Preanalysis

Physiologic Factors

Diurnal variation. This may be encountered when testing for hormones, iron, acid phosphatase, and urinary excretion of most electrolytes such as sodium, potassium, and phosphate (Dufour, 2003). Table 3-2 presents several tests affected by diurnal variations, posture, and stress.

Exercise. Physical activity has transient and long-term effects on laboratory determinations. Transient changes may include an initial decrease followed by an increase in free fatty acids, and lactate may increase by as much as 300%. Exercise may elevate creatine phosphokinase (CK), aspartate aminotransferase (AST), and lactate dehydrogenase (LD) and may activate coagulation, fibrinolysis, and platelets (Garza & Becan-McBride, 2014). These changes are related to increased metabolic activities for energy purposes and usually return to preexercise levels soon after exercise cessation. Long-term effects of exercise may increase CK, aldolase, AST, and LD values. Chronic aerobic exercise is associated with lesser increases in plasma concentration of muscle enzymes such as CK, AST, alanine aminotransferase (ALT), and LD. Decreased levels of serum gonadotropin and sex steroid concentrations are seen in long-distance athletes, while prolactin levels are elevated (Dufour, 2003).

Diet. An individual's diet can greatly affect laboratory test results. The effect is transient and is easily controlled. Glucose and triglycerides, absorbed from food, increase after eating (Dufour, 2003). After 48 hours of fasting, serum bilirubin concentrations may increase. Fasting for 72 hours decreases plasma glucose levels in healthy women to 45 mg/dL (2.5 mmol/L), while men show an increase in plasma triglycerides, glycerol, and free fatty acids, with no significant change in plasma cholesterol. When determining blood constituents such as glucose, triglycerides, cholesterol, and electrolytes, collection should be done in the basal state (Garza & Becan-McBride, 2014). Eating a meal, depending on fat content, may elevate plasma potassium, triglycerides, alkaline phosphatase, and 5-hydroxyindoleacetic acid (5-HIAA). Stool occult blood tests, which detect heme, are affected by the intake of meat, fish, iron, and horseradish, a source of peroxidase, causing a false-positive occult blood reaction (Dufour, 2003). In addition, consumption of bismuth-containing antacids such as Pepto-Bismol also renders false-positive results. Physiologic changes may include hyperchylomicronemia, thus increasing turbidity of the serum or plasma and potentially interfering with instrument readings.

Certain foods or diet regimens may affect serum or urine constituents. Long-time vegetarian diets are reported to cause decreased concentrations of low-density lipoproteins (LDLs), very-low-density lipoproteins (VLDLs), total lipids, phospholipids, cholesterol, and triglycerides. Vitamin B₁₂ deficiency can also occur, unless supplements are taken (Young, 2007). A high-meat or other protein-rich diet may increase serum urea, ammonia, and urate levels. High-protein, low-carbohydrate diets, such as the Atkins diet, greatly increase ketones in the urine and increase the serum blood urea nitrogen (BUN). Foods with a high unsaturated-to-saturated fatty acid ratio may show decreased serum cholesterol, while a diet rich in purines will show an increased urate value. Foods such as bananas, pineapples, tomatoes, and avocados are rich in serotonin. When ingested, elevated urine excretion of 5-HIAA may be observed. Beverages rich in caffeine elevate plasma free fatty acids and cause catecholamine release from the adrenal medulla and brain tissue. Ethanol ingestion increases plasma lactate, urate, and triglyceride concentrations. Elevated high-density lipoprotein (HDL) cholesterol, γ-glutamyl transferase (GGT), urate, and mean corpuscular volume (MCV) have been associated with chronic alcohol abuse.

Serum concentrations of cholesterol, triglycerides, and apoB lipoproteins are correlated with obesity. Serum LD activity, cortisol production, and glucose increase in obesity. Plasma insulin concentration is also increased, but glucose tolerance is impaired. In obese men, testosterone concentration is reduced (Young, 2007).

Stress. Mental and physical stresses induce the production of adrenocorticotropic hormone (ACTH), cortisol, and catecholamines. Total cholesterol has been reported to increase with mild stress, and HDL cholesterol to decrease by as much as 15% (Dufour, 2003). Hyperventilation affects acid-base balance and elevates leukocyte counts, serum lactate, or free fatty acids.

Posture. Posture of the patient during phlebotomy can have an effect on various laboratory results. An upright position increases hydrostatic pressure, causing a reduction of plasma volume and increased concentration of proteins. Albumin and calcium levels may become elevated as one changes position from supine to upright. Elements that are affected by postural changes are albumin, total protein, enzymes, calcium, bilirubin, cholesterol, triglycerides, and drugs bound to proteins. Incorrect application of the tourniquet and fist exercise can result in erroneous test results. Using a tourniquet to collect blood to determine lactate concentration may result in falsely increased values. Prolonged tourniquet application may also increase serum enzymes, proteins, and protein-bound substances, including cholesterol, calcium, and triglycerides, as the result of hemoconcentration when plasma water leaves the vein because of back pressure. After bed rest in the hospital, a patient's hemoglobin (Hb) can decrease from the original admitting value enough to falsely lead a physician to suspect internal hemorrhage or hemolysis (Dufour, 2003). This effect can be amplified by intravenous fluid administration. Patients should be advised to avoid changes in their diet, consumption of alcohol, and strenuous exercise 24 hours before having their blood drawn for laboratory testing.

Age. Age of the patient has an effect on serum constituents. Young defines four age groups: newborn, childhood to puberty, adult, and elderly adult (Young, 2007). In the newborn, much of the Hb is Hb F, not Hb A, as seen in the adult. Bilirubin concentration rises after birth and peaks at about 5 days. In cases of hemolytic disease of the fetus and newborn (HDFN), bilirubin levels continue to rise. This often causes difficulty in distinguishing between physiologic jaundice and HDFN. Infants have a lower glucose level than adults because of their low glycogen reserve. With skeletal growth and muscle development, serum alkaline phosphatase and creatinine levels, respectively, also increase. The high uric acid level seen in a newborn decreases for the first 10 years of life, then increases, especially in boys, until the age of 16 (Young, 2007). Most serum constituents remain constant during adult life until the onset of menopause in women and middle age in men. Increases of about 2 mg/dL (0.05 mmol/L) per year in total cholesterol and 2 mg/dL (0.02 mmol/L) per year in triglycerides until midlife have been reported. The increase in cholesterol seen in postmenopausal women has been attributed to a decrease in estrogen levels. Uric acid levels peak in men in their 20s but do not peak in women until middle age. The elderly secrete less triiodothyronine, parathyroid hormone, aldosterone, and cortisol. After age 50, men experience a decrease in secretion rate and concentration of testosterone, and women have an increase in pituitary gonadotropins, especially follicle-stimulating hormone (FSH) (Young, 2007).

Gender. After puberty, men generally have higher alkaline phosphatase, aminotransferase, creatine kinase, and aldolase levels than women; this is due to the larger muscle mass of men. Women have lower levels of magnesium, calcium, albumin, Hb, serum iron, and ferritin. Menstrual blood loss contributes to the lower iron values (Young, 2007).

Tobacco smokers have high blood carboxyhemoglobin levels, plasma catecholamines, and serum cortisol. Changes in these hormones often result in decreased numbers of eosinophils, while neutrophils, monocytes, and plasma free fatty acids increase. Chronic effects of smoking lead to increased Hb concentration, erythrocyte (RBC) count, MCV, and leukocyte (WBC) count. Increased plasma levels of lactate, insulin, epinephrine, and growth hormone and urinary secretion of 5-HIAA are also seen. Vitamin B₁₂ levels may be substantially decreased and have been reported to be inversely proportional to serum thiocyanate levels. Smoking also affects the body's immune response. Immunoglobulin (Ig)A, IgG, and IgM are lower in smokers, and IgE levels are higher. Decreased sperm counts and motility and increased abnormal morphology have been reported in male smokers when compared with nonsmokers (Young, 2007).

In Vitro

Collection-Associated Variables

On occasion, when there is a problem finding a vein for phlebotomy, the specimen may be hemolyzed as the result of sheer forces on the red blood cells. Hemolysis can also be caused by using a needle that is too small, pulling a syringe plunger back too fast, expelling the blood vigorously into a tube, shaking or mixing the tubes vigorously, or performing blood collection before the alcohol has dried at the collection site. Hemolysis is present when the serum or plasma layer is pink. Hemolysis can falsely increase blood constituents such as potassium, magnesium, iron, LD, phosphorus, ammonium, and total protein (Garza, 2014). Table 3-3 shows changes in serum concentrations (or activities) of selected constituents caused by lysis of RBCs.

TABLE 3-3

Changes in Serum Concentration (or Activities) of Selected Constituents Due to Lysis of Erythrocytes (RBCs)

Constituent	Ratio of Concentration (or Activity) in RBC to Concentration (or Activity) in Serum	Percent Change of Concentration (or Activity) in Serum after Lysis of 1% RBC, Assuming a Hematocrit of 0.50
Lactate dehydrogenase	16 : 1	+272.0
Aspartate aminotransferase	4 : 1	+220.0
Potassium	23 : 1	+24.4
Alanine aminotransferase	6.7 : 1	+55.0
Glucose	0.82 : 1	–5.0
Inorganic phosphate	0.78 : 1	+9.1
Sodium	0.11 : 1	–1.0
Calcium	0.10 : 1	+2.9

Because of the extremely important role of potassium in cardiac excitation, elevations due to hemolysis can be problematic, especially for emergency room patients who are at risk of hemolysis during frantic blood collection. The relationship between level of hemolysis and potassium (as determined on a Siemens ADVIA 1650 chemistry analyzer [Siemens Healthcare Diagnostics, Deerfield, Ill.]) in serum and plasma specimens is shown in Figure 3-1. Even with no hemolysis, the range of potassium concentrations can be broad in a combination of healthy and sick individuals. Low levels of hemolysis cause only minor elevations, but very strong hemolysis can raise the potassium level by 2 to 3 mEq/L into a critical range.

Another special case where pseudohyperkalemia can occur is in patients with extremely high blast counts in acute or accelerated phase leukemias. Those blasts can be fragile and may lyse during standard phlebotomy, releasing potassium. In contrast, specimens with very high WBC counts that are collected gently can show pseudohypokalemia when potassium is taken up by highly metabolically active leukemic cells along with glucose; such specimens can be transported on ice to slow this enzymatically mediated uptake.

Normally platelets release potassium during clotting, so serum has a slightly higher value of potassium than plasma from the same individual; this difference is accentuated when the platelet count is extremely elevated.

To avoid problems with hemoconcentration and hemodilution, the patient should be seated in a supine position for 15 to 20 minutes before the blood is drawn (Young, 2007). Extended application of the tourniquet can cause hemoconcentration, which increases the concentrations of analytes and cellular components. When blood collection tubes that contain various anticoagulants/additives are used, it is important to follow the proper order of draw and to thoroughly mix an anticoagulated tube of blood after it has been filled. Failure to mix a tube containing an anticoagulant will result in failure to anticoagulate the entire blood specimen, and small clots may be formed. Erroneous cell counts can result. If a clot is present, it may also occlude or otherwise interfere with an automated analyzer. It is very important that the proper anticoagulant be used for the test ordered. Using the wrong anticoagulant will greatly affect the test results.

Icteric or lipemic serum provides additional challenges in laboratory analysis. When serum bilirubin approaches 430 mmol/L (25 mg/L), interference may be observed in assays for albumin (4-hydroxyazobenzene-2-carboxylic acid [HABA] procedure), cholesterol (using ferric chloride reagents), and total protein (Biuret procedure). Artifactually induced values in some laboratory determinations result when triglyceride levels are elevated (turbidity) on the basis of absorbance of light of various lipid particles. Lipemia occurs when serum triglyceride levels exceed 4.6 mmol/L (400 mg/dL). Inhibition of assays for amylase, urate, urea, CK, bilirubin, and total protein may be observed. To correct for artifactual absorbance readings, “blanking” procedures (the blank contains serum but lacks a crucial element to complete the assay) or dual-wavelength methods may be used. A blanking process may not be effective in some cases of turbidity, and ultracentrifugation may be necessary to clear the serum or plasma of chylomicrons.

Special Issues That May Impact Analysis

In addition to the preanalytical variables discussed above, there are a variety of special conditions and interferences that may impact sample analysis.

Time of Collection

Sometimes, samples have to be collected at a specific time. Failure to follow the planned time schedule can lead to erroneous results and misinterpretation of a patient's condition. The most common tests in this category are the ASAP and stat collections. ASAP means “as soon as possible,” and stat is an American medical term meaning “immediately” (from the Latin statim). The exact definitions of these terms vary from one laboratory to another. Stat specimens are collected and analyzed immediately. They are given the highest priority and are usually ordered from the emergency department and critical care units (Strasinger & DiLorenzo, 2011). Timed specimens are ordered for a variety of reasons, usually to monitor changes in a patient's condition, to determine the level of a medication, or to measure how well a substance is metabolized. For example, a physician may want to monitor a cardiac marker to determine if it is rising or decreasing. In therapeutic drug monitoring, trough and peak levels of a drug may be ordered. Trough specimens reflect the lowest level in the blood and are generally drawn 30 minutes before the drug is administered. The peak specimen is drawn shortly after the medication is given; the actual collection time varies by medication. Drug manufacturers specify the length of time that must pass between trough and peak collection times.

Specimen Acceptability and Identification Issues

All specimens must be collected, labeled, transported, and processed according to established procedures that include sample volume, special handling needs, and container type. Failure to follow specific procedures can result in specimen rejection. Inappropriate specimen type, wrong preservative, hemolysis, lipemia, clots, and so on, are reasons for rejection. Specimen rejection is costly and time-consuming.

An incorrect sample collection may harm the patient, especially when the blood sample in the tube is mislabeled. The first goal of The Joint Commission 2015 Laboratory National Patient Safety Goals is to “identify patients correctly” (http://www.jointcommission.org/standards_information/npsgs.aspx). Patient wristbands that include bar codes or radiofrequency technology promote positive patient identification. A fully integrated, information technology–based preanalytical process has also been described (Barak & Jaschek, 2014). Misidentification of patients during sample collection for transfusion or at the time of transfusion can be a life-threatening medical error. The incidence of patient misidentification at the time of specimen collection has been reported to be approximately 1 in 1000, with 1 in 12,000 patients receiving a unit of blood that was not intended for that individual (Linden et al, 2000; Dzik et al, 2003). Trained phlebotomists appear to have fewer Wrong Blood in Tube (WBIT) errors (1.1 per 1000) than nondedicated staff (2.7 per 1000) (Bolton-Maggs et al, 2015). As a result, the College of American Pathologists requires laboratories to have a plan to reduce the risk of mistransfusion and suggests as options collecting two samples at separate phlebotomy events or utilizing an electronic identification verification system such as an electronic bar code reader for patient identification wristbands (CAP TRM.30575). It is therefore essential to thoroughly train all medical staff in all aspects of patient identification, specimen collection, transportation, and processing. Box 3-1 lists various reasons for specimen rejection.

Anticoagulants and Additives

Ethylenediaminetetraacetic acid (EDTA) is the anticoagulant of choice for hematology cell counts and cell morphology. It is available in lavender-top tubes as a liquid or is spray-dried in the dipotassium or tripotassium salt form (K₂EDTA in plastic, spray-dried, and K₃EDTA in liquid form in glass tubes). K₃EDTA is a liquid and will dilute the sample about 1% to 2%. K₂EDTA is spray-dried on the walls of the tube and will not dilute the sample. Pink-top tubes also contain EDTA. The EDTA is spray-dried K₂EDTA. Pink tubes are used in immunohematology for ABO grouping, Rh typing, and antibody screening. These tubes have a special cross-match label for information required by the American Association of Blood Banks (AABB) and are approved by the U.S. Food and Drug Administration (FDA) for blood bank collections. White-top tubes also contain EDTA and gel. They are used most often for molecular diagnostic testing of plasma.

For coagulation testing, a light blue–top tube containing 0.105 M or 0.129 M (3.2% or 3.8%, respectively) sodium citrate is commonly used because it preserves the labile coagulation factors. Coagulation tests require the correct ratio of blood and anticoagulant; thus containers should be adequately filled. An insufficient blood collection volume yields excess citrate in plasma and falsely elevated clotting times. This interference is also important in polycythemia, when the hematocrit is abnormally high and the volume of plasma (where citrate distributes) is small. Thus even in an apparently correctly filled tube, polycythemia can lead to falsely prolonged PT and PTT unless the amount of citrate anticoagulant in the tube is reduced proportionally to the decrease in plasma volume. A discard tube (i.e., a partially filled tube that is discarded prior to collecting the one used for testing) is normally only needed when the sample is drawn from a winged set or intravenous catheter; this is because the air space in these collection systems leads to underfilling the first tube (Favaloro et al, 2012). Black-top tubes also contain buffered sodium citrate and are generally used for Westergren sedimentation rates, as are lavender-top tubes. They differ from light blue–top tubes in that the ratio of blood to anticoagulant is 4 : 1 in the black-top tubes and 9 : 1 in the light blue–top tubes.

Heparin, a mucoitin polysulfuric acid, is an effective anticoagulant in small quantities without significant effect on many determinations. Heparin was originally isolated from liver cells by scientists looking for an anticoagulant that could work safely in humans. Heparin is available as lithium heparin (LiHep) and sodium heparin (NaHep) in green-top tubes. Heparin accelerates the action of antithrombin III, neutralizing thrombin and preventing the formation of fibrin. Heparin has an advantage over EDTA as an anticoagulant, in that it does not affect levels of ions such as calcium. However, heparin can interfere with some immunoassays. Heparin should not be used for coagulation or hematology testing. Heparinized plasma is preferred for potassium measurements to avoid an elevation due to the release of potassium from platelets as the blood clots (Garza, 2014). Lithium heparin may be used for most chemistry tests except for lithium and folate levels; for lithium, a serum specimen can be used instead. Sodium heparin cannot be used for assays measuring sodium levels, but it is recommended for trace elements, leads, and toxicology. Sodium heparin is the injectable form used for anticoagulant therapy.

Gray-top tubes are generally used for glucose measurements because they contain a preservative or antiglycolytic agent, such as sodium fluoride, which prevents glycolysis for 3 days (Strasinger & DiLorenzo, 2011). In bacterial septicemia, fluoride inhibition of glycolysis is neither adequate nor effective in preserving glucose concentration. Red-top tubes have no additive, so blood collected in these tubes clots.

Red-top tubes can be used for most chemistry, blood bank, and immunology assays. Integrated serum separator tubes are available for isolating serum from whole blood. During centrifugation, blood is forced into a thixotropic gel material located at the base of the tube. The gel undergoes a temporary change in viscosity during centrifugation and lodges between the packed cells and the top serum layer (Strasinger & DiLorenzo, 2011). Pediatric-sized tubes are also available. Advantages of serum separator tubes include (1) ease of use, (2) shorter processing time through clot activation, (3) higher serum yield, (4) minimal liberation of potentially hazardous aerosols, (5) only one centrifugation step, (6) use of single tube (same one as patient specimen), and (7) ease of a single label. A unique advantage is that centrifuged specimens can be transported without disturbing the separation. Some silica gel serum separation tubes may give rise to minute particles that can cause flow problems during analysis. Filtering the serum solves the problem.

A few specialized tubes exist. Red/gray- and gold-top tubes contain a clot activator and a separation gel. These tubes are referred to as serum separator tubes (SSTs) and are used most often for chemistry tests. Therapeutic drug monitoring specimens should not be collected in tubes that contain gel separators because some gels absorb certain drugs, causing a falsely lowered result. Significant decreases in phenytoin, phenobarbital, lidocaine, quinidine, and carbamazepine have been reported when stored in Vacutainer SST tubes (Becton, Dickinson and Company [BD], Franklin Lakes, N.J.), while no changes were noted in theophylline and salicylate levels. Storage in standard red-top Vacutainer collection tubes without barrier gels did not affect measured levels of the above therapeutic drugs (Dasgupta et al, 1994). Studies indicate that this absorption is time dependent, and therefore speed in processing minimizes absorption. Acrylic-based gels do not exhibit the absorption problems associated with silicone and polyester gels (Garza, 2014).

Tubes containing gels are not used in the blood bank or for immunologic testing because the gel may interfere with the immunologic reactions (Strasinger & DiLorenzo, 2011). Clotting time for tubes using gel separators is approximately 30 minutes, and tubes that have clot activators, such as thrombin, will clot in 5 minutes. Plain red-stoppered tubes with no additives take about 60 minutes to clot completely (Strasinger & DiLorenzo, 2011).

Anticoagulants may affect the transport of water between cell and plasma, thereby altering cell size and constituent plasma concentration. Oxalate anticoagulants may shrink red cells; thus blood anticoagulated with oxalate cannot be used to measure hematocrit. Combined ammonium/potassium oxalate does not have the same effect of shrinking cells.

EDTA, citrate, and oxalate chelate calcium, thereby lowering calcium levels. Fluoride, used for glucose determinations, prevents glycolysis by forming an ionic complex with Mg⁺⁺, thereby inhibiting the Mg⁺⁺-dependent enzyme, enolase (Young, 2007). Table 3-5 lists anticoagulants/additives and their effects on various blood tests.

Sample transfer between tubes. A serious error that is sometimes encountered in sample collection occurs when a portion of a sample in the EDTA-containing lavender-top tube for hematological analysis is poured into a red-top tube for clinical chemistry analysis. EDTA is a potassium salt, artificially increasing the potassium concentration of the sample to values around 25 meq/L, a value that would be incompatible with life. EDTA itself is an ion exchanger that exchanges potassium for calcium. Calcium levels are therefore greatly reduced in the red-top tube sample to values as low as 0. Very low calcium levels and very high potassium levels in red-top tube samples should be taken as a flag for an error due to EDTA contamination.

Blood Collection Devices

The most common blood collection system uses a vacuum to pull blood into a container; it consists of a color-coded evacuated collection tube, a double-headed needle, and an adapter/holder. Small tubes are available for pediatric and geriatric collections. The blood collection holder accommodates various sizes (gauge) of blood collection needles. Needles vary from large (16 gauge) to small (23 gauge). Several types of holders have been designed to eject the needle after use. OSHA policies require that the adapters be discarded with the used needle (OSHA, 2002). Pediatric inserts are available for adapters and accommodate the smaller-diameter pediatric blood collection tubes. Also available are a variety of safety needles that cover the needle after use or retract the needle before it is discarded.

Winged infusion sets (butterfly needles) can be used when blood has to be collected from a very small vein. Butterfly needles come in 21, 23, and 25 gauge. These needles have plastic wings attached to the end of the needle that aid in insertion of the needle into the small vein. Tubing is attached to the back end of the needle, which terminates with an adapter for attachment to a syringe or evacuated collection holder. A retractable needle design has significantly reduced needlestick injuries (Hotaling, 2009).

Blood collected in a syringe can be transferred to an evacuated tube. Special syringe safety shield devices are available to avoid unnecessary contact with the blood sample. If blood requires anticoagulation, speed becomes an important factor, and the blood must be transferred before clot formation begins. Once the blood has been transferred, the anticoagulated tube must be thoroughly mixed to avoid small clot formation.

Several additional pieces of phlebotomy equipment are necessary. A tourniquet, usually a flat latex strip or piece of tubing, is wrapped around the arm to occlude the vein before blood collection and is discarded after each phlebotomy. OSHA guidelines state that gloves should be worn when performing phlebotomy and should be changed between patients. Gloves are available in various sizes and are made of various materials to avoid latex sensitivity as experienced by some individuals. Other supplies include gauze pads, alcohol or iodine wipes for disinfection of the puncture site, and a Band-Aid (Johnson & Johnson, New Brunswick, N.J.) to prevent bleeding after completion of the phlebotomy.

Blood Storage and Preservation

During storage, the concentration of a blood constituent in the specimen may change as a result of various processes, including adsorption to glass or plastic tubes, protein denaturation, evaporation of volatile compounds, water movement into cells resulting in hemoconcentration of serum and plasma, and continuing metabolic activities of leukocytes and erythrocytes. These changes occur, although to varying degrees, at ambient temperature and during refrigeration or freezing. Storage requirements vary widely by analyte.

Stability studies have shown that clinically significant analyte changes occur if serum or plasma remains in prolonged contact with blood cells. After separation from blood cells, analytes have the same stability in plasma and serum when stored under the same conditions. Glucose concentration in unseparated serum and plasma decreases rapidly in the first 24 hours and more slowly thereafter. This decrease is more pronounced in plasma. Two approaches have been used to minimize this effect. First, the serum or plasma may be rapidly separated from the red cells, or the specimen may be collected in a fluoride tube to inhibit glycolysis of the red blood cells, thereby stabilizing the glucose level during transport and storage. Fluoride has little effect on reducing glycolysis within the first hour of storage and may not reach complete inhibition until 4 hours of storage. One study has demonstrated a reduction in glucose concentration by 0.39 mmol/L in specimens collected in fluoride that are not immediately separated. These authors suggest that specimens collected in fluoride have a negative bias in blood glucose levels (Shi et al, 2009). Lactate levels increase, and a greater rise is seen in plasma than in serum. Chloride and total carbon dioxide (CO₂) show a steady decrease over 56 hours, with the degree of change more pronounced in plasma. K⁺ is reported to be stable for up to 24 hours, after which a rapid increase takes place. The degree of change is slightly more pronounced in plasma. Unseparated serum and plasma yield clinically significant increases in total bilirubin, sodium, urea, nitrogen, albumin, calcium, magnesium, and total protein. These changes are attributed to movement of water into cells after 24 hours, resulting in hemoconcentration (Boyanton & Blick, 2002). Other studies found potassium, phosphorus, and glucose to be the analytes that were least stable in serum not removed from the clot within 30 minutes. Albumin, bicarbonate, chloride, C-peptide, HDL-cholesterol, iron, LDL-cholesterol, and total protein were found to be unstable after 6 hours when the serum was not separated from the clot (Zhang et al, 1998).

When serum and plasma are not removed from the cells, lipids (such as cholesterol) and some enzymes increase over time, with the change more pronounced in plasma than in serum. LD activity continuously increases over 56 hours. AST, ALT, and CK were found to be stable over 56 hours. GGT activity in plasma, with and without prolonged contact with cells, was found to be 27% lower than in serum at 0.5 hours; however, plasma GGT activity steadily increases with prolonged exposure to cells. Creatinine can increase by 110% in plasma and by 60% in serum after 48 to 56 hours (Boyanton & Blick, 2002).

Serum and plasma may yield significantly different results for an analyte. For example, when serum and EDTA plasma results for parathyroid hormone (PTH) are compared from specimens frozen within 30 minutes of collection, EDTA plasma results are significantly higher (>19%) than those obtained from serum (Omar et al, 2001). The effect of freeze–thaw cycles on constituent stability is an important consideration. In plasma or serum specimens, the ice crystals formed cause shear effects that are disruptive to molecular structure(s), particularly to large protein molecules. Slow freezing allows larger crystals to form, causing more serious degradative effects. Thus quick freezing is recommended for optimal stability.

Importance of Policies and Procedures

It is essential to establish institution-specific phlebotomy policies and procedures that include personnel standards with qualifications; dress code and evaluation procedures; safety protocols, including immunization recommendations; universal precautions; needlestick and sharps information; personal protective equipment; test order procedures; patient identification; confidentiality and preparation; documentation of problems encountered during blood collection; needlestick site selection and areas to be avoided (mastectomy side, edematous area, burned/scarred areas, etc.); anticoagulants required and tube color; order of draw; special requirements for patient isolation units; and specimen transport. The laboratory should have available all CDC, College of American Pathologists (CAP), Clinical and Laboratory Standards Institute (CLSI), OSHA, and The Joint Commission (TJC) guidelines, as well as other government regulations pertaining to laboratory testing. All employees must be trained about safety procedures, and a written blood-borne pathogen exposure control plan must be available. See Chapter 1 for a more complete discussion of safety.

The OSHA Bloodborne Pathogens Standard concluded that the best practice for prevention of needlestick injury following phlebotomy is the use of a sharp with engineered sharps injury protection (SESIP) attached to the blood tube holder and immediate disposal of the entire unit after each patient's blood is drawn (OSHA, 2001). Information on exposure prevention can be found on the Exposure Prevention Information Network (EPINet), a database coordinated by the International Healthcare Worker Safety Center at the University of Virginia (http://www.healthsystem.virginia.edu/internet/epinet/). OSHA further mandates that employers make available closable, puncture-resistant, leak-proof sharps containers that are labeled and color-coded. The containers must have an opening that is large enough to accommodate disposal of the entire blood collection assembly (i.e., blood tube, holder, and needle). These containers must be easily accessible to employees in the immediate area of use, and if employees travel from one location to another (one patient room to another), they must be provided with a sharps container that is conveniently placed at each location/facility. Employers must maintain a sharps injury log to record percutaneous injuries from contaminated sharps, while at the same time protecting the confidentiality of the injured employee.

Arterial Puncture

Arterial punctures are technically more difficult to perform than venous punctures. Increased pressure in the arteries makes it more difficult to stop bleeding, with the undesired development of a hematoma. In order of preference, the radial, brachial, and femoral arteries can be selected.

Before blood is collected from the radial artery in the wrist, one should do a modified Allen test (Box 3-4) to determine whether the ulnar artery can provide collateral circulation to the hand after the radial artery puncture. The femoral artery is relatively large and easy to puncture, but one must be especially careful in older individuals because the femoral artery can bleed more than the radial or brachial. Because the bleeding site is hidden by bedcovers, it may not be noticed until bleeding is massive. The radial artery is more difficult to puncture, but complications occur less frequently. The major complications of arterial puncture include thrombosis, hemorrhage, and possible infection. When performed correctly, no significant complications are reported except for possible hematomas.

Unacceptable sites are those that are irritated, edematous, near a wound, or in an area of an arteriovenous (AV) shunt or fistula (McCall & Tankersley, 1993). Arterial spasm is a reflex constriction that restricts blood flow with possible severe consequences for circulation and tissue perfusion. Radial artery puncture can be painful and is associated with symptoms such as aching, throbbing, tenderness, sharp sensation, and cramping. At times, it may be impractical or impossible to obtain arterial blood from a patient for blood gas analysis. Under these circumstances, another source of blood can be used, with the recognition that arterial blood provides a more accurate result. Although venous blood is more readily obtained, it usually reflects the acid-base status of an extremity—not the body as a whole.

Finger or Heel Skin Puncture

For routine assays requiring small amounts of blood, skin puncture is a simple method by which to collect blood samples in pediatric patients. In the neonate, skin puncture of the heel is the preferred site to collect a blood sample; in older children, the finger is the preferred site. The large amount of blood required for repeated venipunctures may cause iatrogenic anemia, especially in premature infants. Venipuncture of deep veins in pediatric patients may rarely cause cardiac arrest, hemorrhage, venous thrombosis, reflex arteriospasm followed by gangrene of an extremity, damage to organs or tissues accidentally punctured, infection, and injury caused by restraining an infant or child during collection. Accessible veins in sick infants must be reserved exclusively for parenteral therapy. Skin puncture is useful in adults with extreme obesity, severe burns, and thrombotic tendencies, with point-of-care testing or with patients performing tests at home (blood glucose). Skin puncture is often preferred in geriatric patients because the skin is thinner and less elastic; thus a hematoma is more likely to occur from a venipuncture.

In newborns, skin puncture of the heel is frequently used to collect a sample for newborn screening tests for inherited metabolic disorders. A deep heel prick is made at the distal edge of the calcaneal protuberance following a 5- to 10-minute exposure period to prewarmed water. The best method for blood gas collection in the newborn remains the indwelling umbilical artery catheter. Box 3-6 lists the steps for a skin puncture (CLSI GP42-A6, 2008).

Central Venous Access Devices

Central venous access devices (CVADs) provide ready access to the patient's circulation, eliminating multiple phlebotomies, and are especially useful in critical care and surgical situations. Indwelling catheters are surgically inserted into the cephalic vein or into the internal jugular, subclavian, or femoral vein and can be used to draw blood, administer drugs or blood products, and provide total parenteral nutrition. Continuous, real-time, intraarterial monitoring of blood gases and acid-base status has been accomplished with fiberoptic channels containing fluorescent and absorbent chemical analytes (Smith et al, 1992).

Special Urine Collection Techniques

Catheterization of the urethra and bladder may cause infection but is necessary in some patients (e.g., for urine collection when patients are unable to void or control micturition). Ureteral catheters can also be inserted via a cystoscope into the ureter. Bladder urine is collected first, followed by a bladder washing. Ureteral urine specimens are useful in differentiating bladder from kidney infection or for differential ureteral analysis, and may be obtained separately from each kidney pelvis (labeled left and right). First morning urine is optimal for cytologic examination.

Urine Storage and Preservation

Preservation of a urine specimen is essential to maintain its integrity. Unpreserved urine specimens are subject both to microbiologic decomposition and to inherent chemical changes. Table 3-6 lists common changes that occur as urine decomposes. To prevent growth of microbes, the specimen should be refrigerated promptly after collection and, when necessary, should contain the indicated chemical preservative. For some determinations, the addition of a chemical preservative may be best to maintain analytes when performing 24-hour urine collections. If a preservative is added to the empty collection bottle, particularly if acid preservatives are used, a warning label is placed on the bottle. The concentrated acid adds a risk of potential chemical burns; the patient should be warned about this potential danger, and the container labeled accordingly. In this scenario, the clinician must assess the patient's risk of exposure to the preservative; therefore, refrigeration may be more appropriate, and the preservative may be added upon submission to the laboratory. Light-sensitive compounds, such as bilirubin, are protected in amber plastic bottles. Precipitation of calcium and phosphates occurs unless the urine is acidified adequately before analysis.

It is particularly important to use freshly voided and concentrated urine to identify casts and red and white blood cells, as these undergo decomposition upon storage at room temperature or with decreased concentration (<1.015 specific gravity). They disappear rapidly in hypotonic and alkaline urine. Bilirubin and urobilinogen decrease, especially after exposure to light. Glucose and ketones may be consumed, while bacterial contamination and loss of CO₂ lead to increase of pH, formation of turbidity with precipitates, and change in color. Ideally specimens should be delivered to the laboratory and analyzed within 1 hour of collection.

Urine may be frozen in aliquots to be assayed at a later date for chemical analysis only. When repeat testing is expected, the specimen should be stored in multiple aliquots to circumvent specimen degradation as a result of repeated freeze–thawing of a single specimen. Preservatives may also be added, depending on the substance to be tested. Sodium fluoride can be added to 24-hour urine for glucose determinations to inhibit bacterial growth and cell glycolysis but not growth of yeast. About 0.5 g of sodium fluoride is added to a 3- to 4-L container. Sodium fluoride may inhibit reagent (enzyme-embedded) glucose strip tests. Tablets containing for­maldehyde, mercury, and benzoate (95-mg tablet/20 mL urine) have also been used; however these preservatives elevate specific gravity (0.002/1 tablet/20 mL). Boric acid in a concentration of 1 g/dL preserves urine elements such as estriol and estrogen for up to 7 days. Boric acid maintains the pH at about 6.0 and preserves protein and formed elements well without interfering with routine testing except for pH. Boric acid is a bacteriostatic preservative, not a bactericidal, and it does not inhibit the growth of yeasts. Boric acid has been reported to interfere with drug and hormone analysis (Strasinger & DiLorenzo, 2001). For catecholamines, vanillylmandelic acid (VMA), or 5-hydroxyindoleacetic acid (5-HIAA) collections, 10 mL of 6N HCl is added to a 3- to 4-L container. The HCl establishes a pH of approximately less than 3.0, which is good for chemical testing. However, the low pH destroys formed elements and enhances uric acid precipitation. Table 3-7 lists preservatives commonly used for 24-hour urine specimens. The CLSI-approved guidelines for urinalysis and collection, transportation, and preservation of urine specimens provide useful information on various preservatives recommended for 24-hour urine collections (CLSI, 2009).

Lumbar punctures (LPs) are performed to collect cerebrospinal fluid (CSF) for laboratory evaluation to establish a diagnosis of infection (bacterial, fungal, mycobacterial, or amebic meningitis), malignancy, subarachnoid hemorrhage, multiple sclerosis, or demyelinating disorders. The most common site for lumbar puncture is between the third and fourth lumbar vertebrae or between the fourth and fifth lumbar vertebrae. A serious complication of an LP is cerebellar tonsillar herniation in patients with elevated intracranial pressure, and it should be avoided unless CSF findings are expected to improve treatment or outcome. Patients with spinal cord tumors with paresis may progress to paralysis following LP. Patients with sepsis in the lumbar region (skin infection, cellulitis, or epidural abscess) should not have an LP performed, to avoid introduction of infection. Other complications of LP include asphyxiation in infants due to hyperextending the head forward, thus occluding the trachea, paresthesia, headache, and, rarely, hematomas. CSF is also collected by cisternal puncture. A needle is inserted into the cisternal subarachnoidea, or small space, that serves as a reservoir for CSF between the atlas and the occipital bone in the back of the head or by lateral cervical puncture (Kjeldsberg & Knight, 1993). Specimens can also be collected from ventricular cannulas (shunts) when present.

Before CSF is collected, the pressure should be between 90 and 180 mm Hg; this is measured by allowing fluid to rise in a sterile, graduated manometer. Holding one's breath, abdominal compression, congestive heart failure, inflammation of the meninges, obstruction of intracranial venous sinuses, mass lesions, or cerebral edema may cause the pressure to be elevated (>180 mm Hg). When pressure is normal, 20 mL of specimen may be removed. On closing, the pressure should be between 10 and 30 mm Hg. A marked decrease in pressure following this procedure suggests cerebellar herniation or spinal cord compression; thus no additional CSF should be collected. Patients with partial or complete spinal block may have low pressure (<80 mm Hg), falling to zero after removal of only 1 mL. Again, no additional fluid should be removed. No more than 2 mL can be removed when the pressure is greater than 200 mm Hg. Three aliquots are generally collected in separate, sterile tubes labeled appropriately with name, date, and sequential tube collection number and distributed. Hospital policies differ as to which tube is distributed to which laboratory for analysis. It is generally recommended that Tube #1 goes to chemistry for glucose and protein analysis or to immunology/serology; Tube #2 goes to microbiology for culture and Gram stain; and Tube #3 goes to hematology for cell counts. Tube #3 is the least likely to be contaminated by a bloody tap at collection. Expansion of tests ordered on CSF to include molecular diagnostic procedures has placed even greater demand on proper utilization of the entire volume of CSF collected, with special efforts to conserve specimens in each laboratory section so that additional tests may be performed.

Precentrifugation Phase

Ideally, all measurements should be performed within 45 minutes to 1 hour after collection. If this is not practical, the specimen should be processed to a point at which it can be properly stored to preclude alteration of constituents to be measured. With the exception of blood gases and ammonia determinations, plasma or serum is preferred for most biochemical determinations. In clinical chemistry, serum and plasma are interchangeable except for a few measurements. Serum is required for protein electrophoresis and immunofixation assays, just as plasma is necessary for fibrinogen and other coagulation measurements. Serum is most commonly the specimen of choice, owing to its simplicity in collection and handling. Additionally, interference from anticoagulants is obviated. Plasma may be used in medical emergencies because samples do not have to clot before centrifugation. Usually, a greater volume of plasma than serum is obtained from a given volume of whole blood owing to the clot formation process. Hospitalized patients are likely to be receiving heparin (especially under critical care), which can delay clotting in blood collection tubes even with activators and lead to fibrin strands that can clog up aspiration probes on instrumentation. Accordingly, it is good practice to collect blood from hospitalized patients into tubes with heparin anticoagulant to obtain plasma for chemical tests.

Blood should be stoppered in the original container until ready for separation. For plasma preparation, centrifuge blood within 1 hour after collection, for 10 minutes at a relative centrifugal force (RCF) of 850 to 1000× gravity (g), keeping the container stoppered to prevent evaporation of plasma or serum water. Adequate time for clotting must be allowed to prevent latent fibrin formation, which may cause undesirable clogging of automated chemistry analyzers. Loosening the clot by “trimming” or “ringing” the tube may cause some hemolysis and should be avoided. When glass tubes are used, they should be centrifuged in an aerosol-contained vessel. Serum or plasma must be stored at 4° C to 6° C if analysis is to be delayed for longer than 4 hours. One study suggests that this may not be necessary (Melanson et al, 2004). Many laboratories store samples for 7 days in case a test is added.

Centrifugation Phase

A centrifuge uses centrifugal force to separate phases of suspensions by different densities. It is most frequently used in processing blood to derive plasma or serum fractions. Urine and other body fluids may be centrifuged to concentrate particulate matter as sediment to be examined and to minimize interference with other determinations from the same material. Conditions for centrifugation should specify both the time and the centrifugal force. In selecting a centrifuge, one should look for the highest possible centrifugal force and not the rotational speed. The RCF in g units—that is, multiples of the gravitational force—may be calculated by using the following formula:

RCF=1.118×10−5×r×(rpm)2

where 1.118 × 10⁻⁵ is a constant; r is the radius, expressed in centimeters, between the axis of rotation and the center of the centrifuge tube; and rpm is the speed in revolutions per minute. The RCF can also be obtained from a nomogram that gives the RCF without the need to calculate it from the previous formula.

Several principles must be observed to avoid damage to the centrifuge, or the specimen, and danger to personnel. Tubes, carriers, or shields of equal weight, shape, and size should be placed in opposing positions in the centrifuge head to achieve appropriate balance. Tubes must be balanced across the center of rotation, and each bucket must be balanced with respect to its pivotal axis (Seamonds & Elizabeth, 2001). Specimens must be placed with regard for a geometrically symmetrical arrangement, using water-filled tubes to attain balance.

Recentrifugation of gel separator tubes has been associated with pseudohyperkalemia. One study demonstrated that after initial centrifugation, a new serum layer develops under the gel within the cellular layer. During storage, potassium leaks from the cellular layer into the new serum layer, creating hyperkalemia in this layer. When the tube is recentrifuged, the new serum layer will move above the gel layer and cause a pseudohyperkalemia in the serum for analysis. The same authors also demonstrated that a pseudonormokalemia in patients with true hypokalemia may be erroneously reported after recentrifugation (Hira et al, 2001; 2004).