The Effects of in Vitro Hemolysis on the Comprehensive

Home , Bilirubin, Hemosiderin, Serum chloride

T H E E F F E C T S O F I N V I T R O H E M O L Y S I S O N T H E C O M P R E H E N S I V E M E T A B O L I C P A N E L

A Report of Senior Thesis by Brent Thomas Carl Robinson

Maryville College Fall, 2002

Date Approved , by Faculty Supervisor

, by Editor

ii ABSTRACT

Hemolysis is a common problem in clinical chemistry analyses because it interferes in the determination of several chemical parameters. Using fully mechanized autoanalyzers, the effects of hemolysis in the determination the 16 clinical blood chemistry parameters that are included in the comprehensive metabolic panel were determined quantitatively by adding hemolysate to serum. Free hemoglobin concentrations up to 800 mg/dL did not significantly affect the following analytes: chloride ion, sodium ion, calcium ion, blood urea nitrogen, creatinine, alkaline phosphatase, conjugated bilirubin, and carbon dioxide. However, hemolysis at 800 mg/dL of free hemoglobin and higher caused erroneously high values for: potassium ion, glucose, total bilirubin, unconjugated bilirubin, total protein, and albumin. Hemolysis at 400 mg/dL of free hemoglobin and higher caused significantly high values in aspartate aminotransferase, and significantly low values in alanine aminotransferase. These results differ subtly from those of other studies; therefore, it is extremely important for each laboratory to know how hemolysis affects its chemical analyses because

iii the effects vary among methodologies. After such determination, hemolysis should be accounted for when interpreting results and in diagnosis.

TABLE OF CONTENTS

Page

Chapter

I Introduction ...... 1 Components of the Blood ...... 1 Clinical Chemistry ...... 16 Hemolysis ...... 23 Purpose and Hypothesis ...... 27

II Materials and Methods ...... 28

III Results ...... 32

IV Discussion ...... 38 Potassium ...... 40 Glucose ...... 41 Aspartate Aminotransferase ...... 42 Total, Unconjugated, and Conjugated Bilirubin ...... 43 Total Protein ...... 44 Albumin ...... 44 Alanine Aminotransferase ...... 45 Urea Nitrogen, Calcium, Sodium, Creatinine, Carbon Dioxide, Alkaline Phosphatase, and Chloride ...... 46 MLT Qualitative Grades ...... 47 Conclusions ...... 50

Appendix ...... 53

References ...... 63

v LIST OF TABLES

Table Page

1 Functions of Formed Elements of the Blood...... 2

2 Normal Values of the Individual Determinations within the Comprehensive Metabolic Panel in Adults...... 18

3 The Comprehensive Metabolic Panel Analyses and Methodology in the Ortho-Diagnostics Vitros 950...... 31

4 Intracellular Hemoglobin Concentrations of the 5 Participants...... 32

5 MLT Visual Hemolysis Grades vs. Actual Grades...... 33

6 Mean Results (± 1 SE) of the Comprehensive Metabolic Panels for Each Degree of Hemolysis ...... 35

vi ACKNOWLEDGEMENTS

I would like to thank the following employees of the Blount Memorial Hospital Laboratory who helped in performing the analyses: Mary Corbitt, George Easton, Karen Easton, and James Houser. I would also like to thank the following employees of the Blount Memorial Hospital Laboratory who were participants in the study: Nathena Bruce, Karen Easton, Kim Bailey, Donna Sweitzer, and Kris Loggins. Thank you to John Bleazey, Blount Memorial Hospital Laboratory Manager, for authorizing the conductance and financing of the study. In addition, I would like to thank Dr. Drew Crain for his time and effort in supervising my study and for providing a very structured agenda that was helpful in completing the study. Finally, I would like to thank my wife Michelle and my son Brock for tolerating my diverted attention from them during the course of this study.

vii

9 CHAPTER I

INTRODUCTION

Components of Blood Blood is composed of plasma and formed elements. Plasma is the protein-rich, acellular fluid in the blood. The formed elements are erythrocytes, also called red-blood cells, leukocytes, also termed white- blood cells, and platelets. Functions of the formed elements are summarized in Table 1. Around 55% of blood is plasma, and about 45% is red blood cells; platelets and white blood cells make up only about 1% of the blood (Vick, 1984). When blood is centrifuged, the more dense formed elements go to the bottom and the less dense plasma sits on the top. The percentage of formed elements in the blood is called the hematocrit (Fox, 1999). On average adult humans have just over ten pints of blood. The bright red blood flowing from the heart to the capillaries via the arteries is called arterial blood. It is oxygen enriched which yields the bright red color. After the blood has been oxygen depleted

1 Table 1. Functions of Formed Elements of the Blood. Component Function Erythrocyte Transports oxygen and carbon dioxide Leukocytes Granulocytes 1. Neutrophil Phagocytic 2. Eosinophil Helps detoxify foreign substances; secretes enzymes that dissolve clots; fights parasitic infections 3. Basophil Releases anticoagulant heparin Agranulocytes 1. Monocyte Phagocytic 2. Lymphocyte Provides specific immune response, including antibodies Platelet Enables clotting, releases serotonin causing vasoconstriction Source: (Fox, 1999, p. 368). in the capillaries, it is returned to the heart by the veins. This darker blood is termed venous blood.

2 Erythrocytes Erythrocytes are shaped like flattened biconcave discs (Fox, 1999). This unusual shape increases substrate surface area and aids in oxygen binding, which allows oxygen transport throughout the body. The main function of the erythrocytes is to circulate oxygen, which is bound to hemoglobin in the lungs and transported to the rest of the cells in the body. Another function of hemoglobin is that it is an excellent acid-base buffer, which makes the erythrocytes accountable for roughly 70% of the buffering capacity of the blood. Additionally, red blood cells contain a large amount of carbonic anhydrase, which acts as a catalyst in the reaction of carbon dioxide and water (Guyton & Hall, 2000). The catalysis of this reaction allows blood to react with vast amounts of carbon dioxide and, in turn, transport it back to the lungs. Red blood cells, although having a biconcave disk-like shape, are very amoeboid in nature. They can change shape dramatically when flowing through the capillaries (Guyton & Hall, 2000). The reason the shape of the erythrocytes is so adaptable is that the cell membrane is very excessive compared to the material inside the cell, so the cell is not filled to capacity. This allows the erythrocyte to be plastic.

3 Mature erythrocytes are also anucleated and lack an endoplasmic reticulum. Erythrocyte production In the early fetal months, red blood cells are produced in the yolk sac (Vick, 1984). As the yolk sac recedes, the liver, spleen, and lymph nodes assume the responsibility of erythrocyte production. By the fifth month, the bone marrow begins to produce red cells, and by the time of birth, the bone marrow exclusively produces the erythrocytes (Guyton & Hall, 2000). Early in life all of the bones produce red blood cells, but as they are not needed, the longer bones become infiltrated with fat. By age twenty, only the bones of the axial skeleton produce erythrocytes. The bone marrow contains hematopoetic, or blood-forming, tissue. These tissues are separated by venous sinuses that drain into a central vein. In addition, arterial blood flows through the tissue and drains into the central vein via the venous sinuses (Vick). The bone marrow contains primordial stem cells that can lead to the formation of many types of cells. These are responsible for producing the hemocytoblasts. The genesis of red blood cells begins with the hemocytoblast. The hemocytoblast forms the basophil erythroblast which begins to produce hemoglobin (Guyton & Hall, 2000). The basophil

4 erythroblast gives rise to the polychromatophil erythroblast, and as the nucleus becomes smaller while even more hemoglobin is being produced, the cell becomes a normoblast. After the concentration of hemoglobin in the cell reaches its maximum, the normoblast then completely loses its nucleus, while at the same time the endoplasmic reticulum is shrinking. At this point the cell is an immature red blood cell, or a reticulocyte. The reticulocytes make their way into the capillaries by moving through the pores in the membrane. After the reticulocyte passes into the bloodstream, it continues to make hemoglobin for a couple of more days until the endoplasmic reticulum is completely reabsorbed (Guyton & Hall). At this point, the cell is now a mature red blood cell, or erythrocyte. Once mature, the erythrocytes circulate approximately 120 days before destruction, and red cells are synthesized at a rate higher than any other tissue in the body (Vick, 1984). There are two essential nutrients that aid in erythrocyte genesis: vitamin B12 and folic acid. Both of these vitamins are needed for DNA synthesis; thus, deficiency in these vitamins causes failure of nuclear maturation and division. Without this ability, red cells cannot be produced efficiently. As a result of this deficiency, erythrocytes become macrocytes, which are large, oval, and have fragile cell membranes

5 (Guyton & Hall, 2000). This fragility does not affect the oxygen carrying capability of the cell, but it does result in shorter life span. Erythrocyte regulation Red blood cell production is regulated by a stimulus-response mechanism by which when there is a decrease in the delivery of oxygen to the cells, more red cells are produced. This hypoxia could be the result of several circumstances such as heavy bleeding, being in an oxygen-poor environment, or destruction of bone marrow. Erythrocyte regulation is directed by the glycoprotein hormone erythropoietin (Vick, 1984). When placed in any of the aforementioned circumstances, erythropoietin is formed promptly. Erythropoietin causes the bone marrow to begin producing erythrocytes. Erythropoietin stimulates the production of proerythroblasts from the hemopoietic stem cells in the bone marrow, and erythropoietin also causes these cells to pass through the stages of maturation more quickly. Erythropoietin is produced mainly by the kidney; however, the exact area of the kidney where it is formed is unknown (Guyton & Hall, 2000). Erythrocyte destruction Red blood cells contain enzymes capable of producing some adenosine triphosphate (ATP), which keeps the integrity of the cells in check. The ATP

6 maintains pliability of the membrane, protects proteins, allows transport of ions, and keeps the iron in the hemoglobin in the ferrous form, which allows it to bind to oxygen (Guyton & Hall, 2000). As the erythrocytes age, these processes cease to work, and the cell becomes very delicate. Once this fragility is reached, the cell will rupture either in normal circulation, or while passing through the spleen. The remaining fragments are engulfed by macrophages and reused (Vick, 1984). Hemoglobin Hemoglobin is a globular protein that has a molecular weight of 64,500 daltons. Hemoglobin is composed of two major parts: the heme molecule, which contains iron, and 4 polypeptide chains collectively called globin (Vick, 1984). The maximum concentration of hemoglobin in erythrocytes is near 34 g per 100 ml, and this concentration does not exceed this amount because the red blood cells’ metabolic limit of hemoglobin is met (Guyton & Hall, 2000). When the pre-erythrocyte cells are deficiently producing hemoglobin, it is reflected in low hemoglobin concentrations of the mature erythrocytes. Normal values of hemoglobin range from approximately 12 to 18 g per 100 ml depending on gender and other factors (Fox, 1999). In addition, when hemoglobin combines

7 with oxygen, approximately 20 ml of oxygen can be transported by each 100 ml of blood (Guyton & Hall). Hemogobin production begins during the erythrocyte maturation process at the erythroblast stage. The production of hemoglobin is continued through the reticulocyte stage, even while the reticulocytes have passed through the capillaries into circulation. Hemoglobin can not be formed in the mature erythrocyte as they are enucleated. The main importance of hemoglobin is that it bonds loosely with oxygen. This loose bond is significant because it allows the oxygen to be transported, and the bond is easily broken. Another important feature of hemoglobin is that it carries oxygen in the molecular form and not as an ion. The human body can utilize only molecular oxygen. Each molecule of hemoglobin has four iron atoms each of which is capable of bonding with oxygen, so one hemoglobin molecule can carry four oxygen molecules. The formation of hemoglobin is assisted by several other substances which include copper, pyroxidine, cobalt, and nickel (Guyton & Hall, 2000). These substances catalyze the synthesis of hemoglobin. Iron is essential in hemoglobin formation. Iron is initially absorbed in the small intestine and combines with a transport protein called transferrin, which carries it through the plasma. Transferrin can

8 release the iron into any body tissue. Although iron is stored in all body cells, the liver is responsible for the majority of iron storage. In the liver, iron combines with a protein called apoferritin, becoming ferritin (Guyton & Hall, 2000). This is storage iron. When the liver becomes saturated with ferritin, excess iron is then stored in the body cells in the form called hemosiderin. Daily absorption of iron has to at least equal daily loss of iron. Average daily iron loss of a male is 0.6 mg, while female average daily loss is about double that as a result of menstruation. Iron absorption works on a feedback mechanism. When no more iron can be converted to ferritin, iron can no longer be released into the tissues. When iron is not allowed into the tissue, the transferrin responsible for carrying the iron becomes saturated. Because of this, transferrin will no longer accept iron from the mucosal cells of the duodenum (Guyton & Hall, 2000). White Blood Cells White blood cells, or leukocytes, are different from erythrocytes in several ways, including their function, and in the fact that they have nuclei and mitochondria (Fox, 1999). White blood cells are part of the human immune system, and they have the capability to migrate through the capillaries to a site of infection. There are six different types of

9 white blood cells: neutrophils, eosinophils, basophils, monocytes, lymphocytes, and plasma cells. Plasma cells, which are derived from lymphocytes, create large numbers of antibodies. The eosinophils, basophils, and neutrophils are collectively called granulocytes because they have a granular appearance (Guyton & Hall, 2000). The neutrophils and the monocytes are responsible for destroying foreign bacteria and viruses. The neutrophils are mature cells that destroy foreign bodies even in the blood. However, monocytes are immature cells that cannot fight foreigners in the blood. Instead they enter the tissues and begin to swell becoming macrophages. Both neutrophils and monocytes move by amoeboid motion through the pores of the blood vessels. This process is called diapedesis (Guyton & Hall, 2000). A process called chemotaxis, in which certain chemicals are produced from the area of the invaders, causes the monocytes and neutrophils to move toward the infection. Once the monocytes and neutrophils migrate through the capillaries to the site of infection, they phagocytize, or engulf, the invaders. In order to phagocytize only the invaders, this process must be very selective. There are three factors that determine what will be phagocytized: roughness of the particle, electropositive charge of

10 the particle, and presence of an antibody-antibody complex (Guyton & Hall, 2000). If these factors are present, phagocytosis is more likely to occur. When a neutrophil phagocytizes a particle, it projects pseudopodia around the particle, and then invaginates the particle into the cytoplasm. A neutrophil can digest only 5 to 20 bacteria. Monocytes, which become macrocytes by swelling to a very large size, can phagocytize many more particles than can neutrophils. After the particles have been phagocytized, they are destroyed by digestive enzymes from the lysosomes within the neutrophil or macrophage itself (Guyton & Hall, 2000). Eosinophils are not very important in fighting infection. They are poor phagocytes; however, they do aid in digesting antigen-antibody complexes. Also, eosinophils are thought to clean up remaining elements near the site of infection. Finally, eosinophils aid in blood clot dissolution (Guyton & Hall, 2000). Basophils transport heparin into the blood, which prevents clotting and cleans out fat particles. Basophils also function in allergic reactions. A specific antibody attaches to basophils, and when the antibody encounters the antigen, the basophil ruptures. This rupture releases histamine, bradykinin, and seratonin which causes tissue reactions (Guyton & Hall, 2000).

11 Lymphocytes function in two types of immunity: cell-mediated and humoral immunity. There are two types of lymphocytes, T lymphocytes and B lymphocytes. In cell-mediated immunity, T lymphocytes are produced. When faced with an antigen, the T lymphocytes become immunoblasts called killer T cells that attach to and destroy the invader. In humoral immunity, B lymphocytes are produced. When the B lymphocytes encounter an antigen, they too become immunoblasts, which are larger than normal lymphocytes, called B cells. These B cells then become plasma cells. Plasma cells are much larger than B cells and have more cytoplasm, allowing them to house the mechanisms responsible for antibody production. Once the antibodies are produced, they combine to antigens, acting as a marker so the foreign substance will be phagocytized. Also, the antigen-antibody complex triggers the complement system, which is composed of proteins, that destroys the invaders (Vick, 1984). Leukocyte production The genesis of the different types of leukocytes occurs in separate places in the body. The monocytes and granulocytes are produced in the bone marrow, while the lymphocytes and plasma cells are produced in the lymphogenous organs (Guyton & Hall, 2000). The genesis of leukocytes is called leucopoiesis. Mature leukocytes evolve from myeoblasts, lymphoblasts, and

12 monoblasts. Myeoblasts give rise to the granulocytes, lymphoblasts give rise to the lymphocytes, and monoblasts give rise to the monocytes. There are several chemicals called cytokines that determine which kind of leukocyte will be produced. Cells in the immune system secrete the specific cytokine that will stimulate the type of leukocyte needed (Fox, 1999). Leukocyte life span The life span of white blood cells is short relative to the erythrocyte life span, because the function of the white cells is simply to leave the bone marrow or lymphoid tissue and go to the infection. The life span of a granulocyte is about three days after being released from the bone marrow. Monocytes have a much longer life because they spend little time in circulation, but when they reach the site of infection they become macrophages and can exist in this state for months. Lymphocytes have a life span of about 100 to 300 days because although they spend only a few hours at a time in circulation, they eventually cycle back to the lymph and are used repeatedly (Guyton & Hall, 2000). The destruction of phagocytic leukocytes results from the digestive enzymes released from the lysosome eventually leading to their own death.

13 Platelets and Clotting There are three main avenues by which blood loss is prevented: vasoconstriction, formation of the platelet plug, and clotting (Guyton & Hall, 2000). The bodily process of preventing blood loss is called hemostasis. When vascular trauma occurs, the veins constrict as a response to pain, which slows blood flow to the traumatized area. Blunt trauma, in which a great deal of damage occurs, results in greater vasoconstriction than would result from trauma by sharp objects (Vick, 1984). The second step in hemostasis is plugging the damaged area with platelets. Platelets, or thrombocytes, are small, disk-shaped fractions of megakaryocytes, which are formed in the bone marrow (Guyton & Hall, 2000). Platelets are anucleated, and exhibit amoeboid motion. Platelets circulate for a maximum of about 9 days, and then they are destroyed by the liver and spleen. The function of platelets relies on their adhesive ability. When platelets come in contact with a damaged vessel, they enlarge and develop sticky protrusions that adhere to the damaged area. After the platelets adhere to the damaged area, phospholipids in the platelet cell membrane activate clotting factors that are in the plasma (Fox, 1999). In addition, platelets release adenosine diphosphate (ADP) and serotonin. They attract more platelets and

14 cause the aforementioned vasoconstriction. When enough platelets adhere to each other, a plug is formed preventing further blood loss. The final way that blood loss is prevented is through the coagulation process. Blood plasma contains many proteins, some of which have clotting functions. They are called clotting factors. In the clotting process resulting from tissue damage, first tissue thromboplastin is released. Thromboplastin activates a clotting factor known as proconvertin. Proconvertin forms a complex that with tissue thromboplastin, calcium, and the phospholipids from the cell membranes of the platelets. This in turn activates yet another clotting factor, clotting factor X, which forms a complex with calcium, phospholipids, and clotting factor V. Finally, this converts prothrombin to thrombin, which as a result converts fibrinogen to fibrin which forms the clot. This is called the extrinsic pathway because the tissue thromboplastin, which activates the process, is not part of the blood. There is also another pathway for clotting known as the intrinsic pathway, which occurs in a test tube. Contact with a negatively charged surface activates this pathway (Fox, 1999). Plasma and Serum Plasma is the fluid portion of the blood that transports ions, such as potassium and sodium, and

15 many organic molecules, such as glucose, lipids, and proteins. Plasma is made up of between 7% and 9% protein. Albumins, globulins, and fibrinogen are the main types of proteins found in the plasma. Albumins are manufactured by the liver. They produce the osmotic pressure needed to maintain blood volume. This pressure acts on the interstitial fluid between cells either pulling from the fluid to increase blood volume, or adding to the fluid in order to decrease blood volume. Alpha globulins, beta globulins, and gamma globulins are the different globulins found in plasma. Alpha and beta globulins are generated by the liver and transport lipids and fat-soluble vitamins. Gamma globulins have immunity functions and are created by the lymphocytes. Finally, plasma contains fibrinogen, which is a clotting factor. When blood clots, fibrinogen, as well as other clotting factors are transformed and are contained in the clot itself (Fox, 1999). As a result, when blood is allowed to clot, the remaining fluid lacks these clotting factors. This fluid is called serum. Often, plasma and serum are incorrectly considered synonyms. Clinical Chemistry Many biochemical tests are done on plasma and serum for diagnostic purposes. The reason for this is that many constituents, such as glucose or sodium ions, are dispersed less uniformly in erythrocytes

16 compared to plasma or serum (Henry, 1991). Serum analysis is usually the method of choice because no anticoagulant is required. In the past, after blood was collected and centrifuged, the serum was not allowed to sit on the red cells for longer than two hours because some analytes were skewed as a result (Tietz, 1976). Now, with the advent of the serum separator tube, which has a silicone gel that moves between the serum and the cells after being centrifuged, analyses may be conducted longer than two hours after collection. Except for blood gases, venipuncture is usually the preferred method of collection because it usually requires less skill than arterial puncture, and risk of hematoma is considerably less. In serum determinations, usually a glass or plastic test tube is used for collection, and sometimes a clot activator is added to accelerate result turnout. There are many clinical chemistry tests. Tests commonly ordered by physicians are certain liver enzymes, kidney function tests, and basic metabolites. The comprehensive metabolic panel, abbreviated “CMPN” includes all three of these tests. Each of these tests has a specific diagnostic significance, so accurate results are very important. Normal values for the comprehensive metabolic panel can be seen in Table 2. The individual determinations included in

17 Table 2. Normal Values of the Individual Determinations within the Comprehensive Metabolic Panel in Adults. Analyte Normal Valuea Albumin 3.3-5.0 g/100 ml Bilirubin, total 0.2-1.0 mg/100 ml Calcium (Ca++) 8.5-10.4 mg/100 ml Carbon Dioxide, total 23-30 mmol/l Chloride (Cl-) 98-106 mmol/l Creatinine Males: 0.9-1.5 mg/100ml Females: 0.8-1.2 Glucose 70-105 mg/100 ml Alkaline Phosphatase 3.5-13 units/100 ml Potassium (K+) 3.5-5.3 mmol/l Protein, total 6.0-8.5 g/100 ml Sodium (Na+) 135-148 mmol/l BUN (urea nitrogen) 7-18 mg/100 ml AST Males: 7-21 U/l Females: 6-18 U/l ALT Males: 6-21 U/l Females: 4-17 U/l aSome of these values are dependent on specific specimen conditions such as temperature and methodology used. Source: (Tietz, 1976, pp. 1206-1226)

18 the CMPN are described hereafter. Albumin is a protein that is found in the plasma. Albumin has many functions. They include the transport of large organic anions normally insoluble in aqueous fluids (in particular, long-chain fatty acids and bilirubin), the binding of toxic heavy metal ions, the transport of poorly soluble hormones such as cortisol, aldosterone, and thyroxine when the capacities of their more specific binding proteins are exceeded, the maintenance of plasma colloidal osmotic pressure, and the provision of a reserve stored of protein. Albumin levels can be depressed for several reasons. A barrier to albumin synthesis such as malnutrition or liver disease can result in low albumin. Also, losing protein in urine and feces can result in low albumin. Many times during infection albumin levels are lowered, as well as in carcinomatosis and congestive heart failure. Proteinuria causes the lowest serum albumin concentrations. Edema can result from low albumin (Tietz, 1976). Bilirubin, the bile pigment, is derived from hemoglobin that has been released from destroyed or damaged erythrocytes. Bilirubin binds to albumin and is transported through the blood. Increased bilirubin causes jaundice. This can result from decreased liver function, or common bile duct obstruction (Tietz, 1976). Hepatitis and cirrhosis are common liver

19 diseases that result in jaundice because the damaged cells can no longer remove and metabolize bilirubin. Calcium is a mineral that is found mostly in the bones; however, it is also found in blood plasma. Calcium has several functions including skeletal mineralization, blood coagulation, neuromuscular conduction, excitability of muscle, glandular synthesis, and preserving cell membrane integrity. It is regulated by the parathyroid gland. Increased serum calcium can result from hyperparathyroidism, hypervitaminosis D, and multiple myeloma. Low serum calcium can result from hypoparathyroidism, steatorrhea, nephrosis, nephritis, and pancreatitis (Tietz, 1976). Carbon dioxide is dissolved in blood as a result of respiration. Knowledge of the carbon dioxide level, along with other chemistry levels, allows one to evaluate the acid-base status of the blood. High total carbon dioxide can be seen in compensated respiratory acidosis, or in metabolic alkalosis. Low carbon dioxide can be seen in compensated respiratory alkalosis, or metabolic acidosis (Tietz, 1976). Chloride is the most abundant extracellular anion. It functions in water distribution, osmotic pressure, and in ion balance in the extracellular fluid. Low serum chloride may result from salt-losing nephritis, diabetic acidosis, renal failure, and

20 prolonged vomiting. High chloride can be seen in dehydration and congestive heart failure (Tietz, 1976). Creatinine is a nitrogenous compound that is a waste product derived from creatine, a muscular energy compound excreted by the kidneys. As a result, when the kidneys are not functioning properly, there is an increase in serum creatinine. This can result from kidney damage or disease. Creatinine values are usually used in concordance with urea nitrogen levels in order to diagnose renal dysfunction (Tietz, 1976). Glucose is a carbohydrate that is the energy source of the body. Irregular glucose can result from diabetes, hyperglycemia, and hypoglycemia. Some other long-term effects of diabetes can include hyperlipidemia, artherosclerosis, and kidney disease. Alkaline phosphatase is an enzyme found in all tissues of the body. It facilitates the movements of analytes across membranes, aids in lipid transport, and aids in the calcification of bone. Increased alkaline phosphatase can indicate hepatobiliary disease, bone disease, and hyperparathyroidism (Tietz, 1976). Potassium is the major intracellular cation. Low potassium causes excitatory changes in cardiac muscle function, which is reflected in electrocardiographs. Low serum potassium results from alkalosis, or

21 potassium deficiency from inadequate intake or from loss through prolonged vomiting or diarrhea. High potassium can be observed in oliguria, anuria, renal dysfunction or failure, and acidosis (Tietz, 1976). Total protein is a measure of all of the proteins found in the plasma. There are several types of proteins present in the plasma with an array of functions including regulating osmotic pressure, blood clotting, and acid-base balance. High total protein can be observed in multiple myeloma and in dehydration, because less water is present resulting in higher protein concentration. Low serum protein can result from a dysfunction in intestinal absorption, kidney damage from nephrotic syndrome, and prolonged bleeding (Tietz, 1976). Sodium is the major extracellular cation. It functions in maintaining osmotic pressure in fluid compartments. Low sodium levels are sometimes attributed to severe polyuria, diabetic acidosis, Addison’s disease, and severe diarrhea. High sodium concentrations may be attributed to hyperadrenalism, dehydration resulting in higher concentration of sodium, and specific brain injuries (Tietz, 1976). Urea nitrogen, along with creatinine, is chiefly used as a means to evaluate kidney function. Urea is produced by the liver from ammonia present as a result of deamination of amino acids. Urea nitrogen is

22 usually written as blood urea nitrogen, abbreviated BUN. Among causes for elevated BUN are cardiac compensation, dehydration, high protein intake, chronic nephritis, and urinary tract obstruction (Tietz, 1976). Aspartate aminotransferase, AST or SGPT, and alanine aminotransferase, ALT or SGOT, are enzymes found throughout the body; however, they are often called liver enzymes because they are used to evaluate liver function. A significant amount of AST is found in the heart and in skeletal muscle. Consequently, AST is increased after heart attacks and in muscular dystrophy. Although in liver disease both enzymes are elevated, ALT is a more liver specific enzyme. The concentration of ALT in the liver is higher than any other area of the body. Examples of liver diseases affecting ALT and AST are hepatitis, cirrhosis, and liver cancer (Tietz, 1976). Hemolysis Hemolysis is the rupture of the erythrocyte cell membrane, resulting in a release of the intracellular contents into the plasma. Some of these components include intracellular ions, proteins, and hemoglobin. Hemolysis may occur as a result of several factors. Intravascular hemolysis can occur as a result of hemolytic anemias, such as sickle cell anemia, transfusion reactions, certain drug reactions, and

23 immune reactions against the cells (Gayler, 1999). Extravascular, or in vitro, hemolysis can result from contact of erythrocytes with a foreign surface, such as metal, using a small gauge needle during phlebotomy, exerting excess negative pressure during phlebotomy, and difficult venipuncture. These are collectively termed mechanical or physical trauma. Hemolysis is not visible until the specimen is centrifuged. Once centrifuged, hemolysis is visible only at concentrations of 20 mg/dl or higher, and is indicated by a reddish tint in the plasma or serum (Dorner, Hoffman, & Filipov, 1983). Figure 1 shows different degrees of hemolysis. Hemolysis is a common problem in hospitals, independent laboratories, and doctor’s offices. Hemolysis is known to have skewing effects on several chemistry metabolites. Therefore rejection of all hemolyzed specimens for all determinations in laboratories is common. Rejection of hemolyzed specimens is often based on subjective criteria, and not actual quantitative criteria (Dorner, Hoffman, & Lock, 1981). Potential problems of rejecting all hemolyzed specimens are recollecting specimens, resulting in further patient discomfort, increased costs, and longer result turnout time. Toledo Hospital in Ohio estimates that when a specimen is rejected, there is a 60 to 90 minute delay in result turnout, and when all hemolyzed specimens are

24 0 Trace 1+ 2+ 3+ 4+

0 50 100 200 400 800

Figure 1. Different Degrees of Hemolysis. (The top values are qualitative measurements of hemolysis, while the bottom values are quantitative amounts of free Hb in mg/dl. The samples were graded by a medical laboratory technician.) rejected, an annual cost of approximately $15,000 results (Spencer & Rogers, 1995). Hemolysis is especially a problem in emergency rooms and outpatient settings, because emergency rooms rely on fast results, and in outpatient settings, patients are no longer available to recollect the specimen. Hemolysis makes test result interpretation more difficult for the clinician. Therefore, it is important that laboratory technicians know whether or not the test ordered by the clinician is affected by hemolysis, and if so what degree of hemolysis affects the test, and

25 how much is the test affected (Yucel & Dalva, 1992). Exact effects of hemolysis vary in different analytical methods, so in turn, each laboratory should determine the effects of hemolysis on each test performed by the method it uses for analysis (Morgan, Vann Bonn, Jensen, & Ridgway, 1999). Many times the test assay books accompanying autoanalyzers indicate whether or not hemolyzed specimens can be used, but they do not quantify the effects of hemolysis on the assay. Also, some newer model analyzers indicate hemolysis and correct for it (Szamosi, 2001). The effects of hemolysis on a specific determination are usually directly related to a difference in intracellular and extracellular concentrations of the analytes. If the analyte being tested is in greater concentration in the erythrocytes than in the plasma, hemolysis results in a greater plasma concentration of the analyte. However, if the cells have a lower concentration of the analyte than the plasma, the plasma becomes diluted resulting in lower plasma concentration of the analyte. Also, hemolysis results in the release of hemoglobin, which can interfere with many colorimetric determinations. Finally, hemoglobin can inhibit or interfere with certain chemical reactions, such as diazotization in the analysis of bilirubin (Sonntag, 1986).

26 Purpose and Hypothesis The purpose of this study is to examine the effects of hemolysis on analysis of the metabolites seen in Table 2, collectively called the comprehensive metabolic panel. The comprehensive metabolic panel is commonly ordered by clinicians, and it is known that some of the metabolites in this panel are affected by hemolysis. It is hypothesized that the analytes greater in concentration within the cytoplasm compared to plasma, such as potassium, after hemolysis will result in greater serum concentrations. It is also hypothesized that those analytes with greater extracellular concentrations will be only mildly affected by diluting effects of hemolysis. In addition, it is hypothesized that the determination of some of the analytes will be affected by interference from hemoglobin. Finally, it is hypothesized that different degrees of hemolysis will yield different results in the analysis of the electrolytes.

27 CHAPTER II

MATERIALS AND METHODS

Blood samples were obtained from the antecubital vein of five medical technologists at Blount Memorial Hospital Laboratory using proper phlebotomy technique. The Human Participants Review Committee approval form, research approval by Blount Memorial Hospital Laboratory, and consent forms from each participant can be found in the Appendix. Twenty-one gauge needles and evacuated tubes were used insuring minimal trauma to the erythrocytes. The blood was placed in five 10 ml Vacutainer® serum tubes and in one 7 ml Vacutainer® tube containing the anticoagulant EDTA (ethylenediaminetetraacetic acid). The blood samples in the serum tubes were allowed to clot for 30 minutes. Serum was separated from the erythrocytes by centrifugation for 10 minutes at 3500 rpm. Serum was removed from the tubes and refrigerated. Before lysis, the whole blood in EDTA was analyzed for intracellular hemoglobin (Hb) concentration in an automated hematology analyzer (Coulter MAXM, Coulter Corporation, Miami, FL 33196) using a method in which a lytic reagent destroys the

28 erythrocytes and converts the hemoglobin to a stable cyanide-containing pigment. Using this technique, the absorbance of the pigment at 525 nm is directly proportional to the Hb concentration of the sample. The accuracy of this method equals that of the hemiglobincyanide method, the reference method of choice for hemoglobinometry recommended by the International Committee for Standardization in Hematology (Coulter MAXM Reference, 1993, p. 5). The whole blood was then centrifuged and the erythrocytes were harvested. Erythrocytes were washed three times with isotonic saline solution for 10 minutes at 3500 rpm, removing the supernatant each time. The washed erythrocytes were then frozen at –30°C for 30 minutes to induce hemolysis. The lysed erythrocytes were thawed in a water bath at 37ºC for 15 minutes. The hemolysate was examined microscopically to ensure lysis of every cell. Based on the Coulter Hb concentration, the hemolysate was added to the serum achieving concentrations of 50, 100, 200, 400, and 800 mg/dl of free Hb. The original serum contained negligible, if any, free Hb. One aliquot of serum was unadulterated serving as a control. For accuracy comparison, serum samples were visually and qualitatively graded by a medical laboratory technologist as 0, trace, 1+, 2+, 3+, and 4+ hemolysis, which correspond to the respective

29 aforementioned concentrations, by comparison with a chart of standards (Vitros Sample Integrity Chart, Ortho-Clinical Diagnostics, Rochester, NY 14626). Comprehensive metabolic panels were performed on each serum aliquot in the Vitros 950 chemistry auto analyzer (Ortho-Clinical Diagnostics, Rochester, NY 14626). The comprehensive metabolic panel includes the analyses listed in Table 3. The listed methods are those used by the Vitros 950 (Vitros Chemistry, 2002). For each analyte listed in Table 3, an Analysis of Variance (ANOVA)(Statview 5.0, 2000) was conducted in order to determine if significant difference existed. If a significant difference was established, Fisher’s PLSD post hoc test (Statview 5.0) was used to determine if significant difference existed between each degree of hemolysis and the control.

30 Table 3. The Comprehensive Metabolic Panel Analyses and Methodology Used by the Ortho-Diagnostics Vitros 950. Analysis Methodology Potassium Ion Potentiometric Chloride Ion Potentiometric Sodium Ion Potentiometric Calcium Ion Colorimetric, 680 nm Glucose Colorimetric, 540 nm BUN Colorimetric, 670 nm Creatinine Two-point rate, 670 nm AST Multiple-point rate, 670 nm ALT Multiple-point rate, 340 nm Alkaline Phosphatase Multiple-point rate, 400nm Total Bilirubin Colorimetric, 540, 460 nm Conjugated Bilirubin End-point colorimetric, 400, 460 nm Unconjugated Bilirubin End-point colorimetric, 400, 460 nm Total Protein Colorimetric, 540 nm Albumin Colorimetric, 630 nm Carbon Dioxide Enzymatic end-point, 340 nm

31 CHAPTER III

RESULTS

Intracellular hemoglobin concentrations given by the Coulter MAXM automated hematology analyzer were used in order to calculate dilutions of free hemoglobin in serum after hemolysis. Intracellular hemoglobin values are presented in Table 4 for each participant.

Table 4. Intracellular Hemoglobin Concentrations of the 5 Participants. Participant Hb Concentration (g/dL) 1 12.2 2 13.6 3 12.7 4 13.3 5 13.4

After the hemoglobin dilutions were made, a medical laboratory technician (MLT) visually and qualitatively graded the samples according to a

32 reference chart as 0, trace, 1+, 2+, 3+, or 4+ hemolysis. These qualitative grades correspond to free Hb concentrations of 0, 50, 100, 200, 400, and 800 mg/dL. Each sample was numerically coded with a number and a letter. The number represented the participant and the letter represented the actual grade of hemolysis. This code was established to ensure the MLT did not know the actual grade of the specimen. Table 5 presents the visual grades by the MLT and the qualitative grades that correspond to the actual concentration of free Hb in the sample. The MLT was inaccurate in 26% of the grades.

Table 5. MLT Visual Hemolysis Grades vs. Actual Grades. Participant A-0 B-4+ C-3+ D-2+ E-1+ F-Tr 1 0 4+ 3+ 2+ 1+ Tr 2 0 4+ *4+ 2+ 1+ Tr 3 0 *3+ *2+ *Tr *Tr Tr 4 0 4+ *1+ *Tr 1+ *1+ 5 0 4+ 3+ 2+ 1+ Tr *Samples in which the MLT was incorrect.

The ANOVA for the effects of each analyte in the comprehensive metabolic panel was used to determine if significant difference at a 95% confidence level existed between hemolyzed samples and non-hemolyzed

33 samples (i.e., the control). A significant difference is established by a p-value less than 0.05. If a significant difference was recognized, a Fisher’s PLSD post hoc analysis was used in order to determine where variation existed among the different degrees of hemolysis. In the Fishers PLSD post hoc analysis, a 95% confidence level was also used to distinguish significant difference. Table 6 presents the mean results and standard error of each comprehensive metabolic panel. Significant differences established by ANOVA and by Fishers PLSD post hoc analysis are denoted with an asterisk. The units in which the analyte is reported is displayed under the abbreviation.

34 Table 6. Mean Results (± 1 SE) of the Comprehensive Metabolic Panels for Each Degree of Hemolysis (mg/dL free Hb). Analyte Control 50 100 200 400 800 *K+ (mmol/L) 4.2 4.34 4.46 4.62 5.26 *9.04 ±0.071 ±0.060 ±0.093 ±0.146 ±0.275 ±1.108 Cl- (mmol/L) 111 110.2 111.4 111 110.8 110.2 ±2.025 ±2.154 ±2.249 ±2.025 ±2.177 ±2.177 Na+ (mmol/L) 146.76 146.56 146.76 146.4 146.02 141.9 ±2.254 ±1.963 ±2.098 ±1.959 ±1.931 ±2.498 Ca++ (mmol/L) 10.16 10.16 10.24 10.3 10.18 9.52 ±0.157 ±0.214 ±0.196 ±0.202 ±0.203 ±0.383 *Glu (mg/dL) 89.5 88.75 89 88.5 90 *103.25 ±2.723 ±2.626 ±2.483 ±2.533 ±1.683 ±5.588 BUN (mg/dL) 11.8 12.2 12.2 12.4 12.6 12.8 ±0.49 ±0.663 ±0.49 ±0.51 ±0.51 ±0.49

35 Analyte Control 50 100 200 400 800 Crea (mg/dL) 0.994 0.97 0.968 0.948 0.936 0.88 ±0.064 ±0.073 ±0.069 ±0.936 ±0.062 ±0.065 *AST (U/L) 20.6 24.8 27.6 33.4 *58 *188 ±1.166 ±1.2 ±1.631 ±4.045 ±9.529 ±28.515 *ALT (U/L) 21.6 20.6 20 17.2 *12.6 *3.6 ±2.315 ±2.159 ±2.074 ±2.557 ±3.696 ±1.47 ALKP (U/L) 75.2 69.2 67.8 66.6 *63.2 *63.4 ±3.338 ±2.245 ±1.934 ±2.804 ±3.216 ±5.105 *TBIL (mg/dL) 0.26 0.42 0.48 0.68 1.56 *7.72 ±0.04 ±0.037 ±0.058 ±0.124 ±0.394 ±1.966 *Bu (mg/dL) 0.12 0.18 0.18 0.2 *0.28 *0.38 ±0.037 ±0.049 ±0.058 ±0.045 ±0.037 ±0.049 Bc (mg/dL) 0 0 0 0 0 0 *TP (g/dL) 7.26 7.4 7.48 7.48 7.94 *10.8 ±0.093 ±0.055 ±0.124 ±0.136 ±0.244 ±0.093

36 Analyte Control 50 100 200 400 800 *Alb (g/dL) 4.46 4.52 4.6 4.68 *5.12 *6.26 ±0.117 ±0.107 ±0.126 ±0.156 ±0.307 ±0.240

*CO2 (mOsm/kg) 19.4 20.2 19 18.8 19.6 17.8 ±0.245 ±0.97 ±0.707 ±0.663 ±0.678 ±1.241 *An asterisk by the analyte indicates significant difference established by ANOVA, while an asterisk by the value indicates significant difference compared to the control as established by Fishers PLSD. Both employ a 95% confidence level.

37 CHAPTER IV

DISCUSSION

All of the intracellular hemoglobin concentrations obtained from the Coulter MAXM were in the normal range of 12-15 g/dL, and there was only a small difference between the highest Hb concentration and lowest Hb concentration in this study. Very small volumes of the hemolysate were used to make the dilutions in serum, on the order of 10-100 µL. As more hemolysate was added to unadulterated serum, the brightness of the red tint increased. In those samples that had only 50 or 100 mg/dL of free Hb, the appearance of the serum was only minutely affected and the red tint was very faint. However, the serum samples that contained 400 and 800 mg/dL had a very bright, cherry red appearance. This color was a result of the heme molecule in the Hb. An example of the samples can be seen in Figure 1. Several studies have been published on the interference of hemolysis in chemical analyses; however, the specimens were only graded qualitatively, with no intention to know the exact concentration of

38 free Hb in the serum in most of those studies. Also, the range of concentrations in the present study is greater than that of most studies which only had one or two different concentrations of free Hb and a control. The present study is unique in that it focuses specifically on the comprehensive metabolic panel, a diagnostic chemistry panel commonly ordered by physicians while other studies have included many other analyses, and excluded some of the analyses within the comprehensive metabolic panel. Depending on the mechanism involved in the very broad spectrum of methods used to determine concentrations of the analytes within the comprehensive metabolic panel, there are several ways interference by hemolysis can occur. First, if an analyte has greater intracellular concentration than extracellular concentration, upon lysis the contents are released into the extracellular fluid resulting in an increased concentration of the analyte in the plasma (Benson, Paul-Murphy, & MacWilliams, 1999). Secondly, several different hemoglobin derivatives are found within the erythrocyte resulting in spectrophotometric absorption at several wavelengths. Hemoglobin absorbs light at wavelengths of 417-575 mm (Ramer, MacWilliams, & Paul-Murphy, 1995, p. 64). Many of the analyses use wavelengths in this range resulting in chromogenic interference by hemoglobin.

39 In addition, a third way hemolysis affects plasma concentrations is by the release of some intracellular constituents that interfere with the chemical reactions of the assay method (Sonntag, 1986). Finally, if the concentration of an analyte is higher in the plasma, then upon lysis of the cells, the plasma becomes diluted with intracellular fluid, resulting in lowered plasma concentrations (Caraway, 1962). In the present study, statistically significant increases (p < 0.05) in serum analyte concentration following hemolysis were found in potassium, glucose, aspartate aminotransferase, total bilirubin, unconjugated bilirubin, neonatal bilirubin, total protein, and albumin. Urea nitrogen had a visible linear increase with increased hemolysis; however, the increase was not statistically significant. Alanine aminotransferase was significantly decreased. Also, there were visible decreases in calcium, sodium, creatinine, carbon dioxide, and alkaline phosphatase; however, they were not significantly different either. Hemolysis up to 800 mg/dL had no significant or visible effect on chloride or conjugated bilirubin. Potassium There was a very significant increase in mean K+ concentrations at high degrees of hemolysis, as was expected. Potassium concentrations in this study were

40 only significantly increased at 800 mg/dL of free Hb. Hemolysis causes interference because the intracellular K+ concentration is about 20 times higher than serum (Sonntag, 1986). Although an increase in K+ concentrations is in accordance with the findings of other investigators, other studies have shown that K+ is statistically increased at even lower degrees of hemolysis (Sonntag). However, Frank, Bermes, Bickel, and Watkins (1978) found that lesser degrees of hemolysis, such as 50 or 100 mg/dL, would only be a cause for concern for those values of K+ at the extremes of the normal range. Glucose Glucose was significantly increased at 800 mg/dL of free Hb. This result contradicts many prior studies. Most previous studies had significantly lower serum glucose concentrations following hemolysis, or no effect at all. Although glucose was only significantly increased at 800 mg/dL of free Hb, there was a linear increase as free Hb increased. The reason for the contradiction of previous studies is the method of analysis. For example, Sonntag (1986) used two different methods to determine glucose and in neither method did hemolysis up to 660 mg/dL affect glucose concentrations; however, the methods used involved dialysis and deproteinization before analysis which resulted in the removal of Hb, thus there was no

41 interference. In the present study, glucose was determined colorimetrically at 540 nm, with no deproteinization. Hemoglobin also absorbs light at this wavelength resulting in an interference with the analysis, but in actuality glucose concentration does not increase. This difference is made clear in a study evaluating the effects of hemolysis on clinical chemistry analysis in ratites. This experiment employed two different methods of glucose analysis. Both were endpoint absorbances; however, the first reaction was evaluated at 340 and 380 nm. This resulted in a lowered glucose concentration. The second was evaluated using a wavelength of 550 (Andreasen, Andreasen, & Thomas, 1997, p. 168). Dorner, Hoffmann, and Filipov (1983) predicted that the selection of wavelengths between 505 and 590 nm would enhance the interference by hemoglobin in the samples whose endpoints were read in this range (p. 16). Because of Hb interference, there was an apparent increase in serum glucose which agrees with the present study. Aspartate Aminotransferase Aspartate aminotransferase increased dramatically in the present study from a mean concentration in the control of 20.6 U/L to 188 U/L at 800 mg/dL free Hb. There was a significant difference from the control at both 400 and 800 mg/dL of free Hb. This increase was

42 expected, and has been observed in several previous studies chiefly because the enzymic activity of AST is 40 times greater in erythrocytes than in plasma (Sonntag, 1986, p. 133). A previous study observed only slight, insignificant increases in AST concentration following hemolysis; however, the degree of hemolysis in that study was much lower than 400 mg/dL which would be required for a significant increase in AST (Yucel & Dalva, 1992). Total, Unconjugated, and Conjugated Bilirubin Total bilirubin was significantly increased by 800 mg/dL of free Hb in the present study. Unconjugated bilirubin was also significantly increased at both 400 and 800 mg/dL of free Hb. This was expected because increases in Bu are seen in hemolytic anemia (Vitros Chemistry, 2002). Conjugated bilirubin remained unchanged. This contradicts many previous studies. Several prior studies have shown that all bilirubin fractions decrease because the released hemoglobin competes with nitrite for sulfanilic acid in the diazotization reaction (Yucel & Dalva, 1992). However, in most of those studies the method used to determine bilirubin contained a blanking procedure for hemoglobin (Sonntag, 1986). The method in the present study did not subtract a sample blank, thus the hemoglobin interfered with absorption of those fractions between 417 and 525 nm.

43 The determination of total bilirubin and unconjugated bilirubin were at wavelengths in this range, resulting in interference. However the determination of conjugated bilirubin was at a lower wavelength; therefore, hemoglobin did not interfere. Total Protein Total protein was significantly increased by hemolysis at 800 mg/dL. This was expected as hemolysis results in the release of all of the intracellular proteins including hemoglobin (O’Neill & Feldman, 1989). Proteins have a higher concentration in the erythrocytes than in plasma (Yucel & Dalva, 1992). However, several studies have seen a decrease in total protein following hemolysis presumably due to a sample blank for hemoglobin being subtracted (Sonntag, 1986). Hemoglobin is the major protein within the erythrocytes. Therefore, in the studies that observed a decrease in total protein, the serum was diluted by the intracellular fluid and hemoglobin was not accounted for in the result. Albumin Albumin was significantly increased by 400 and 800 mg/dL of free Hb. This result was unexpected because little albumin exists within the red blood cells. Albumin is the protein with the highest concentration in the plasma. Prior studies have found increases, decreases, and no change in serum albumin

44 resulting from hemolysis. The study by Sonntag (1986) showed a decrease in serum albumin; however, a sample blank was subtracted resulting in the free Hb having no effect. Therefore, the intracellular fluid diluted the plasma, resulting in a lower plasma albumin. Ramer et al. (1995) proposes that assays for albumin are frequently measured at wavelengths in which Hb also absorbs, thus having an additive effect which accounts for the observed serum albumin concentration. The study by Yucel and Dalva (1992) showed no change in serum albumin resulting from hemolysis; however, they did not use concentrations of free Hb that were as high as those in the present study. Alanine Aminotransferase ALT was significantly decreased in 400 mg/dL and 800 mg/dL of free Hb. Previous studies also do not agree with respect to the effect of hemolysis on serum ALT levels. The study by Sonntag (1986) showed a significant increase in ALT concentration following hemolysis, as was expected because there is a sevenfold higher concentration of the enzyme in erythrocytes. However, other studies such as Ramer et al. (1995) and O’Neill and Feldman (1989) are in agreement with the present study, although none of those studies offer an explanation for the decrease. The decrease could possibly be a due to inhibition of the reaction by hemoglobin. Also, the pH of blood is

45 slightly decreased by hemolysis (Caraway, 1962). A large enough decrease in plasma pH could result in inactivation of the enzyme. Whatever the case, the effects of hemolysis on serum ALT are dependent on the method of analysis used. Urea Nitrogen, Calcium, Sodium, Creatinine, Carbon Dioxide, Alkaline Phosphatase, and Chloride Urea nitrogen had a visible linear increase as degree of hemolysis increased, but the increase was not statistically significant. None of the previous studies that were examined resulted in any significant change in urea. However in Laessig, Hassemer, Paskey, and Schwartz (1975), and in Ramer et al. (1995) a very slight, although not statistically significant, increase was observed at higher concentrations of free Hb. It is possible that the increase is simply artifactual and has no implications, especially since there is no statistical significance. Calcium, sodium, creatinine, carbon dioxide, and Alkaline phosphatase all had slight visible decreases at high degrees of hemolysis; however, none were significant. This is in accord with most of the previous studies examined. The small decreases evident in the present study probably resulted from dilutatory effects of the hemolysate. Frank et al. (1978, p. 1968) said that dilutatory effects are minimal because the increase in volume of a grossly

46 hemolyzed specimen is only 1.5%. Sonntag found a significant decrease in alkaline phosphatase, but admits there have been contradictory reports. Chloride was not significantly or visibly affected by hemolysis, which is also in agreement with the aforementioned studies. MLT Qualitative Grades The purpose behind having each specimen graded by a medical laboratory technician was to evaluate human error in determining the acceptability of a specimen. In 26% of the grades, the MLT was incorrect. This could result in the acceptance of a hemolyzed specimen that would compromise the test results, thus possibly affecting diagnosis. Conversely, if a specimen that is acceptable is rejected, turn-around times are lengthened, and schedules for timed analyses are affected. In Hemolysis Meets Quality Improvement Process it is suggested that each time a hemolyzed sample is rejected, there is a 60- to 90-minute delay in providing results to physicians (Spencer & Rogers, 1995, p. 41). There are several ways laboratories can alleviate problems caused by hemolyzed specimens. The best way is to find the root of the problem. Phlebotomy is an underrated skill. The value of test results depends on sample integrity. Integrated blood collection systems, such as Becton Dickinson

47 Vacutainer®, are much better at preventing hemolysis than traditional syringe and bottle methods. The reason for this is that the blood only has to pass through the needle one time in the Vacutainer®, and must pass twice through the syringe, once into the syringe and once out into the bottle. Also, sometimes nurses and emergency medical technicians draw blood from the catheter after inserting an IV. This results in severe hemolysis often because too much negative pressure is applied to the plastic catheter, resulting in a reduction of the radius of the catheter. This in turn creates shear stress on the red blood cells, lysing them. A way to avoid this is only to draw blood through large gauge IVs. In all cases, a smooth phlebotomy procedure will minimize in vitro hemolysis. Hemolysis can never be completely avoided because difficult venipuncture is sometimes unavoidable. Therefore, other measures must be taken in order to ensure accurate results. Presently, there are auto analyzers that detect and correct for hemolysis. This is not the best method, however, because the auto analyzers use the fraction of free Hb to make the calculation. There are instances when serum is left in contact with the erythrocytes too long. If serum is not separated within a short period of time, on the order of an hour, diffusion can occur across the cell membrane of the erythrocytes resulting in increased

48 plasma concentrations. An example of this is the loss of intracellular potassium through leak channels. In this instance no free Hb would be present, and no correction would be made. The best way to avoid this is prompt centrifugation and removal of serum or plasma. Free Hb can be measured and a calculation can be made to compensate for the effects of hemolysis; however, this method has been found unsuitable (Yucel & Dalva, 1992). This is an unreliable procedure primarily because intracellular concentrations can vary among individuals. Sonntag (1986) suggested that if hemoglobin interferes, errors can be corrected by using alternative methodologies. For example, if hemoglobin acts a chromogen, errors can be eliminated by using deproteinization, dialysis, sample blanking, and measuring at two different wavelengths. However, if an increased concentration of the analyte being tested is released from the erythrocytes into the plasma, such as potassium, the sample must be recollected. The elimination of errors by the aforementioned methods is a good approach; however, it would be too expensive for most laboratories to have several methods available for the same analyte. It has been shown that the determination of many clinical chemistry analytes is extremely method dependent. Different methods provide very different

49 results in some cases. Therefore, the effect of hemolysis on chemical assays should be determined by each individual laboratory for its particular methods of analyses (O’Neill & Feldman, 1989). Once the effects are made clear, then an accurate method for the determination of the degree of hemolysis should be employed. As previously shown, visual evaluation is not a very efficient method. There is a spectrophotometric method available for the determination of free Hb. It is based on the catalytic action of hemoglobin on the oxidation of TMB (3,3’,5,5’-tetramethylbenzidine) by hydrogen peroxide. The color formed in the reaction is proportional to the hemoglobin concentration (Sigma Biochemicals and Reagents Catalogue, 2000, p. 2679). This is an accurate method which would allow acceptance or rejection of the sample based on the previously determined criteria. Conclusions The results supported the original hypotheses. Increased serum concentrations resulted from the release of analytes that have high intracellular concentrations, such as potassium and AST. Also, those analytes with very low intracellular concentrations, such as sodium, were only mildly affected by dilution; however, the effect was not significant. The determination of some analytes, such

50 as glucose, was affected by direct interference from hemoglobin. The only result that did not agree with the hypothesis was alanine aminotransferase. Although ALT has higher intracellular concentration, hemolysis resulted in a significant decrease which was unexpected and remains unexplained. It is extremely important for each laboratory to know how hemolysis affects its chemical analyses. In the Vitros 950 chemistry auto analyzer (Ortho-Clinical Diagnostics, Rochester, NY 14626) used by Blount Memorial Hospital Laboratory to perform chemical analyses, including the comprehensive metabolic panel examined in this study, hemolysis at 800 mg/dL significantly affects potassium, glucose, aspartate aminotransferase, alanine aminotransferase, total bilirubin, unconjugated bilirubin, total protein, and albumin. In addition, albumin, neonatal bilirubin, unconjugated bilirubin, alanine aminotransferase, and aspartate aminotransferase were significantly affected at 400 mg/dL free Hb. There were no significant effects found below 400 mg/dL of free Hb. Thus, medical laboratory technicians at Blount Memorial Hospital Laboratory should reject specimens with 400 mg/dL of free Hb or greater for the comprehensive metabolic panel. However, invalidation of the individual analyses depends on the amount of free Hb present. Therefore if the physician orders a

51 comprehensive metabolic panel and is concerned only about potassium, total bilirubin, or glucose, a concentration of 400 mg/dL free Hb would be acceptable for analysis. Although these analyses are not affected significantly at this degree of hemolysis, the hemolysis should be accounted for when interpreting the result and in diagnosis.

APPENDIX

REFERENCES

Andreasen, C., Andreasen, J., & Thomas, J. (1997). Effects of hemolysis on serum chemistry analytes in ratites. Veterinary Clinical Pathology, 26(4), 165-171. Benson, K., Paul-Murphy, J., & MacWilliams, P. (1999). Effects of hemolysis on plasma electrolyte and chemistry values in the common green iguana (Iguana iguana). Journal of Zoo and Wildlife Medicine, 30, 413-415. Caraway, W. (1962). Chemical and diagnostic specificity of laboratory tests. American Journal of Clinical Pathology, 37, 445-464. Coulter MAXM reference manual. (1993). Miami, FL: Coulter Corp. Dorner, J., Hoffman, W., & Filipov, M. (1983). Effect of in vitro hemolysis on values for certain porcine serum constituents. Veterinary Clinical Pathology, 12(1), 15-19 Dorner, J., Hoffman, W., & Lock, T. (1981). Effects of in vitro hemolysis on equine serum chemical values. American Journal of Veterinary Research, 42, 1519-1522.

63 Fox, S. (1999). Human physiology (6th ed.). Boston: Mcgraw-Hill. Frank, J., Bermes, E., Bickel, M., & Watkins, B. (1978). Effect of in vitro hemolysis on chemical values for serum. Clinical Chemistry, 24, 1966- 1970. Gayler, M. (1999, May 26). Haemolysis of blood samples: What is it and how to avoid it. Nursing Times, 95, 54-55. Guyton, A., & Hall, J. (2000). Textbook of medical physiology (10th ed.) Philadelphia: W.B. Saunders. Henry, J. (1991). Clinical diagnosis & management by laboratory methods (18th ed.). Philadelphia: W. B. Saunders. Laessig, R., Hassemer, D., Paskey, T., & Schwartz, T. (1976). The effects of 0.1 and 1.0 per cent erythrocytes and hemolysis on Serum chemistry values. American Journal of Clinical Pathology, 66, 639-644. Morgan, L., Vann Bonn, W., Jensen, E., & Ridgway, S. (1999). Effects of in vitro hemolysis on serum biochemistry values of the bottlenose dolphin (Tursiops truncatus). Journal of Zoo and Wildlife Medicine, 30(1), 70-75.

64 O’Neill, S., & Feldman, B. (1989). Hemolysis as a factor in clinical chemistry and hematology of the dog. Veterinary Clinical Pathology, 18(3), 58-68. Ramer, J., MacWilliams, P., & Paul-Murphy, J. (1995). Effects of hemolysis and frozen storage on serum electrolyte and chemistry values in cotton-top tamarins (Saguinus Oedipus). Journal of Zoo and Wildlife Medicine,26(1), 61-66. Sigma biochemicals and reagents catalogue. (2000). St. Louis, MO: Sigma Co. Sonntag, O. (1986). Haemolysis as an interference factor in clinical chemistry. Journal of Clinical Chemistry and Clinical Biochemistry, 24, 127-139. Spencer, L., & Rogers, L. (1995, October). Hemolysis meets QIP. Medical LaboratoryObserver. [Online–subscription only]. Retrieved February 11, 2002 from Infotrac Web: http://infotrac .galegroup.com/itweb/tel_a_maryville. Statview 5.0. (2000). Cary, NC: SAS Institute. Szamosi, D. (2001, July). Phlebotomy Standards. Medical Laboratory Observer. Retrieved February 11, 2002 from Infotrac Web: http://infotrac .galegroup.com/itweb/tel_a_maryville. Tietz, N. (1976). Fundamentals of clinical chemistry. Philadelphia: W. B. Saunders.

65 Vick, R. (1984). Contemporary medical physiology. Menlo Park, CA: Addison-Wesley. Vitros chemistry products instructions for use. (2002). Rochester, NY: Ortho-Clinical Diagnostics. Yucel, D., & Dalva, K. (1992). Effect of in vitro hemolysis on 25 common biochemical tests. Clinical Chemistry, 38, 575-577.