.E.. WaIker-Smith. J.. Stevens, B.J.. Gillespie. J.M.. Blomfield. J. and Turner, B. (1972) Lancet i. 1100-1102 7 Sandstead. H.H.. Prasad, AS.. Schulert, A.R., Farid. Z., Miale, A. Jr.. BassiIIy, S. and Derby, W.S. (1967) Am. J. Clin. Nutr. 20,422442 8 Prasad. A.S., Schoomaker, E.B., Ortega. J., Brewer, G.J.. Oberleas. D. and OeIshlegel. F.J. Jr. (1975) CIin. Chem. 21.582-587 9 Mandelbrote. B.M., Stainer. W.M.. Thompson. R.H.S. and Thurston, M.N. (1948) Brain 71. 212-228 10 Roeser. H.P., Lee, G.R., Nacht, S. and Cartwright. G.E. (1970) J. CIin. Invest. 49. 2408-2417 11 Prasad, A.S., Halstead. J.A. and Nadimi. M. (1961) Am. J. Med. 31. 532-546 12 Porter, K.G.. McMaster, D.. Elmer, M.E. and Love, A.H.G. (1977) Lancet ii, 744 13 Helman, E.Z., WalIick. D.K. and Reingold, E.M. (1971) CIin. Chem. 17,61+2 14 Reinhold. J.G. (1975) CIin. Chem. 21.476-500 15 Cartwright. G.E. (1968) in Diagnostic Laboratory Hematology, 4th Ed. PP. 210, Grune and Stratton, New York 16 CerkIewski, F.L.A. (1978) Nutr. Rep. Int. 18,1-4 17 Sinha, S.N. and Gabrieli, E.R. (1970) Am. J. Clin. Pathol. 54, 570-577 18 Pekarek, R.S.. Beisel, W.R.. Bartelloni, P.J. and Bostian. K.A. (1972) Am. J. CIin. Pathol. 57. 506510 19 Prasad, A.S., Oberleas, D. and Halstead. J.A. (1965) J. Lab. CIin. Med. 66. 508-516
Clinica Chimica Acta, 99 (1979) 23-29 @ Elsevier/North-Holland Biomedical Press
CCA 1151
TRACE METAL DETERMINATIONS IN BLOOD OBTAINED EVACUATED COLLECTION TUBES
DARRYL
IN
M. WILLIAMS
Department of Medicine, University of Utah College of Medicine, Salt Lake City, Utah 84132 and the Department of Medicine, Louisiana State University School of Medicine in Shreveport, Shreveport, LA 71130 (U.S.A.) (Received June 7th, 1979)
Summary Using atomic absorption spectrophotometry, we determined levels of iron, copper and zinc in serum or plasma samples collected in acid-washed containers and in commercially available evacuated tubes including one model equipped with rubber stoppers prepared by a process which reduces zinc contamination of the rubber. Serum or plasma from all evacuated tubes yielded iron and copper values which were comparable to those from serum obtained in acidwashed glass. Zinc values were higher from all evacuated tubes than from acidwashed glassware. Values obtained from samples collected in the speciallyprepared tubes were 10% higher than values obtained from samples collected in acid-washed glassware. Thus, the specially-prepared evacuated tubes may only be useful in large-scale clinical screening studies. Traditional blood collection methods for zinc determination should be used when accuracy is required.
Introduction The measurement of serum levels of the trace metals, iron, copper, and zinc has assumed increasing clinical importance and there is a need for reliable screening methods for large-scale clinical surveys. Analytical methods have been developed which improve the accuracy of these metal determinations, but there are numerous sources of contamination which lead to erroneously high values [l]. A common cause of contamination is the use of inadequate blood collection techniques. In particular, samples collected in rubber-stoppered evacuated tubes or with plastic syringes containing rubber plungers may give falsely elevated zinc values [ 21. Butyl rubber, used in such products, is manufactured by Correspondence should be addressed to: Dr. Williams, Department of Medicine, Louisiana sity Medical Center at Shreveport, P.O. Box 33932, Shreveport.. LA 71130 (U.S.A.).
State
Univer-
24
a process which uses zinc salts,including zinc oxide, zinc diethyldithiocarbamate and zinc dimethyldithiocarbamate [3] which result in zinc contamination of the stopper. In order to overcome this drawback, an evacuated bloodcollection tube has been developed which is sealed by a rubber stopper made by a process which does not utilize these salts. The purpose of this study is to compare the results of iron, copper, and zinc determinations in samples collected in these tubes and in other tubes commonly used for the collection of blood. Materials and methods Blood collection Blood samples were obtained from normal volunteers in the fasting state after receiving written, informed consent and with the approval of the Human Use Committee. A single specimen was drawn in an acid-washed, glass syringe mounted with a 20 gauge stainless steel needle (Monoject, Rutherford Medical Industries, Deland, FL) and transferred to an acid-washed glass test tube. Additional samples were drawn in evacuated tubes by an immediate second venipuncture. These tubes were obtained from a commercial source (Vacutainer, Becton Dickson, Inc., Rutherford, NJ) and included plain (red stoppered), leadfree (amber stoppered), heparinized (green stoppered), and special “metalfree” (blue stoppered) tubes. The order of blood drawings in evacuated tubes was varied sequentially from patient to patient. In one half of the subjects, blood was drawn in evacuated tubes in the first venipuncture before a second venipuncture with the glass syringe. Blood obtained with glass syringe or in unheparinized evacuated tubes was allowed to clot at room temperature. Blood samples obtained from cancer patients were drawn at regularly scheduled clinic visits. These samples were obtained with the patients’ signed consent and with the approval of the Human Use Commit~e. Serum or plasma was separated by centrifugation and transferred to acid-washed glass test tubes. If visible hemolysis was present in any specimen, all specimens obtained from that subject were excluded from analysis. In some studies, 2.0 ml of distilled, deionized water was added to collection tubes in place of blood. In additional tubes, 2.0 ml of solution containing 120 pg/dl of each of Fe, Cu, and Zn in 0.1 mol/l HCl were added to representative tubes. These tubes were placed on a rotary mixer for 5 min and then processed in an identical fashion to serum samples. Analytical methods Iron, copper, and zinc determinations were carried out in 2.0 ml serum or plasma samples deproteinized by the addition of trichloroacetic acid at a final concentration of 5.0%. Samples were heated under reflux in a boiling water bath for 5 min and centrifuged [43. The resulting clear supernatant was analyzed for iron, copper, or zinc using a Fer~n-Elmer model 305 atomic absorption spectrophotometer fitted with a 3-slot burner head with wavelength settings of 248.3, 324.8 and 213.9 nm respectively. The instrument was calibrated with standards prepared from metallic salt solutions (Fisher, Fair Lawn, NJ) containing 5% trichloroacetic acid in 0.15 mol/l NaCl. Recorder
25
response was linear over the range measured. Distilled, deionized water and standard metal solutions added to collection tubes were subjected to the same deproteinizing procedure before analysis. Results Serum trace metal determinations The mean values for serum iron, copper, and zinc in 40 normal volunteers are shown in Table I. Values obtained from serum collected in plain, acidwashed glass tubes are in close agreement with values obtained in other series. The values appear to have a normal distribution, and the 95% confidence limits, without regard to sex, are: iron, 67-175 pg/dl; copper, 94-166 pg/dl; and zinc, 77-133 pgldl. Values obtained from samples collected in special rubberstoppered tubes are: iron, 66-174 pg/dl; copper 91-169 pg/dl; and zinc, 88144 pg/dl. Mean values for both iron and copper are comparable regardless of the tubes used for collection. In contrast, mean zinc values are consistently higher when plain evacuated tubes, lead-free evacuated tubes, or heparinized evacuated tubes are used for collection. These values are significantly different (P < 0.001) and there is considerable variability of values obtained from samples collected in the various evacuated tubes. There is also a statistically significant difference between values obtained from samples collected in plain, acid-washed tubes and in special rubber-stoppered evacuated tubes (P < 0.01). Values from the latter are approximately 10% higher than values obtained from samples in plain acidwashed tubes with a range of lOO-124% of these values. The relationship of samples collected by the two methods from 40 normal subjects and 10 cancer patients is shown in Fig. 1. Serum zinc levels are consistently higher in blood collected in the special rubber-stoppered tubes, but there is a good correlation between the two methods (slope = 0.998, r = 0.973). Eight zinc determinations from samples contained in acid-washed tubes are identified as less than the normal range. All matched samples contained in special rubber-stoppered tubes also fall below the normal range determined for samples thus collected.
TABLE I TRACE METAL DETERMINATIONS IN SERUM OR PLASMA UNTEERS IN PREPARED TEST TUBES Tube
FROM NORMAL
Metal concentration (I.rg/dI)
plain, acid washed Plain, evacuated Lead-free. evacuated Heparinized
COLLECTED
, evacuated
Special rubber. evacuated * Refers to mean + S.E.
Iron (n = 40)
(n = 35)
Zinc (n = 40)
121 f 4.3 * 122 f 4.2 120 f 4.5
130 + 3.1 1312 3.4 128 f 2.8
105 f 2.3 194f 6.4 250 f 12.7
copper
121 2 4.4
128 f 2.7
184f
7.2
120 f 4.2
130 ? 3.3
1161
2.3
VOL-
26
IO
20
?bo 40
50
60
SERUM ZINC
7J
80
so
IJO
CONCENTRATION
ACID-WASHED
110 I20
130
140
I30
(j~g/dl)
GLASS TUBES
Fig. 1. Correlation of zinc determinations carried out in human serum obtained from blood drawn in special rubberstoppered evacuated tubes (ordinate) and in glass syringes and acid-washed glassware (abscissa). The dashed line represents hypothetically perfect correlation and the solid line represents the calculated correlation (slope = 0.998. r = 0.973). Values are consistently higher in samples obtained in
evacuated
tubes.
Trace metal determinations in distilled deionized water When distilled, deionized water is agitated within the various collection tubes, it is not remarkably contaminated by iron or copper (Table II) and iron and copper determinations are only minimally elevated above those obtained in water used as a blank. However, water samples agitated in evacuated tubes show mean zinc levels of 9-175 pg/dl greater than the zinc level found in water
TABLE
II
TRACE METAL TEST TUBES
DETERMINATIONS
Tube
Plain, acid washed Plain. evacuated Lead-free, evacuated Heparinized. evacuated Special rubber. evacuated * Refers to mean ? S.E.
IN DISTILLED
Metal concentration
DEIONIZED
WATER
STORED
IN PREPARED
(Mg/dl)
IIOn (n = 6)
Copper (n = 6)
Zinc (n = 6)
0* 0 0 8 f1.8 0.6 + 0.60
3.1 8.4 6.6 5.4 9.0
0.6 9.6 158 175 12
? k 2 k +
0.85 1.75 1.15 1.12 2.3
? 0.60 f 1.12 ? 11.4 f 10.3 f 0.9
27 TABLE III TRACE
METAL
DETERMINATIONS
IN STANDARD
SOLUTIONS
STORED
IN PREPARED
TEST
TUBES Tube
Plain, acid washed Plain, evacuated Lead-free, evacuated Heparinized. evacuated Special rubber, evacuated
Metal concentration (pg/dl) IlWn (n= 6)
Copper (n = 6)
Zinc (n = 6)
122 126 136 175 140
120 f 124 + 120+ 121 + 118?
120+ 0 1572 + 199.4 2604 f 322.3 1608 f 127.9 158 + 6.2
+ f f f +
5.6 * 4.2 4.1 4.4 5.6
1.9 2.4 0 1.2 1.5
* Refers to mean f S.E.
stored in plain acid-washed tubes. Lead-free and heparinized tubes are clearly contaminated with zinc. The special rubber-stoppered tubes yield higher readings than do acid-washed plain tubes. This difference is comparable to the difference observed in clinical samples collected in the two kinds of tubes. Trace metal determinations in standard solutions When trace metal solutions in dilute HCl are agitated within the various collection tubes, there is no remarkable copper contamination from any of the tubes (Table III). However, there is measurable iron contamination in all tubes except the plain acid-washed tubes. There is marked zinc contamination of all tubes except the plain acid-washed tubes. Discussion Within recent years, a number of disorders have been recognized to be associated with disordered trace metal metabolism. In particular, acrodermatitis enteropathica [ 51 and Menkes’ kinky hair disease [6] have been shown to be inherited abnormalities of zinc and copper absorption, respectively. Zinc deficiency has also been recognized as a consequence of poor nutrition [7] and in sickle cell disease [ 81. Disordered copper metabolism has long been recognized in Wilson’s disease [ 91. It has also become apparent that there are important metabolic interactions of the three metals, iron, copper, and zinc. The copper enzyme, ceruloplasmin, appears to play a role in the transport mechanism for iron [lo]. Zinc and iron deficiencies may be observed in the same individual [ 111, and high doses of zinc may result in decreased serum copper levels [ 121. Thus, there is an increasing clinical demand for the determination of trace metal concentrations in the blood and for techniques which permit the rapid, efficient screening of large numbers of patients. However, it is not widely recognized that trace metal8 are ubiquitous and that inadequately prepared laboratory equipment results in the contamination of specimens. In particular, zinc contamination has been traced to aspirating needles [ 131 as well as to the use of applicator sticks for dislodging clot from serum [ 11. Traditionally, trace metal analysis has been carried out in acid-washed glass-
28
ware or in metal-free plastic [14]. These methods do not lend themselves to routine analysis in most clinics since the usual blood collection system is geared to the use of multiple samples gathered in commercially available evacuated tubes. Our studies have shown that collection methods which have been advocated as satisfactory for the measurement of serum iron and copper [15] are inadequate for the measurement of serum zinc. In an attempt to overcome the zinc contamination associated with the rubber used in the stopper; of evacuated tubes, stoppers have been made by a commercial process with low zinc contamination. Tubes equipped with such stoppers appear to be adequate for iron and copper determinations. However, zinc determinations are significantly and generally consistently higher than determinations obtained from serum collected in plain, acid-washed tubes. In clinical samples, the difference of means is 10 pg/dl representing a difference of approximately 10%. This compares with a mean difference of 12 pg/dl observed when deionized distilled water is analyzed after storage in the two different tubes. Values are clearly discordant when stored standard solution is analyzed. The reasons for this discrepancy are not clear, although it is likely due to leaching of metal ions from the stopper by the dilute HCl of the standard solution
[161The discrepancy of serum zinc observed in clinical specimens is sufficient that in studies which require great accuracy, the use of blood samples collected in the special rubber-stoppered tubes cannot be justified. The normal range for serum zinc obtained from such samples (88-144 pg/dl) is higher than the normal range which we obtained from samples collected by traditional methods (76134 pg/dl) and it is higher than other published values (plasma zinc 60-116 pg/dl [l], plasma zinc 80-130pg/dl [2], serum zinc 70-180 pg/dl [17], serum zinc 68-136 pg/dl [18], serum zinc 76-132 pg/dl [19]). However, zinc determinations from blood samples obtained in the special rubber-stoppered tubes may be of sufficient reliability to be useful in large-scale clinical screening studies. This is supported by our observation that patients with hypozincemia detected with conventional collection methods are also detected when the special rubber-stoppered tubes are used for blood collection, providing that the values are compared with normal values obtained from similarly collected samples. Acknowledgements The author wishes to thank Ms. Joan Wiseman of Beckton, Dickson, Inc. for providing special rubber-stoppered Vacutainer tubes. These studies were supported in part by a research grant, AM 04495 of the National Institutes of Health, Bethesda, MD. References 1 Reimold. E.W. and Besch. D.J. (1978) Clin. Chem. 24.675-680 2 Kay, R.G., Tasman-Jones, C.. Pybus, J., Whiting. R. and Black, H. (1976) Ann. Surg. 183, 331-340 3 Shanker. J., Gibaldi. M., Kanig, J.L., Parker. A.P. and Lachman. L. (1967) J. Pharmaceut. Sci. 56, 100-108 4 Williams. D.M.. Atkin, C.L., Frens, D.B. and Bray. P.F. (1977) Pediat. Res. 11. 823-826
29 5 Moynahan.
E.J. (1974)
Lancet ii. 399400
6 Danks. D.M., CampbeII. P.E.. WaIker-Smith. J.. Stevens, B.J.. Gillespie. J.M.. Blomfield. J. and Turner, B. (1972) Lancet i. 1100-1102 7 Sandstead. H.H.. Prasad, AS.. Schulert, A.R., Farid. Z., Miale, A. Jr.. BassiIIy, S. and Derby, W.S. (1967) Am. J. Clin. Nutr. 20,422442 8 Prasad. A.S., Schoomaker, E.B., Ortega. J., Brewer, G.J.. Oberleas. D. and OeIshlegel. F.J. Jr. (1975) CIin. Chem. 21.582-587 9 Mandelbrote. B.M., Stainer. W.M.. Thompson. R.H.S. and Thurston, M.N. (1948) Brain 71. 212-228 10 Roeser. H.P., Lee, G.R., Nacht, S. and Cartwright. G.E. (1970) J. CIin. Invest. 49. 2408-2417 11 Prasad, A.S., Halstead. J.A. and Nadimi. M. (1961) Am. J. Med. 31. 532-546 12 Porter, K.G.. McMaster, D.. Elmer, M.E. and Love, A.H.G. (1977) Lancet ii, 744 13 Helman, E.Z., WalIick. D.K. and Reingold, E.M. (1971) CIin. Chem. 17,61+2 14 Reinhold. J.G. (1975) CIin. Chem. 21.476-500 15 Cartwright. G.E. (1968) in Diagnostic Laboratory Hematology, 4th Ed. PP. 210, Grune and Stratton, New York 16 CerkIewski, F.L.A. (1978) Nutr. Rep. Int. 18,1-4 17 Sinha, S.N. and Gabrieli, E.R. (1970) Am. J. Clin. Pathol. 54, 570-577 18 Pekarek, R.S.. Beisel, W.R.. Bartelloni, P.J. and Bostian. K.A. (1972) Am. J. CIin. Pathol. 57. 506510 19 Prasad, A.S., Oberleas, D. and Halstead. J.A. (1965) J. Lab. CIin. Med. 66. 508-516
