VARIOUS PREANALYTICAL VARIABLES AND THEIR EFFECTS ON THE QUALITY OF LABORATORY RESULTS
Livija Cvitkovic, Ranko Mesic
Vuk Vrhovac Institute, University Clinic for Diabetes,
Endocrinology and Metabolic Diseases,
Dugi dol 4a, 10000 Zagreb, Croatia
Received: May 28, 1999
Key words: preanalytical factors, laboratory results
The subject of this paper are a great number of different preanalytical factors that include age, gender, race, pregnancy, diet, starvation, physical activity, caffeine, cigarettes, alcohol, timing of sampling, diagnostic and therapeutic measures, posture, tourniquet, site of sampling, anticoagulants, transportation of samples, storage, processing, centrifugation, distribution, and effects of lipemia, hemolysis and hyperbilirubinemia. All these factors influence laboratory results, and proper information is needed for accurate interpretation of the results.
The physician need not be familiar with technical details of laboratory analyses, although it is favorable to know their analytical precision, reproducibility, and range of physiologic variations. Furthermore, the results are impacted by an array of preanalytical variables. The purpose of this paper is to highlight the most common of these variables (Table 1).
Table 1. Preanalytical variables
- age, gender, race, and pregnancy
- diet, starvation, and physical activity
- caffeine, cigarettes, and alcohol
- timing of sampling
- diagnostic and therapeutic measures
- posture and tourniquet
- site of sampling
- transportation of samples
- storage, processing, centrifugation, and distribution
- effects of lipemia, hemolysis, and hyperbilirubinemia
All these factors should taken into consideration to accurately interpret laboratory results. The most important issue is good cooperation between the physician and laboratory staff, thus avoiding misunderstanding in interpreting results and allowing the best treatment possible to be offered to the patient.
It is not possible to accurately supervise long-term biologic variables, but it is important to recognize them and to include them in the evaluation of laboratory results by setting up a reference range for the variables of age, gender and race.
Blood and urine analyte concentrations vary due to age factors from infancy to old age. Thus, increaesd hemoglobin concentrations are found in newborns compared with adults. Also, increased bilirubin concentrations are found in the first few days of life, reaching the highest plateau after 3-5 days. In newborn, the concentration of blood glucose is low due to their small glycogen reserves, yet some attribute it to adrenal immaturity.
Serum alkaline phosphatase activity rises with growth before puberty, and it correlates better with skeletal growth and sexual maturity than with chronological age (1). The activity is greatest at the time of maximal osteoblastic activity occurring with bone growth. After puberty, the activity rapidly decreases. From infancy to puberty, serum creatinine concentration uniformly increases, depending on the skeletal muscle development.
In the elderly, the kidney concentrating ability is decreased, which results in reduced creatinine clearance. This is due to a decreased urinary creatinine excretion as the result of reduced lean body mass rather than to renal dysfunction. The maximal tubular capacity for glucose is reduced. Plasma urea concentration and urinary protein excretion increase with age. Age does not affect basal insulin concentration, although the response to glucose is decreased (2). Alkaline phosphatase activity is greater in males than in female, however, the opposite applies when females reach over 50 years of age (3).
In addition to differences in gender specific hormones, differences also exist in hematology and clinical chemistry parameters. The concentrations of albumin, calcium, magnesium and hemoglobin are lower in women, whereas serum iron concentration is lower in women of child-bearing age. Plasma concentrations of amino acids, urea, uric acid and creatinine are higher in males than in females (2). Reticulocyte count is greater in males than in females over 20 years of age (4). The activities of aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, and creatine kinase are higher in males than in females (Fig. 1) (2).
Figure 1. Effects of race and gender on creatine kinase
It has been found that blacks have lower counts of white blood cells in comparison to whites, and this difference is explained by a decreased number of granulocytes (5). Blacks have a higher activity of creatine kinase than whites, due to a greater muscle mass in the former (Fig. 1) (6). Between the two there also is a difference in the metabolism of carbohydrates and lipids. Whites have a higher glucose tolerance compared with other races of the same age and gender (2). Studies have shown that blacks have a 1.35-fold serum concentration of vitamin B12 (7) and 2-fold concentration of Lp(a) (8) in comparison to whites.
It has been observed that plasma concentrations of cholesterol, total protein, albumin and fibrinogen, serum calcium, serum phosphate, and plasma iron concentrations as well as plasma concentrations of various female sex hormones are affected by the menstrual cycle.
The plasma iron concentration may be considerably decreased on the first day of menstruation, which also holds for the magnesium concentration which is lowest during this period. At the time of ovulation, the plasma cholesterol concentration is lowest, while the concentrations of total protein and albumin are decreased. Serum calcium decreases simultaneously with albumin decrease, however, it applies to total calcium only (2).
During pregnancy, changes in hormone production and plasma concentrations of fertility hormones entail modifications in various analytes (metabolites, electrolytes, proteins, some lipids, enzymes and thyroid hormone). The mechanisms of changes are illustrated by two examples: changes in thyroxine, lipids, copper and ceruloplasmin are caused by increased plasma transport proteins, while changes in total protein and albumin are due to hemodilution. The erythrocyte sedimentation rate is increased five-fold, and mean plasma volume rises from about 2600 mL to 3900 mL. A 50% physiologic increase in the glomerular filtration rate takes place in the last trimester (1).
EFFECT OF DIET, EXERCISE, SMOKING, CAFFEINE, AND ALCOHOL
Clinical chemistry analytes are greatly influenced by diet and drinking. The greatest increase is seen in serum concentrations of glucose, iron, lipids and alkaline phosphatase, the latter as a predominantly intestinal isoenzyme being further increased by a high-fat meal. Hydrochloric acid is produced in response to the intake of food causing a reduction in plasma chloride anions (2). A high-fat diet leads to an increased concentration of triglycerides and reduces serum urate (9). A high-protein diet increases plasma urea, serum cholesterol, and phosphate concentrations. In addition, there is an increase in ammonia and uric acid (2). The changes induced by a carbohydrate meal (75 g) are used in glucose tolerance testing for diagnostic purposes. Figure 2 shows changes of some analytes 2 h after a standard meal (10).
Figure 2. Change (%) in serum concentration of some analytes after a standard meal
In contrast, starvation, fasting and malnutrition also induce clinically significant changes in analyte concentrations. Short-term starvation (48 h) leads to significant changes that include an increase in organic acids, mainly ketone bodies (acetoacetic acid, b-hydroxybutyric acid), which causes metabolic acidosis with a decrease of both pH and bicarbonates (11). Long-term starvation entails decreased concentrations of blood protein, cholesterol, triglycerides, apolipoproteins, and urea. On the other hand, the concentrations of creatinine and uric acid increase (Fig. 3) (12).
Figure 3. Change (%) in serum concentration of some analytes after 4-week starvation
Within the first three days of starvation, the blood glucose concentration decreases by as much as 1 mmol/L, insulin secretion is reduced, and glucagon secretion rises in an attempt to maintain normal glucose concentration (2). Stimulation of lipolysis and hepatic ketogenesis commences and the main sources of energy intended for muscle now are ketoacids and fatty acids. There is an increase in the plasma concentrations of ketone bodies, fatty acids, and glycerol. It is obvious that prolonged starvation causes a reduced energy consumption, thus serum concentrations of T4 and T3 are reduced. Urinary excretion of ammonia and creatinine is increased, whereas the excretion of urea, calcium and phosphate is reduced (13).
In malnutrition, total serum protein, albumin and b-globulin concentrations are decreased. At the very beginning of malnutrition, the concentrations of complement C3, retinol-binding globulin, transferrin and prealbumin rapidly decrease (14). There is also a decrease in the plasma lipoprotein concentration, serum cholesterol and triglycerides. The concentrations of cholesterol and triglycerides may be only 50% of the concentration found in a healthy individual. Due to the decreased skeletal mass, serum concentrations of urea and creatinine as well as creatinine clearance are reduced (2). In protein calorie malnutrition, the erythrocyte count and plasma concentration of folate as well as blood hemoglobin are reduced, while the concentration of iron is slightly affected (15). Partly due to the reduced concentration of thyroxine binding globulin and prealbumin, there is a decreased plasma concentration of total triiodothyronine, thyroxine, and thyroid-stimulating hormone (2).
Changes in analyte concentrations during exercise are due to volume shifts between the intravasal and interstitial compartments, volume loss by sweating, and changes in hormone concentrations (increase in the concentrations of epinephrine, norepinephrine, glucagon, somatotropin, cortisol and ACTH, and decrease in insulin concentration) (16,17). The activities of creatine kinase have been shown to rise 4 times, pyruvate kinase 2.6 times, AST and bilirubin concentration 1.4 times, and urea 1.3 times. Blood sampling was performed on the day before and 45 min after a marathon race (18). This change depends on many individual and/or environmental factors, such as air temperature, intake of electrolyte- and carbohydrate-containing liquids during the race, and training status. It is apparent that serum uric acid concentration rises, which is due to the reduced urinary excretion, which in turn is caused by the increased lactate concentration. In addition, the increase in creatine kinase activity is greater in individuals with lower physical endurance (1).
Blood constituents are greatly affected by the intake of caffeine containing beverages such as coffee, tea, and soft drinks. The impact of caffeine on various analytes has not yet been fully investigated, however, caffeine has been found to inhibit phosphodiesterase and thus cyclic AMP degradation. Cyclic AMP leads to a rise in the blood glucose concentration. Caffeine is an activator of triglyceride lipase and causes an elevation of non-esterified fatty acids (19). Plasma renin activity and catecholamine concentrations were found to be increased 3 h after the intake of 250 mg caffeine (20). Caffeine intake over prolonged periods of time causes a decrease in serum cholesterol and an increase in serum triglycerides (2).
Smoking is the main cause of acute and chronic changes in analyte concentrations. Its effects include elevated concentrations of fatty acids, epinephrine, free glycerol, aldosterone, and cortisol. These effects occur within one hour of smoking 1-5 cigarettes. Chronic smoking entails changes in blood leukocyte and erythrocyte counts, lipoproteins, activities of some enzymes, hormones, vitamins, tumor markers, and heavy metals. Some chronic effects on blood analytes are illustrated in Fig. 4 (21).
Figure 4. Change (%) in serum composition in smokers: chronic effects
After 10 min of smoking a cigarette, glucose concentration may increase by 0.56 mmol/L and persist so for one hour. After one hour of smoking one cigarette, plasma insulin concentration shows a delayed response to the increased blood glucose. Therefore, plasma glucose concentration is higher in smokers than in nonsmokers, and glucose tolerance is slightly lower in smokers (2).
Ingestion of alcohol can lead to acute and chronic effects on clinical chemistry analytes (Fig. 5) (21-23).
Figure 5. Acute and chronic effects of alcohol consumption on some analyte concentrations
The acute effects (within 2-4 h) include a decrease in serum glucose concentration, increase in plasma lactate concentration due to inhibition of hepatic gluconeogenesis, and increase in osmolality. Ethanol is metabolized to acetaldehyde and then to acetate, which leads to a rise in hepatic uric acid formation (24). Acetate with lactate lowers serum bicarbonates, which leads to metabolic acidosis. The high lactate concentration reduces urinary excretion of uric acid, and following acute ingestion of alcohol there is an increase in serum concentration of uric acid (25). The chronic effects include, among others, increased serum activity of liver enzymes, serum triglyceride concentration, and MCV. The high activity of g-glutamyl transferase is caused by the enzyme induction. Aminotransferases (AST, ALT) as well as glutamate dehydrogenase activities are increased due to the direct liver toxic effect. Increased serum triglyceride concentrations are caused by the impaired triglyceride breakdown. The rise in MCV may be related to the direct toxic effect of ethanol on erythropoietic cells or to the shortage of folate (26).
FACTORS TO CONSIDER ON SAMPLING
Timing of sampling
It has been observed that various analytes have a circadian rhythm of their plasma concentrations, e.g., the concentration of potassium is lower in the afternoon than in the beginning of the day, and cortisol concentration rises during the course of the day to decline toward the evening (27). The individual's eating habits, rate of physical activity, and resting patterns have an impact on the circadian rhythm, however, these impacts should not be confounded with the real circadian rhythm. Sampling should be performed 12 hours after the last meal to eliminate influences of the food metabolic products (21,28). Due consideration should be paid to various possibly scheduled procedures that will probably interfere with diagnostic test results (e.g., operation, infusion and transfusion, venipuncture, injection and palpation, endoscopy, dialysis, physical stress, function tests, psychological stress, ionizing radiation, immunoscintigraphy, etc.) (29-31). It has been found that psychological stress has an impact on various blood components, such as creatine kinase, lactate dehydrogenase, and total protein. Females tend to be more sensitive than males to this kind of influence (32).
Body posture can influence the concentration of various analytes in blood. The effective filtration pressure, i.e. difference between capillary pressure and osmotic colloidal pressure in plasma, rises in the lower parts of the body when the person changes from supine to upright position. This entails water migration from the intervasal compartment to the interstitium, causing a plasma volume reduction by about 12%. When changing from upright to supine position, a reduction in the effective filtration pressure takes place, which in turn causes volume shift in the reverse direction. In the upright position, many body particles with a diameter greater than 4 nm cannot pass through the membranes and cannot follow this volume shift, thus most cellular and micromolecular analytes increase by approximately 5% to 15% compared with supine position. The change from supine to upright position leads to an increase in the secretion of catecholamines, aldosterone, angiotensin II, renin, and antidiuretic hormone (33).
A tourniquet is consistently used to obstruct the return of venous blood to the heart, and it should not be left in position for more than one minute, because it causes the moving of fluid and low molecular compounds from the intravasal space to the interstitium. Fluid migration is due to the increased filtration pressure, similar to that in changing posture from lying to standing. Macromolecules, compounds bound to protein and blood cells cannot follow this, therefore their concentration rises (1). The plasma/serum analyte concentrations and coagulation factors are not influenced by the constriction time of one minute with successive releases of the tourniquet (34). Prolonged tourniquet application entails changes in the concentration of many analytes, e.g., total protein, calcium, alanine aminotransferase, aspartate aminotransferase, creatine kinase, bilirubin, and potassium (35).
An anticoagulant must be added to the sample if whole blood or plasma is to be analyzed.
Heparin causes least interference with various tests. It is available as sodium, potassium, lithium, and ammonium salt. Lithium salt of heparin is most widely used to obtain plasma for clinical chemistry analyses. Ammonium salt of heparin can disturb the blood urea nitrogen assay when ammonium ions are to be determined. Heparin inhibits acid phosphatase activity and obstructs binding of calcium to EDTA in some methods, and influences the binding of triiodothyronine and thyroxine to their carrier proteins (1,2). Heparin used in a liquid form for determination of ionized calcium in plasma or whole blood can cause preanalytical errors by dilution and by changing the original value through binding or re-equilibration with calcium in the anticoagulant solution (36,37).
EDTA is useful in hematologic analyses because of its ability to maintain blood cells. The International Council for Standardization in Hematology (ICSH) recommends dipotassium EDTA for the collection of blood samples to be used for counting and sizing blood cells (38). EDTA inhibits activities of alkaline phosphatase and creatine kinase (2).
Sodium citrate has little application in clinical chemistry, but it is a standard anticoagulant in global coagulation tests (2,39).
Sodium fluoride and lithium iodoacetate are used to preserve glucose. Mannose and fluoride can be used as well. Fluoride inhibits the enolase enzyme activity in the glycolytic pathway and therefore efficiently preserves glucose concentration. Mannose is used in combination with fluoride to better preserve glucose. Mannose is a short-lived inhibitor which inhibits up to 4 hours from blood collection, whereas fluoride becomes effective beyond 3 hours after sampling. The loss of glucose is more effectively minimized by using a mixture of mannose and sodium fluoride than with fluoride or lithium iodoacetate alone (40,41).
SAMPLE TRANSPORTATION AND STORAGE
Timing and temperature of sample transportation have a considerable effect on the samples, therefore separation of serum and plasma from blood cells in a refrigerated centrifuge should be performed for all temperature sensitive analytes. Plasma and serum should be separated from blood cells within 2 hours. If this is not possible, the samples should be kept at room temperature rather than cooled at 4 °C, in order to decrease hemolysis. Stability and minimal evaporation can be achieved by storing separated serum in closed tubes at 4 °C, if analyses cannot be immediately performed. When there are indications that the sample may not be stable at this temperature, it should be maintained at -20 °C (1,2).
EFFECTS OF LIPEMIA
Lipemia is characterized by turbidity of samples, which may range from slightly opaque through translucent, turbid to milky appearance. The factor causing turbidity is the rise in serum lipoprotein content. Sample turbidity is always clinically relevant, as normal samples never show any turbidity unless the samples is taken after consumption of a high-fat meal, and this should always be reported and noted by the laboratory (1). Lipemia can cause false low or false high test results due to inhomogeneity, water displacement, optical interference by turbidity, and interference of physicochemical mechanisms. Inhomogeneity is characterized by a disproportionately high concentration of lipids in the upper phase (due to centrifugation), which may cause interferences in other methods such as total protein. Lipids may displace water in the upper phase of the sample, and this is responsible for the higher concentrations of sodium and potassium observed in direct ion-sensitive electrode measurement compared with flame photometry (42). Therefore, lipemia is the interference factor for many tests determining different blood constituents (a-amylase, total calcium, copper, iron, lactate, lactate dehydrogenase, potassium, sodium, total protein, and hemoglobin) (43). When the above mentioned interferences take place during testing, triglycerides may be removed from the sample by ultracentrifugation (44) or precipitation (45), and then it is necessary to repeat the test with a clarified sample.
EFFECTS OF HEMOLYSIS
Hemolysis occurs when the constituents of blood cells are released into plasma or serum, and it is characterized by a reddish appearance of the plasma or serum. This discoloration is due to the release of hemoglobin from erythrocytes. Various effects of hemolysis include an increase of intracellular constituents in the extracellular fluid (46), optical interference due to the color of hemoglobin, and chemical, biochemical and immunologic interferences by intracellular constituents with the reaction mechanism of the assay. For this reason, hemolysis is an interference factor for the analysis of albumin, total protein, bilirubin, uric acid, inorganic phosphate, creatinine, lipase, aspartate aminotransferase, creatine kinase, and g-glutamyl transferase (47). Prevention of hemolysis can be achieved through standardization of the preanalytical phase, i.e. use of standardized needles, closed tubes, and calibrated centrifuges. Also, the use of plasma instead of serum can minimize hemolysis (48).
EFFECTS OF HYPERBILIRUBINEMIA
A high concentration of serum bilirubin is visible as an intensive yellow color of serum. Bilirubin can cause direct optical interference because of its intensive color, but it also has indirect effects. Bilirubin is an interference factor for determination of cholesterol. It can cause positive interference with the Liebermann-Burchardt and iron salt methods, and negative interference in enzymatic methods. Also, bilirubin can cause false positive results in creatinine determination (49).
HANDLING OF SAMPLES
Patient identification is a major concern which needs to be paid special attention. Using patient names has proved quite questionable, as they are insufficient and may lead to confusion and mix-ups, therefore it is more practical to assign each patient a number (50). This number should contain the date of sampling. Each sample and subsample container has to be labeled with the patient's number to ensure accuracy.
There are some general guidelines and recommendations that should be followed to ensure quality of the samples and laboratory results: time between sampling and analysis should be reduced to minimum; samples should be kept at low temperature (does not apply to whole blood); samples should be stored in capped containers; hemolysis can be prevented by avoiding shaking of sample vessels; avoiding whole blood storage; and samples should be kept in dark to prevent the effects of light (1,51).
Serum is obtained from clotted blood (blood clots approximately 30 min after sampling) by centrifugation. Centrifugation is usually performed at 20 to 22 °C, however, the analytes known to be labile on centrifugation at 20 °C should be centrifugated at a refrigerated temperature. Refrigeration, however, may lead to leakage of potassium from the cell and leads to an increased value. The obtained serum can then be submitted to analysis (52).
Diagnostic methods, including biochemistry analyses, are becoming ever more numerous and more sophisticated. The physician's reliance and dependency on these methods are so great that occasionally the basic clinical approach to the patient seems to become secondary. Due to the importance and role of diagnostic methods, their performance and interpretation are associated with considerable responsibility. While the analyses are technically becoming ever more perfect, it seems to also be ever more important to pay due consideration to different preanalytical factors that can influence the laboratory end result. While some of these factors are well known, a number of factors have been paid quite little attention. It is of great importance that physicians be aware of these factors, because the final interpretation of the analyses and decisions on therapeutic procedures based on this interpretation actually lead to the physician's final judgment. In this process, the laboratory should act as a reliable and consistent collaborator.
- Guder WG, Narayanan S, Wisser H, Zawta B, eds. Samples: from the patient to the laboratory. Darmstadt: GIT Verlag, 1996.
- Young DS, Bermes EW. Specimen collection and processing: sources of biological variation. In: Tietz NW, ed. Textbook of clinical chemistry. Philadelphia: WB Saunders Company, 1987:58-101.
- Schiele F, Henny J, Hitz J, Petitclerc C, Gueguen R, Siest G. Total bone and liver alkaline phosphatases in plasma: biological variations and reference limits. Clin Chem 1983;29:634-641.
- Tarallo P, Humbert JC, Mahassen P, Fournier B, Henny J. Reticulocytes: biological variations and reference limits. Eur J Haematol 1994;53:11-15.
- Karayalcin G, Rosner F, Sawitsky A. Pseudoneutropenia in American negros. Lancet 1972;i:387.
- Harris EK, Wong ET, Shaw ST Jr. Statistical criteria for separate reference intervals: race and gender groups in creatine kinase. Clin Chem 1991;37:1580-1582.
- Saxena S, Carmel R. Racial difference in vitamin B12 levels in the United States. Am J Clin Pathol 1987;88:95-97.
- Guyton JR, Dahlen GH, Patsch W, Kautz JA, Gotto AM Jr. Relationship of plasma lipoprotein Lp(a) levels to race and to apolipoprotein B. Arteriosclerosis 1985;5:265-272.
- Rifai N, Merill JR, Holly RG. Postprandial effect of a high fat meal on plasma lipid, lipoprotein, cholesterol and apolipoprotein measurement. Ann Clin Biochem 1990;27:489-493.
- Steinmetz J, Panek E, Sourieau F, Siest G. Influence of food intake on biological parameters. In: Siest G, ed. Reference values in human chemistry. Basle: Karger, 1973:195-200.
- Lamers KJB, Doesburg WH, Gabreëls FJM, et al. The concentration of blood components related to fuel metabolism during prolonged fasting in children. Clin Chim Acta 1985;152:155-163.
- Ditschuneit H, Wechsler J. Probleme der ambulanten Nulldiätbehandlung. Dtsch Arztebl 1979;75:871-880.
- Wisser H. Praeanalytische Faktoren bei der Harnanalytik. In: Guder WG, Lang H, eds. Pathobiochemie und Funktionsdiagnostik der Niere. Berlin, Heidelberg, New York: Springer Verlag, 1989:171-182.
- Olusi SO, McFarlane H, Osunkoya BO. Specific protein assays in protein calorie malnutrition. Clin Chim Acta 1975;62:107-116.
- Khalil M, Tanios A, Moghazy M. Serum and red cell folates, and serum vitamin B12 in protein calorie malnutrition. Arch Dis Child 1973;48:366-369.
- Alexander S. Physiologic and biochemical effects of exercise. Clin Biochem 1984;17:126-131.
- Cumming DC. Hormones and athletic performance. In: Felig P, Baxter JD, Frohman LA, eds. Endocrinology and metabolism. New York, Toronto: McGraw-Hill, 1995:1837-1885.
- Stansbie D, Begley JP. Biochemical consequence of exercise. J IFCC 1991;3:87-91.
- Narayanan S. Physiological variables in blood sampling. Mitt Klin Chem 1993;24:130-134.
- Robertson D, Frölich JC, Carr RK, et al. Effects of caffeine on plasma renin activity, catecholamines and blood pressure. N Engl J Med 1978;298:181-186.
- Young DS. Effects of preanalytical variables on clinical laboratory tests. Washington, DC: AACC Press, 1993.
- Leppaluoto J, Vuolteenaho O, Arjamaa O. Plasma immunoreactive atrial natriuretic peptide and vasopressin after ethanol intake in man. Acta Physiol Scand 1992;144:121-127.
- Ratge D, Brugger G, Wehr M, Bode JCh, Wisser H. Catecholamines in plasma and urine of patients with alcoholic liver damage under resting and exercise conditions. J Clin Chem Clin Biochem 1985;23:447-452.
- Zilva JF, Pannall PR, Mayne PD, eds. Clinical chemistry in diagnosis and treatment. 5th Ed. Sloyd-Luke Publication, 1988.
- Statland BE, Winkel P. Selected preanalytical sources of variation. In: Gräsbeck R, Alström T, eds. Reference values in laboratory medicine. Chichester: John Wiley & Sons, Ltd., 1981:127-137.
- Freer DE, Statland BE. Effects of ethanol (0.75 g/kg of body weight) on the activities of selected enzymes in sera of healthy young adults. 2. Interindividual variations in response of g-glutamyl transferase to repeated ethanol challenges. Clin Chem 1977;23:2099-2102.
- Touitou Y, Haus E, eds. Biologic rhythms in clinical and laboratory medicine. Berlin, Heidelberg, New York: Springer Verlag, 1992.
- Statland BE, Winkel P. Effects of preanalytical factors on the intraindividual variation of analytes in the blood of healthy subjects: considerations of preparation of the subject and time of venipuncture. Crit Rev Clin Lab Sci 1977;8:105-144.
- Hagemann P. Auftrag/Specimen/Befund, Bindeglied zwischen Klinik und Laboratorium. Darmstadt: GIT Verlag, 1989.
- Keller H. Klinisch-chemische Labordiagnostik für die Praxis. Analyse, Befund, Interpretation. Stuttgart, New York: G. Thieme, 1991.
- Wisser H. Einflubgröben und Störgröben. In: Greiling H, Gressner AM, eds. Lehrbuch der klinischen Chemie und Pathobiochemie. Stuttgart, New York: Schattauer Verlag, 1995:50-71.
- Dugue B, Leppanen EA, Zhou HP, Grasbeck R. Preanalytical factors and standardized specimen collection: influence of psychological stress. Scand J Clin Lab Invest 1992;52:43-50.
- Felding P, Tryding N, Hyltoft Petersen P. Effects of posture on concentrations of blood constituents in healthy adults: practical application of blood specimen collection procedures recommended by the Scandinavian Committee on Reference Values. Scand J Clin Lab Invest 1980;40:615-621.
- Witt I, Beeser H, Müller-Berghaus G. Minimanforderungen zur Gewinnung von Citratplasma für hämostaseologische Analysen. Lab Med 1995;19:143-145.
- Statland BE, Bokelund H, Winkel P. Factors contributing to intraindividual variation of serum constituents. 4. Effects of posture and tourniquet application on variation of serum constituents in healthy subjects. Clin Chem 1974;20:1513-1519.
- Sachs C, Rabouine P, Chaneac M, Kindermans C, Dechaux M, Falch-Christiansen T. Preanalytical errors in ionized calcium measurements induced by the use of liquid heparin. Ann Clin Biochem 1991;28:167-173.
- Sachs C, Rabouine P, Chaneac M, Kindermans C, Dechaux M. Anticoagulant-induced preanalytical errors in ionized calcium determination on blood. Scand J Clin Lab Invest 1989;49:647-651.
- Anonymous. Recommendations of the International Council for Standardization in Hematology for ethylene diamine tetra-acetic acid anticoagulation of blood for blood cell counting and sizing. Am J Clin Pathol 1993;100:371-372.
- Koepke JA, Rogers J, Allwier MJ. Pre-instrumental variables in coagulation testing. Am J Clin Pathol 1975;61:591-596.
- Chan AYW, Swaminathan R, Cockram CS. Effectiveness of sodium fluoride as a preservative of glucose in blood. Clin Chem 1989;35:315-317.
- Liss E, Bechtel S. Improvement of glucose preservation in blood samples. J Clin Chem Clin Biochem 1990;28:689-690.
- Külpmann WR. Determination of electrolytes in serum and serum water. Wien Klin Wochenschr 1992;Suppl 1992:34-38.
- Grafmeyer D, Bondon M, Manchon M, Levillain P. The influence of bilirubin, haemolysis and turbidity on 20 analytical tests performed on automatic analysers. Eur J Clin Chem Clin Biochem 1995;33:31-52.
- Bermes EW, Young DS. Specimen collection and processing: sources of biological variation. In: Tietz NW, ed. Fundamentals of clinical chemistry. Philadelphia: WB Saunders Company, 1987:266-286.
- Lamprecht J, Seidel D. Enzymdiagnostik bei Patienten mit Hyperlipoproteinämie: Beseitgung von Plasmatrübungen durch selektive Polyanionenpräzipitation von Plasma-Lipoproteinen. Z Klin Chem Klin Biochem 1974;12:154.
- Sonntag O. Haemolysis as an interference factor in clinical chemistry. J Clin Chem Clin Biochem 1986;24:127-139.
- Guder WG. Haemolysis as an influence and interference factor in clinical chemistry. J Clin Chem Clin Biochem 1986;24:125-126.
- Lutomski DM, Bower RH. The effect of thrombocytosis on serum potassium and phosphorus concentrations. Am J Med Sci 1994;307:255-258.
- Straus B. Medicinska biokemija. Zagreb: Medicinska naklada, 1992.
- Bonini PA, Alpert N, Luzzana M, Rubin R. Guidelines for the identification and distribution of patient samples in the medical laboratory. J Autom Chem 1994;16:25-32.
- Alström T, Grasbeck R, Lindblad B, Solberg HE, Winkel P, Viinikka L. Establishing reference values from adults: recommendation on procedures for the preparation of individuals, collection of blood, and handling and storage of specimens. Committee on Reference Values of the Scandinavian Society for Clinical Chemistry. Scand J Clin Lab Invest 1993;53:649-652.
- National Committee for Clinical Laboratory Standards. Procedures for the handling and processing of blood specimens. Villanova, Pennsylvania, 1984:4:No. 9.