Charles D. Bolan, Roger J. Kurlander and Geraldine P. Schechter
CELLULAR ANALYSIS OF THE PERIPHERAL BLOOD AND BONE MARROW
The practice of hematology is characterized by safe and rapid access and analysis of the cellular elements of blood and bone marrow. Essential clinical information can be derived in less than an hour from a spun hematocrit and microscopic evaluation of peripheral blood or marrow using technology available for more than three quarters of a century,1 while an automated complete blood count, including the hemoglobin concentration, hematocrit, white blood cell and platelet count, white cell differential, and estimates of red cell size, can be provided within minutes.2
Complete Blood Count
Red Cell Parameters
Using traditional methodology, the hematocrit, defined as the packed red cell volume or percent of the blood occupied by red blood cells, was first measured by direct visual inspection of a tube of blood following centrifugation. The hemoglobin concentration and red cell count were also directly measured, by spectrophotometry and use of a counting chamber, respectively. Other red cell indices were indirectly calculated, including the mean corpuscular volume, MCV (hematocrit divided by the red cell count), mean corpuscular hemoglobin, MCH (hemoglobin divided by the red cell count), and mean corpuscular hemoglobin concentration, MCHC (MCH divided by the MCV). Automated cell counters, which directly analyze and count individual cells by changes in impedance or light scatter, provide information for many of the same parameters provided by traditional methods. Automated methods differ in that the hematocrit is derived indirectly, by multiplying the directly measured red cell count and mean corpuscular volume. The hemoglobin and MCV are thus more reliable, reproducible assays than the hematocrit when measured from an automated counter. The MCV is of great clinical utility in evaluating anemia according to red cell size (macrocytic, microcytic, and normocytic anemias), while the RBC count is useful in comparison of iron deficiency and thalassemia. The MCH, which represents the average total hemoglobin content of the red cells, is highly dependent on cell size, while the MCHC represents the average total hemoglobin concentration or degree of “redness” of the red cells. These parameters have less clinical utility than the MCV and RBC count, although the MCHC is classically increased in hereditary spherocytosis.
The RDW, red cell distribution width, is a measure of the variability in size within the red cell population that is provided by automated cell counters. The precise mathematical equation used for calculation varies with the specific instrument, but in all cases elevated values indicate anisocytosis. For casual observers, the RDW is more reliable than is inspection of the blood film for detecting this characteristic.3An elevated RDW is a valuable clue to abnormalities in red cell morphology; when the RDW is markedly increased, inspection of the blood film is essential. The RDW is sensitive to the presence of small subpopulations of large or small red cells, thus it may be more useful than the MCV for the early detection of nutritional deficiencies. The RDW often remains normal in thalassemia, and the combination of a high or normal RBC count, a low MCV, and a normal RDW is a common pattern in thalassemia trait.4 The use of the RDW in conjunction with other red cell indices has had limited acceptance in the classification of anemias because results may be unreliable in complex settings.
Erythrocytes newly released from the bone marrow contain intracellular RNA, which usually disappears within a day of circulation in the peripheral blood. These young red cells, termed reticulocytes based on the microscopic appearance of RNA-containing reticulum caused by supravital staining, are quantitated by microscopic examination of the peripheral smear. Optical and fluorescent cell-counting methods are also able to detect the uptake of RNA-binding dyes by reticulocyte red cells, and quantitation of the reticulocyte count by automated cell counters has largely replaced the older manual method. Although traditional microscopic and automated techniques produce similar values, the automated technique is more precise. Both methods are vulnerable to interference from intracellular organisms, basophilic stippling, and other artifacts; thus, microscopic inspection of a standard blood film, and/or of a reticulocyte preparation is warranted whenever a reticulocyte count seems inappropriately high for the clinical scenario.
The absolute reticulocyte count (absolute reticulocyte count = % reticulocytes X RBC count/100) is a more clinically useful parameter than the simple percentage of reticulocytes, which is influenced by patient total red cell count and hematocrit. Because a highly stimulated marrow produces larger, more polychromatophilic, RNA-rich reticulocytes (termed stress reticulocytes),5 which are released earlier from the marrow and stain with supravital dyes for up to 3 days after release into the peripheral circulation, the absolute reticulocyte count may overestimate the true rate of reticulocyte production in these clinical situations. Some automated counters may also provide the capability to detect reticulocyte subpopulations containing increased levels of RNA, designated as the immature reticulocyte fraction (IRF), which can indicate the early response to erythropoiesis-stimulating agents and early, pre-myeloid marrow recovery after chemotherapy or stem cell transplantation.6 Automated counters are also capable of selectively measuring the hemoglobin content of reticulocytes, which diminishes rapidly as accessible iron stores are exhausted, and this test may be of use in identifying early iron deficiency in blood donors undergoing repeated phlebotomy, iron-depleted dialysis patients receiving erythropoietin for the treatment of anemia, or iron-stressed women during the late stages of pregnancy.6
White Blood Cell Parameters
The automated white blood cell (WBC) count is measured using the same methods employed to count red cells and platelets. Since the peripheral blood in healthy individuals has approximately 1,000 fold more red cells and 40 fold more platelets than white cells, it is essential for the automated counter to distinguish among cell types. To facilitate the process, red cells are usually destroyed by osmotic lysis and residual red cells or platelets are excluded from the count based on size and granularity. Automated counting is faster, more reproducible, and usually more accurate than the older manual methods, particularly for leukopenic samples in which automated counters can quantitate WBC counts as low as 100 cells per μL.
Automated leukocyte differentials are quite accurate in characterizing white cells from normal controls or patients with qualitatively normal white cell morphology. In patients with hematologic disease or qualitative leukocyte abnormalities, automated differentials may still be accurate, but their reliability cannot be assumed. Thus the white cell differential of new patients with complex clinical findings should also be evaluated on the peripheral smear. The automated absolute granulocyte count plays a reliable and important role in evaluating the risk of infection in neutropenic patients, in grading chemotherapy-induced toxicity, and in following the recovery from hematopoietic stem cell transplantation.
Platelets
Similar to RBC and WBC counts, the automated platelet count is measured by impedance- and/ or optical-based techniques. Platelets are distinguished from red cells by size and in some cases by resistance to osmotic lysis. Automated counters are more accurate than older manual counting methods in monitoring thrombocytopenia, and they often can accurately count less than 10,000 platelets per microliter. On occasion, it may be difficult even with advanced optical techniques to distinguish true platelets from debris derived from precipitated proteins or cell fragments.
The mean platelet volume (MPV) is a parameter that is routinely measured by automated analyzers. The MPV has modest value as a measure of increased platelet turnover and/or activation due to the fact that young platelets are larger than older platelets. However, the MPV may increase rapidly during the first 2 hours after collection because of shape changes and swelling in EDTA, and the reference standards are not adequate for comparing values from different institutions.7
Some automated counters can measure the RNA content of new released platelets, expressed as the immature platelet fraction (IPF) or absolute immature platelet number (per cent IPF times the platelet count). These measurements may be useful clinical parameters in assessing marrow recovery after hematopoietic transplantation and marrow thrombopoiesis in cases of suspected idiopathic thrombocytopenic purpura (ITP).8,9
Sources of Artifact
Clinically relevant artifact at the point of sample collection may occur due to clot formation or due to dilution of blood with intravenous fluid.10,11 Although most automated cell counters are equipped to detect clot formation, some may not be detected, resulting in falsely lower platelet and WBC counts. In contrast, fluid dilution generally is apparent by some degree of change in all three cell lines (red cells, white cells, and platelets) as compared to a previous sample. Both clot formation and dilution may occur more frequently in samples drawn from indwelling catheters.
Artifact may also be introduced by factors associated with the storage of blood samples. When stored at room temperature, absolute RBC, WBC, and platelet counts are stable for up to 3 days, but the red cell MCV increases within 24 hours due to cell swelling, resulting in associated increases in the hematocrit and red cell distribution width.12 After 2 days, the relative proportion of monocytes on the WBC differential also decreases, while relative leukocyte, lymphocyte, and eosinophil counts increase.12
Red cell agglutination within the test tube caused by cold or warm reactive antibodies or rouleaux formation can artifactually reduce the red cell count, while extreme leukocytosis, particularly with a WBC count of more than 500,000/μL, can falsely increase the RBC count and calculated hematocrit.11 Similarly, RBC counts in patients with extremely small or fragmented red cells may be artifactually low. Falsely decreased platelet counts, or pseudothrombocytopenia, caused by platelet clumping and platelet satellitism induced by the EDTA anticoagulant, may occur in up to 0.1% of normal samples and may be present in up to 15% of patients referred for evaluation of a low platelet count.13 In most cases, pseudothrombocytopenia can be eliminated by complete blood count performed on samples collected into a tube containing citrate anticoagulant, in which instance the measured platelet count should be multiplied by 1.1. Platelet clumping may sometimes also falsely elevate the white blood cell count, while small white or red cell fragments, or protein precipitates may artifactually increase the platelet count.10,11
Peripheral Blood Film
One common use of review of the peripheral blood film is by laboratory personnel when values reported by automated counters are “flagged” on the device result print out, or by hematologists in clinical situations circumstances in which verification of the automated result is indicated. The automated value can be compared to a manual estimate obtained by counting the average number of platelets present in 5 to 10 oil immersion high power (1,000×) fields (HPF) and/or the number of leukocytes in a similar number of low power (100×) fields (LPF). The selected fields are chosen from representative regions that contain an even monolayer of cells, avoiding the edge of the film. The platelet count and leukocyte count are estimated by using the formulas:
Platelet count (platelets/mm3) = average platelets per HPF × 15,000
Leukocyte count (cells/mm3) = average leukocytes per LPF × 250
Evaluation of the peripheral blood smear is valuable in the evaluation of hematologic disease and should always be performed in conjunction with review of the bone marrow aspirate and core biopsy. Examination of the peripheral smear is not necessary in simple cases of anemia due to iron, vitamin B12, or folate deficiency. However, it can be very useful in cases that do not respond to therapy and in instances of suspected hemolysis or complex anemias, and it is essential in the evaluation of microangiopathic disorders such as thrombotic thrombocytopenic purpura (TTP). The presence of spherocytes, microspherocytes, schistocytes, targets, sickle cells, red cell membrane irregularities, polychromasia, and intracellular inclusions or parasites should be sought in cases of suspected hemolysis, while tear drop cells and nucleated red cells may be present in marrow fibrosis or tumor invasion. Evaluation of white blood cell abnormalities such as the presence of immature cells and blasts are indicated in cases of suspected malignancy, while other changes such as hypogranulated and hypersegmented granulocytes, Pelger-Huet cells, toxic granulation, and Döhle bodies may point towards a previously unsuspected underlying process.
Review of granulocyte morphology by review of the peripheral smear is superior to automated methods for evaluation of a shift to the left in the myeloid series, or an increased proportion of immature forms, which may be a common finding during the early response to stress or infection. The neutrophil band count is performed manually by grading 1,000 neutrophils on a Romanowsky-stained peripheral blood film, and expressed in percent or as an absolute number.
Routine inspection of platelet morphology is indicated in all patients with newly diagnosed thrombocytopenia in order to exclude platelet clumping and artifactual thrombocytopenia. Review of the blood film may also reveal morphologic abnormalities such as hypogranularity, as in gray platelet syndrome or myelodysplasia, or the presence of very large platelets in inherited syndromes such Bernard-Soulier disease, and May-Hegglin anomaly.
Bone Marrow Aspirate/Biopsy
Indications
Examination of the bone marrow aspirate and biopsy may establish or exclude diagnoses such as aplastic anemia, myelodysplastic syndrome, hemophagocytosis or marrow replacement by non-hematopoietic cells in patients with unexplained persistent or severe thrombocytopenia and granulocytopenia. Bone marrow evaluation is not indicated in cases of isolated anemia in which clinical information and review of the peripheral smear suffice to make the diagnosis. However, in patients with unexplained anemia, particularly if requiring transfusions, a bone marrow study is appropriate. Although marrow aspiration and biopsy are invaluable in hematologic malignancies, and in screening for metastatic spread of non-hematologic malignancies, radiologic or nuclear medicine techniques may be more sensitive methods for detecting metastases. Bone marrow can also be used to culture and identify infectious agents such as mycobacteria and fungi. Disorders such as Gaucher disease or amyloidosis may also be discovered, but biochemical tests for the former or less invasive biopsies for the latter are more useful.
Risks
Bone marrow aspirate biopsy can be performed with minimal discomfort using local anesthesia and is generally very safe when performed by an experienced operator. Informed consent should always be obtained, for the procedure as well as the performance of research studies or specialized laboratory tests. Attention should be paid to both patient comfort, as well as toward obtaining adequate sample volume—particularly of the core biopsy—to conduct the indicated analyses. Infiltration of the skin surface and periosteum with several milliliters of lidocaine provides sufficient anesthesia for most patients; conscious sedation should be reserved for those with significant anxiety and apprehension. Thrombocytopenia is not a contraindication provided adequate landmarks are identified and sufficient direct pressure is applied to the site post-procedure; some patients with platelet function defects or thrombocytopenia due to myelodysplasia may require platelet transfusion for control of bleeding. Anticoagulation should be discontinued or held for an adequate time prior to the procedure, and coagulation factor replacement be utilized for patients with serious factor deficits in order to avoid significant bleeding.
Hemorrhage and local infection occur rarely. In one study of more than 50,000 biopsies, there were 14 instances of serious hemorrhage with one death, and 6 instances requiring transfusion.14 Infection can be safely managed by the use of careful sterile technique and proper local care even in patients with neutropenia and compromised immune function.
Technique
In adults, the posterior iliac crest is the site of choice for most hematologists; it can be located using direct palpation with the patient in the prone position and generally provides adequate core and aspirate samples. The anterior iliac crest is a reasonable alternative when obesity, local irradiation, or local skin conditions preclude a posterior approach. Sternal aspiration may be justified when an iliac approach is not possible. However, the sternal approach cannot be used to obtain a core biopsy specimen, may be less well accepted by patients, and is more vulnerable to complications. Traumatic penetration or fracture of the sternum with damage to underlying structures is a rare but serious complication of particular concern when underlying structural damage secondary to malignancy or multiple myeloma is present.
Bone Marrow Aspirate
The bone marrow aspirate is obtained by advancing a specially designed needle fitted with an obturator through the cortex into the medullary space with use of negative pressure to withdraw cells into a syringe after removal of the obturator. The presence of individual marrow particles or particles is determined by visual inspection. The particles are spread on a glass slide, and they may be stained immediately to determine the presence and morphology of cellular marrow elements, such as megakaryocytes or blasts in the evaluation of suspected ITP or leukemia. Cell suspensions obtained from an aspirate may also be used in special studies such as cytogenetic and flow cytometric analyses, while the aspirate clot can also be fixed in a clot section, to be stained with hematoxylin and eosin, or other stains in the same way as a bone marrow biopsy. The marrow may be inaspirable in some patients due to hypercellularity, hypocellularity, fibrosis, or metastatic cancer cells. Aspirates are less reliable than biopsies in detecting marrow involvement with malignancy and not useful at all to detect myelofibrosis or granulomas.
Bone Marrow Biopsy
It is almost always possible to obtain a biopsy specimen, even when hematopoietic cells have been totally replaced by fibrous tissue or tumor. Hematoxylin and eosin-stained biopsy sections show less cytoplasmic and nuclear detail than Romanowsky-stained aspirate films, but provide other essential information about marrow architecture and cellularity. The bone marrow biopsy specimen, or core, is obtained by further advancing the bone marrow biopsy needle through the medullary space after removal of the obturator. In some situations, the core may be obtained from the same puncture site used to obtain the aspirate by advancing deeper into the marrow cortex after obtaining cells for the aspirate; this technique is convenient for the operator and patient but may be associated with artifactual hypocellularity, hemorrhage, and distorted architecture.15 Aspiration artifact may be reduced by obtaining a larger core sample size (1.5–2.5 cm), which may also enhance the evaluation of the marrow composition. The core biopsy sample is ejected from the needle by using a metal probe, fixed, decalcified, sectioned, and stained with hematoxylin and eosin, histochemical stains (for reticulin, collagen, iron), or a variety of immunohistochemical stains.
Marrow Cellularity
Marrow cellularity should be estimated from review of a large bone marrow core biopsy specimen. Because normal marrow cellularity decreases with age; normal percent cellularity can be estimated by subtracting patient age from 100. Inappropriately low cellularity suggests marrow damage, and high cellularity may be consistent with a proliferative disorder, dysplasia, stress reaction, or the use of growth factors.
Megakaryocyte Number and Appearance
A normocellular biopsy and aspirate specimen should contain multiple megakaryocytes per low power (100×) field. An increase in megakaryocytes is consistent with increased turnover secondary to peripheral destruction, inflammation, iron deficiency, or myeloproliferative/myelodysplastic disorders. Reduced megakaryocyte numbers may reflect a primary marrow disease such as aplastic anemia, amegakaryocytic thrombocytopenia, or suppression secondary to chemotherapy. Normal megakaryocytes appear as large cells with multilobated (three or more) attached nuclei. The presence of a substantial number of smaller megakaryocytes with less than three nuclear lobes indicates a shift to the left in megakaryocyte maturity due to increased platelet turnover or dysplasia; megakaryocytes with a single nucleus or multiple small separated nuclei and mature cytoplasm are particularly suggestive of myelodysplasia.
Myeloid to Erythroid Ratio
The myeloid to erythroid (M:E) ratio is quantitated by counting 300 to 500 cells from a marrow aspirate sample. In adults, the normal M:E ratio varies from 1:1 to approximately 3:1. The M:E ratio should be interpreted in the context of overall cellularity, the qualitative appearance of the affected cells, and the clinical setting. Erythroid hyperplasia, characterized by a low M:E ratio in a cellular marrow, suggests an erythroid response to anemia, (especially hemolytic anemia), myelodysplasia, or erythropoietin administration.
Erythroid hypoplasia may indicate red cell aplasia, decreased erythropoiesis due to autoimmunity, medication side effects, or major-ABO incompatible hematopoietic stem cell transplantation. Myeloid hyperplasia may be seen in response to physiologic stress, infection, exogenous growth factors, or in myeloproliferative disorders. Myeloid hypoplasia with a “maturation arrest” (an absence of myeloid precursors beyond the promyelocyte or myelocyte stage) may reflect drug-induced or autoimmune agranulocytosis.
Myeloid and Erythroid Cells
Myeloid and erythroid cell lines should demonstrate normal morphology, a normal distribution of maturation, and synchronous maturation of nucleus and cytoplasm. Disruption of the normal maturation sequence (such as a complete maturation arrest or a preponderance of immature cell forms, or a “shift to the left” of the differentiation pattern), the presence of excess numbers of blasts, or dysplastic changes affecting at least two of the three major hematopoietic cell-lines suggests a serious hematologic disorder. A mild shift to the left and mild megaloblastic changes are more nonspecific.
Lymphocytes and Plasma Cells
There is substantial normal variation in the number of lymphocytes encountered in a marrow specimen, and these elements may be distributed diffusely or in well-defined lymphoid aggregates. Paratrabecular lymphoid aggregates may occur in follicular lymphoma. Benign lymphoid aggregates typically contain more T cells than B cells when immunohistochemical stains are performed, while B cell-rich lymphoid collections are more likely due to clonal B cell lymphoproliferative disorders.
Plasma cells usually constitute less than 2% of marrow cells; increases may be noted in inflammatory disease, benign monoclonal gammopathy, and multiple myeloma. Distinguishing reactive from neoplastic plasma cells in a routine aspirate film may be difficult, as reactive plasmacytosis caused by inflammation or liver disease may reach 20% to 30%, and hypocellular marrows are often rich in plasma cells. Extensive multinuclearity or plasma cell aggregates of more than 5 to 10 cells are suspicious for malignancy, while conspicuous variation in cell size and immature nuclei with nucleoli may be present in myeloma. Immunohistochemical staining for intracellular kappa and lambda chains within plasma cells in the marrow biopsy can indicate the presence of a monoclonal process; distinguishing benign monoclonal gammopathy from multiple myeloma may require additional clinical and laboratory data. The presence of concentrated collections of plasmacytoid lymphocytes suggests Waldenström macroglobulinemia or lymphoplasmacytic lymphoma.
Other Cells Anomalous Present in the Marrow
Malignant cells of epithelial or mesenchymal origin adhere more tightly together and aspirate poorly, forming tightly clumped groups of unusually large cells with a very high nuclear-cytoplasmic ratio. Such clumps may be rare; therefore, when screening an aspirate for malignant cells, the whole slide should be scanned at low power including the leading edge of the film. Tumor cells can also be readily identified in marrow biopsies, and specialized immunohistochemical stains can sometimes help identify the site of origin.
Bone Marrow Iron Stores
When marrow iron is markedly increased, yellow granules of hemosiderin may be seen in routinely stained aspirate and biopsy preparations. The Prussian blue stain, which specifically detects iron, is necessary to evaluate lesser iron stores and iron granules in erythroid cells. Ringed sideroblasts are erythroid precursors containing coarse iron granules immediately surrounding at least half the nuclear circumference due to accumulation of iron within mitochondria. They are always abnormal and imply anomalous porphyrin synthesis secondary to a congenital abnormality, pyridoxine deficiency, exposure to toxins (such as lead or alcohol), medication, or myelodysplasia.
Although tissue iron stores can be assumed to be adequate when stainable marrow iron is demonstrated, this does not indicate that the stores can be mobilized for effective erythropoiesis. For example, when hepcidin levels are elevated by chronic inflammation, iron will be trapped in the macrophages and unavailable for hemoglobin synthesis. In contrast, the absence of stainable iron suggests depleted iron stores. The sensitivity of this method depends heavily on the quantity of sample obtained; for maximal yield, at least 7 separate particles should be evaluated for stainable iron.16 In addition, the absence of stainable marrow iron may incorrectly suggest iron depletion due to hypocellularity, technical error, or presence of cellular antigens binding to antibodies labeled with dyes.17
Additional Studies
Flow cytometry, which allows computerized digital analyses of cells according to size, granularity, and light scattered from colored dyes attached to specific cell markers, is an essential element in the analysis of hematologic malignancy, clonality, and other abnormalities, and can be performed on cells from peripheral blood and marrow.18
Because a very wide variety of possible analyses can be performed, the clinician should specify the clinical question and appropriate differential diagnosis to the laboratory prior to collection to facilitate proper labeling and use of appropriate device settings. As with the automated CBC, flow cytometry is subject to errors and artifacts due to clotting and cell clumping; thus, samples should be carefully collected into appropriately prepared tubes and rapidly and safely transported to the laboratory.
Molecular diagnostic studies such as cytogenetics and fluorescent in situ hybridization (FISH) have an important role in diagnosis and monitoring of hematologic disorders, especially myeloid malignancies, while immunohistochemical stains and molecular analysis of T and B cell receptor rearrangements are critical in documenting clonality in lymphoid malignancies.
SERUM TESTS TO EVALUATE NUTRITIONAL AND HYPOPROLIFERATIVE ANEMIAS
Serum Iron and Total Binding Capacity Assays
Serum iron is measured by automated chemical assays following dissociation from transferrin.19 Total iron binding capacity (TIBC), which is mainly binding to transferrin, can be measured by the addition of excess iron to the sample. Unbound iron is removed by absorption, and the iron bound to protein is again dissociated and measured by the serum iron assay. In many laboratories, unsaturated iron binding capacity is now measured directly by automated chemistry analyzers and the TIBC is calculated.19 Measurement of serum iron can be falsely elevated in specimens containing hemoglobin (hemolysed specimens) and also for many hours transiently after transfusion of older units of red blood cells due to clearance of red cells damaged in storage.20 The percent saturation of transferrin by iron is calculated by dividing the serum iron by the total iron binding capacity and then multiplying by 100.
There is a marked diurnal variation of serum iron in healthy people, with highest levels in the morning; by midnight, serum iron levels can be very low and may fall into the iron deficiency range. Iron levels also vary with the menstrual cycle, with a 10% to 30% increase premenstrually and a similar decrease at the time of menstruation.19 Iron deficiency is characterized by low serum iron values and elevated serum total iron binding capacity and therefore, low iron saturation of transferrin. Serum iron levels also fall below normal in patients with the anemia of inflammation/chronic disease, but since total iron binding capacity also falls, percent iron saturation may remain normal. Levels of serum iron and the percent iron saturation in patients with severe chronic inflammation frequently overlap with those found in iron deficiency anemia.21 The high TIBC levels (more than 300 μg/dL) in patients with uncomplicated iron deficiency are helpful in distinguishing the two entities, but when iron depletion and inflammation coexist, the TIBC is frequently low, and serum ferritin and possibly transferrin receptor assays become necessary to help confirm the diagnosis of iron deficiency21 (see below).
Elevated serum iron levels and iron saturation occur in multiple inherited and acquired conditions (Table 27.1). Genetic hemochromatosis of many different types results in hepcidin deficiency, which leads to high gastrointestinal iron absorption despite high levels of iron. The iron overload of thalassemia major is mainly due to multiple transfusions but is also compounded by increased gastrointestinal iron absorption that is stimulated by hepcidin suppression by increased erythropoiesis22 The latter mechanism accounts for iron overload in patients with intermediate thalassemias and hemolytic anemias. Elevated serum iron levels also results from the failure of reutilization of iron, as in megaloblastic and sideroblastic anemias, and with pure red cell aplasia. Other causes include fulminant acute hepatitis due to severe hepatocyte injury and also chronic hepatitis, particularly hepatitis C. In patients with iron deficiency, a rise in serum iron to levels greater than 100 mcg/mL may occur 1 to 2 hours after ingestion of 325 mg of ferrous sulfate in the form of a tablet or elixir which indicates appropriate bioavailability and normal small bowel absorption.21
Soluble Transferrin Receptor
Soluble transferrin receptor, a truncated form of tissue transferrin receptor, is measured by a “sandwich type” enzyme-linked immunoassay. Serum transferrin receptor levels are elevated in states of increased erythropoiesis such as hemolytic anemias, megaloblastic anemia, thalassemia, and also in iron deficiency anemia. Transferrin receptor levels are decreased in conditions of reduced erythropoiesis such as aplastic anemia and renal insufficiency. The assay is reported to distinguish iron deficiency from the anemia of chronic inflammation/disease,21,23 but overlap in receptor levels between the two conditions is often noted. The ratio of the serum transferrin receptor level to the ferritin level or to log ferritin level may be more useful in distinguishing between the two entities in confirming when they coexist.23Elevated transferrin receptor levels observed in some patients with anemia of inflammation likely reflect limited availability of functional iron pools for erythropoiesis rather than depleted iron stores.24Automated assays are now available.23,25 As discussed previously, changes in reticulocyte hemoglobin content, which is a direct measure of new red cell hemoglobinization, may be the most dynamic and sensitive measure of functional iron availability during active hematopoiesis.
Table 27.1 Causes of High Serum Fe and Fe Saturation
Iron Overload States
Genetic hemochromatosis
HFE, TfR, HJV, hepcidin, ferroportin mutations
Transfusion hemosiderosis
Impaired Reutilization of Iron
Pure red cell aplasia
Aplastic anemia
Ineffective erythropoiesis
Nutritional megaloblastic anemias
Sideroblastic anemias
Thalassemia intermedia and major
Congenital dyserythropoietic anemias
Unknown or Multiple Mechanisms
Hepatitis C
Liver cirrhosis
Chronic alcoholism
Serum Ferritin
Serum ferritin can be measured by a variety of immunoassays.19 The level of serum ferritin generally correlates with body stores of iron. The ferritin assay is most useful clinically at the extremes: low values (<15 ng/mL) are very sensitive to storage iron depletion and very high levels (>1,000 ng/mL) usually indicate iron overload states. Since serum ferritin is an acute phase reactant, patients who have both iron deficiency and anemia of inflammation or liver disease often have serum ferritin levels in the normal range, usually between 40 and 70 ng/mL,21 and less than 7% of anemic patients with depleted iron stores exceeded a serum ferritin level of 100 ng/mL in one study.26 The soluble transferrin receptor/log ferritin index was reported to diagnose iron deficiency in 25% of patients with anemia of inflammation and ferritin levels >100 ng/mL.23
Very high levels of ferritin usually indicate iron overload states resulting from genetic hemochromatosis, liver disease, particularly chronic hepatitis C, or transfusion hemosiderosis. Patients with these disorders will have elevated serum iron levels. Ultrahigh levels above 5,000 ng/mL are often observed in severe inflammatory states, such as disseminated fungal or mycobacterial disease, or in disorders associated with macrophage activation, for example, Still’s disease or syndromes associated with hemophagocytosis. Table 27.2 lists the causes of high ferritin levels not associated with high levels of serum iron, which include genetic causes such as hyperferritinemia cataract syndrome27 and apparently acquired causes such as the “metabolic” syndrome associated with diabetes, obesity and hepatosteatosis.28 In the presence of concurrent inflammation or liver disease, ferritin levels may not accurately gauge the response to iron chelation therapy in patients with transfusion hemosiderosis and very high levels (>1,000 ng/mL).
Serum Vitamin B12 (Cobalamin)
Serum vitamin B12 is generally assayed by an enzyme-linked competitive binding assay most frequently based on binding to intrinsic factor.19 Serum levels below 100 pg/mL are almost invariably associated with cellular cobalamin deficiency, as reflected by elevated serum methylmalonic acid levels (see below).29 Fifty percent of patients with cobalamin levels between 100 and 200 pg/mL and up to 10% of patients between 200 and 300 pg/mL have elevated methylmalonic acid levels indicating cellular deficiency. Above 300 pg/mL, only 0.1% of patients have cellular cobalamin deficiency. Low serum cobalamin in patients with normal methylmalonic acid levels may indicate early depletion of cobalamin stores or reduced levels of transcobalamin I, the major cobalamin-binding protein in the plasma, which however does not play a role in hematopoietic cell utilization of cobalamin. Clinical conditions where myeloid cells, the major producer of transcobalamin I, are severely depleted such as in aplastic anemia may result in low serum cobalamin levels. Patients with myeloma and those with HIV infection frequently have unexplained low levels of cobalamin, possibly due to a reduced myeloid cell mass. Patients whose levels remain low after parenteral treatment with cyanocobalamin are likely to have genetic transcobalamin I deficiency. Elevated cobalamin levels are seen after treatment with parenteral vitamin B12, in patients with hepatic necrosis, or because of increased B12 binding proteins associated with myeloproliferative disorders, particularly untreated chronic myeloid leukemia.19
Table 27.2 Causes of Hyperferritinemia without Increased Serum Fe Saturation
Inflammation
Infection
Metabolic syndrome
Malignancy
Macrophage activation syndrome
Ferroportin mutations (classical type)
Hyperferritinemia-cataract syndrome ( L ferritin gene (FTL) mutation)
Parenteral iron treatment of iron-refractory iron deficiency anemia (TEMPRSS6 mutation)
Serum Methylmalonic Acid
Methylmalonyl dehydrogenase is a cobalamin-dependent enzyme required for transformation of methylmalonate into succinate in mammalian cells. Methylmalonic acid is assayed by gas-liquid chromatography or mass spectrometry.19 Serum and urine methylmalonic acid increase in more than 95% of patients with cellular cobalamin deficiency. Most patients with normal methylmalonic acid levels and low serum cobalamin in the absence of macrocytosis or anemia are presumed to have depleted stores without cellular cobalamin deficiency. In renal insufficiency, reduced methylmalonic acid excretion can lead to elevated serum levels in the absence of cellular cobalamin deficiency. A diagnosis of cellular cobalamin deficiency can be confirmed by demonstrating a decrease in serum methylmalonic acid levels to normal after initiation of cobalamin treatment.
Serum Homocysteine
Deficiency of folic acid, cobalamin, or pyridoxine (vitamin B6) prevents the methylation of homocysteine to form methionine and leads to increased serum homocysteine levels, which can be measured by a number of methods.19Cellular cobalamin deficiency usually causes elevated levels of both methylmalonic acid and serum homocysteine, but in 5% of cobalamin-deficient patients only serum homocysteine will be elevated. In usual practice, it is not cost effective to use homocysteine to confirm cellular deficiency of cobalamin or folate. Homocysteine levels are more often obtained for the clinical evaluation of arterial or venous hypercoagulability risk (see Chapter 22). Other causes of elevated levels of homocysteine include renal insufficiency and inherited abnormalities in the enzymes required for the folic acid cycle and sulfur-containing amino acid metabolism.
Serum Intrinsic Factor Antibody Assay
A positive result in this assay is highly specific for a diagnosis of malabsorption of cobalamin due to autoimmune depletion of intrinsic factor (pernicious anemia) but the sensitivity of the serum assay is less than 50%.
Serum and Red Cell Folate Assays
These levels are determined by a competitive receptor-binding assay.19 Serum folate levels reflect recent dietary intake while red cell folate levels reflect body folate stores at the time when the red cell was formed. Because cobalamin is required for cellular uptake of folate, reduced red cell folate levels are found with either folate or cobalamin deficiency; therefore, a serum cobalamin level is always required to interpret a low red cell folate level. Red cell folate is measured from a hemolysate prepared from whole blood and elevated levels of serum folate may therefore affect the red cell folate value. Elevated serum folate is found in cobalamin deficiency states and following treatment with folic acid. Testing for cobalamin levels is essential in the evaluation of patients with presumed folate deficiency, since folic acid replacement may improve the anemia but not the potentially irreversible neurologic complications of cobalamin deficiency. Folate deficiency in the United States has become rare, even in nutritionally deficient and alcoholic individuals, since cereal grain products were fortified with folate to reduce the risk of fetal neural tube defects. Therefore, since folic acid deficiency is infrequent in this country and so easily treated, it is likely that the assessment of serum and red cell folate levels in patients with anemia is no longer cost effective, and should be limited to those suspected of having malabsorption.
Serum Erythropoietin
In patients with refractory anemia due to marrow failure, markedly elevated erythropoietin levels (over 500–1,000 U/mL) usually predict failure of recombinant erythropoietin therapy. This immunoassay can therefore be useful in identifying a subset of patients unlikely to benefit from therapy with erythropoiesis-stimulating agents. On the other hand, it may not be worthwhile to assay erythropoietin levels in anemic patients with renal insufficiency, malignancy, or inflammation since erythropoietin levels in these conditions are commonly low (under 100 U/mL) and deficiency is treatable.
Erythropoietin assays which are sensitive to low levels are useful for distinguishing polycythemia vera from other causes of erythrocytosis. Patients with polycythemia vera have levels below the normal range, consistent with autonomous erythroid progenitor cell proliferation. Low erythropoietin levels may also be seen in rare patients with primary erythrocytosis due to genetic abnormalities in erythropoietin signaling pathway and in some patients in whom the cause cannot be identified (idiopathic erythrocytosis).30 Erythropoietin levels in patients with secondary polycythemia are frequently within the normal range but will often become elevated following phlebotomy treatment.
TESTS FOR THE EVALUATION OF ABNORMAL HEMOGLOBINS AND HEMOLYTIC ANEMIAS
Hemoglobin Electrophoresis
Historical methods used to differentiate and quantitate abnormal hemoglobins and the minor hemoglobins included alkaline and acid electrophoresis, and have been replaced by cation high- performance liquid chromatography (HPLC), isoelectric focusing (IEF) and capillary electrophoresis in large clinical laboratories.31 Smaller clinical laboratories may continue to use alkaline cellulose acetate or agarose gel electrophoresis at pH 8.6 to screen patient samples and to identify the common hemoglobins A, S, and C. When aberrant hemoglobins are observed, levels are quantified by using densitometry (Fig. 27.1). The minor hemoglobins F and A2 are also separated by this method, but their levels in adults cannot be measured accurately by densitometry. Since a number of G and D hemoglobins comigrate with Hb S, a solubility test is routinely used to confirm the presence of S (see below). Acid citrate agar electrophoresis at pH 6.0, also separates hemoglobins A, S, C, and F and is used routinely for confirmation of Hb S and C. It can distinguish the Hb D (containing mutated beta globin chains) and Hb G (containing mutated alpha globin chains) hemoglobins from Hb S. Acid citrate agar electrophoresis also separates Hb C from Hb E and Hb O Arab, variants that comigrate with C on alkaline electrophoresis and HLPC.31
IEF and HPLC are used in neonatal screening programs because they have greater power than do alkaline or acid agar electrophoresis in separating fetal hemoglobin from S and A; IEF and HPLC also are more expensive and require greater expertise for interpretation.19
Sickle Solubility Test
The insolubility of deoxygenated Hb S in a concentrated phosphate buffer can be exploited to confirm that a hemoglobin with the appropriate electrophoretic mobility is actually Hb S. Since the solubility test cannot distinguish between sickle trait and sickle cell disease, it is not useful to diagnose sickle cell disease in clinical settings.
FIGURE 27.1 Hemoglobin electrophoresis, cellulose acetate, pH 8.6.
Sickle Cell Prep
Red cells from sickle cell trait or homozygous S individuals will take on the sickle shape when deoxygenated. This test has been replaced by the sickle cell solubility test to confirm the presence of Hb S. It also does not distinguish sickle cell trait from sickle cell disease.
Hb A2 Quantitation
HPLC is the method of choice for this measurement.31 Column chromatography is also frequently used but may be unreliable in the presence of Hb S. Elevated levels of Hb A2 above 3.5%, will generally confirm a diagnosis of β thalassemia trait. The presence of iron deficiency may lower the Hb A2 level into the normal range. Patients with α or δβ thalassemia trait have normal levels of Hb A2. When it is clinically necessary to confirm alpha thalassemia, DNA-based methods are necessary, for example, in the case of a Southeast Asian couple with microcytosis who are at risk to have a fetus with hydrops fetalis or Hgb H disease.31
Hb F Quantitation
Many clinical laboratories continue to use the alkali-denaturation test to quantitate the percentage of Hb F. This test exploits the persistent solubility of Hb F under alkaline conditions, which precipitate most other hemoglobins. After treatment with alkali, the residual Hb F can be separated by filtration and quantified spectrophotometrically. The assay is accurate, with samples containing as much as 10% to 15% Hb F but will often underestimate higher levels. HPLC-based methods are more accurate for the assay of higher levels of Hb F.
Hb F Cells
Hb F can also be measured immunologically to determine the amount of Hb F in red cells and distinguish high Hb F containing cells (“F cells”) from red cells containing low levels of Hb F. A uniform distribution of Hb F is found in the red cells of patients with hereditary persistence of fetal hemoglobin (see Chapter 4).
Tests for Unstable Hemoglobins
Some unstable hemoglobins such as Hb Zurich and Hb Koln can be recognized by their propensity to precipitate when hemolysates are exposed to heat (50°C) or to 17% isopropanol. Unstable hemoglobins may also be detected by the formation of Heinz bodies (denatured hemoglobin) in intact red blood cells after exposure to oxidizing conditions. These inclusions, located near the red cell membrane, are detected microscopically by their blue staining after incubating red cells with supravital stains such as brilliant cresyl blue or new methylene blue (see Chapter 3).
Glucose-6-Phosphate Dehydrogenase
Either qualitative or quantitative tests are used in clinical laboratories to detect G6PD deficiency. The assays depend on the generation of NADPH from NADP. The most common screening test is the fluorescent spot test, which depends on the intrinsic fluorescence of NADPH. Reticulocytes from individuals with the most common variant of G6PD deficiency seen in the United States (G6PD A-) have much higher quantities of enzyme than are present in mature red cells. Deficiency may not be diagnosed by these tests in G6PD A- individuals if reticulocytosis has developed in response to hemolysis produced by an oxidant chemical or drug. Heinz bodies (denatured hemoglobin, see above) may also be recognized in the red cells in this setting. Female heterozygotes may also escape detection by screening tests.32
Serum Haptoglobin
This hemoglobin-binding protein can be assayed by nephelometric or turbidometric methods.19 Haptoglobin is an acute phase reactant, but its main usefulness is that very low levels are an indicator of acute or chronic hemolysis. Free hemoglobin bound to haptoglobin is cleared by the reticuloendothelial system in less than 30 minutes. In vivo hemolysis of as little as 50 mL of red blood cells will deplete the blood of haptoglobin. In the absence of continuing hemolysis, it will take at least 5 days to regenerate to normal levels. The normal range varies widely due to genetic differences in the alpha chains. Rare patients have very low haptoglobin levels on a genetic basis. Patients with severe liver disease may have decreased haptoglobin levels due to failure of hepatic synthesis.
Urine Hemosiderin
In patients with chronic intravascular hemolysis such as paroxysmal nocturnal hemoglobinuria or cardiac valve hemolysis, renal excretion of hemoglobin leads to uptake of heme, with subsequent accumulation of hemosiderin in the renal tubular cells. After staining the urine sediment with Prussian blue, microscopic evaluation of the sediment will demonstrate blue stained particles in renal casts indicating iron deposition in the tubular cells.
HEMOSTASIS AND COAGULATION ASSAYS
Activated Partial Thromboplastin Time
This assay measures the time required to initiate clotting after citrated plasma is incubated with calcium, a partial thromboplastin (a phospholipid source devoid of tissue factor) and a surface activating agent. Automated instruments detect clot initiation mechanically or based on turbidometric changes.33 The activated partial thromboplastin time (aPTT) is particularly sensitive to deficiencies of factors VIII and IX, but will also be prolonged by less common deficiencies of factors XI, X, V, prothrombin, and fibrinogen. Mild reductions in factor VIII (>30%) and fibrinogen (>100 mg%), however, may not be detected by this assay.33 Very rare deficiencies of factors not associated with bleeding, Factor XII, prekallikrein, and high molecular weight will also prolong the aPTT.
The aPTT is also commonly prolonged by anticoagulants (antithrombins such as heparin, hirudin, argatroban, bivalirudin, dabigatran), lupus anticoagulants and less commonly factor-specific antibodies (usually against factor VIII; antibodies against other factors are rare). Persistence of a prolonged aPTT in 1:1 mixing studies (50% patient plasma mixed with 50% normal plasma) suggests the presence of an antibody. Detection of antibodies against factor VIII may require incubation of the 1:1 mix with normal plasma for 1 hour at 37° to allow antibody to bind to Factor VIII. The presence of lupus anticoagulants can be confirmed by demonstrating correction of a prolonged aPTT by the addition of phospholipid (see below). Pan-inhibitors may rarely occur in patients with Waldenström macroglobulinemia or multiple myeloma due to M paraproteins inactivating multiple clotting factors. A rare cause of acquired severe factor IX and X and V deficiencies is absorption of the clotting factors on amyloid deposits of primary systemic amyloidosis.33
Unfractionated heparin therapy is most frequently monitored using the aPTT to avoid subtherapeutic or supratherapeutic levels. A recurring problem is the variability of the sensitivity of the many available partial thromboplastins to heparin as well as differing results depending on the instruments used. A standard curve must be generated to compare PTT levels with heparin levels as measured by an anti-Xa assay in a group of patients receiving full dose heparin therapy, in order to confirm the therapeutic range of any new aPTT reagent and the instrument used. After testing a number of aPTT reagents in this manner, one study suggested that aPTT test to control ratio of 2.0 to 3.0 is generally a good target range for therapeutic heparin levels.34 Due to concern regarding the variability of the sensitivity of aPTT reagents, many laboratories now monitor heparin levels solely by its capacity to inhibit anti-Xa. The aPTT is not sensitive to low molecular weight heparins (LMWH). When monitoring of LMWH is indicated, as in patients with renal insufficiency, obesity, pregnancy, or prior to elective surgery, an anti-Xa assay for the specific LMWH must be used.
Spurious results of the aPTT are usually due to poorly filled tubes, low plasma to citrate ratios due to high hematocrits, delays in delivery of samples to the laboratory, or contamination with intravenous fluids or heparin.
Prothrombin Time
The prothrombin time (PT) utilizes a more powerful thromboplastin to detect deficiencies of factors VII, V, X, prothrombin, and fibrinogen,33 and therefore it is valuable in assessing liver function and monitoring warfarin therapy which lowers the Vitamin K dependent factors II, VII, IX, and X, which are synthesized by the liver. Inherited deficiencies of these factors and autoantibodies are rare. Inherited factor VII deficiency, the most common, will be recognized by the isolated prothrombin time prolongation. Deficiencies of factors X, V, prothrombin, and fibrinogen will prolong both the aPTT and the PT.
Some lupus anticoagulants may affect the PT as well as the aPTT; in rare instances, anti-prothrombin antibodies develop, which clear prothrombin from the circulation, causing severe deficiency and bleeding. Also rarely, exposure to bovine thrombin will trigger the development of anti-thrombin and anti-factor V antibodies.
The international normalized ratio (INR) has been useful in standardizing the control of warfarin therapy.33 The INR is the ratio of the patient PT to the mean normal PT raised to the power of the International Sensitivity Index (ISI) of the thromboplastin reagent utilized. Commercial thromboplastins are calibrated and given an ISI value, which reflects their sensitivity to warfarinized plasma. It may be misleading to use the INR to describe the prolongation of the PT in a patient not receiving warfarin, particularly in patients with liver disease.33 The PT is vulnerable to the same preanalytic artifacts described above for the aPTT assay.
Activated Clotting Times
Activated clotting times (ACT) are used on site for cardiac surgery and cardiac catheterization procedures.
Thrombin Time
This assay measures the time to initiate clotting by low concentrations of thrombin, which release only fibrinopeptide A and B. The thrombin time is prolonged by low fibrinogen levels or the presence of heparin, paraproteins, dysfibrinogens, or fibrin(ogen) split products.33 The reptilase time is prolonged by similar molecules but is insensitive to heparin, since only fibrinopeptide A is released from soluble fibrinogen. Dysfibrinogenemia may be confirmed by comparing the antigenic level of fibrinogen with that determined by the thrombin time-based assay. It can be inherited or acquired usually secondary to liver disease.33
Specific Factor Assays
Most specific factor assays are based on the ability of patient plasma to correct clotting times of specific factor-deficient plasma in PTT or PT-based assays (see Chapter 20). Deficiency of factor XIII, which is important to cross-link fibrin monomers, thereby increasing stability of the clot is not associated with abnormalities of aPTT, PT, or thrombin time. Factor XIII is measured immunologically or by the solubility of the clot formed in urea.33
Fibrinogen
Fibrinogen levels are routinely determined by using a thrombin time-based assay using high concentrations of thrombin, but chemical or immunologic methods may also be used.33 As fibrinogen is an acute phase reactant, levels are frequently raised in patients with inflammation and malignancy. Decreased levels are found with disseminated intravascular coagulation, the hemophagocytic syndrome, advanced liver disease, treatment with asparaginase, or rarely as an inherited condition.
Euglobulin Clot Lysis
Euglobulin, a plasma precipitate containing fibrinogen, plasminogen, and plasminogen inhibitor without most fibrinolysis inhibitors, is prepared from the patient sample, and clotted with thrombin; time required for clot lysis is then determined. Because euglobulins lack inhibitors, clot lysis normally occurs quite rapidly (within 90–300 minutes). Abnormally short lysis times occur in states of hyperfibrinolysis such as severe liver disease, but may also simply reflect poor clot formation due to hypofibrinogenemia.
D-Dimer
This immunologic assay detects fibrin split products that are cross-linked because of the action of thrombin and factor XIIIa. Elevated levels indicate extensive local fibrin formation (deep vein thrombosis, pulmonary emboli, pneumonia, postoperative states) or disseminated intravascular coagulation. The value of a positive value in predicting localized thrombosis is poor, particularly in the presence of comorbid conditions such as infection, inflammation, or malignancy. A negative high sensitivity or moderately sensitive D-dimer assay is more helpful in ruling out thrombosis.35,36 A positive D-dimer level has been also used to predict recurrence of thrombosis after discontinuing treatment of an initial unprovoked thrombosis,37 but factors such as age and sex are also important to consider along with the D-dimer level.38
Tests of Platelet Function
Mucocutaneous bleeding in the presence of a normal platelet count may indicate platelet dysfunction. Platelet aggregation in response to epinephrine, ADP, and collagen assessed by light transmission or impedance aggregometry is the most effective method of testing for platelet dysfunction. The availability of aggregometry assays is often limited. They require the patient to be present at the testing laboratory, since platelet-rich plasma or whole blood must be tested rapidly after venipuncture in order to prevent in vitro platelet activation and spurious results.
The bleeding time was frequently used in the past to screen for suspected platelet dysfunction. In the modified Ivy technique, a template is used to make two incisions on the volar surface of the forearm parallel to the antecubital fold, while a blood pressure cuff applied to the upper arm is inflated to 40 mm Hg. Blood is gently removed from the incisions every 30 seconds. The technique is affected by the depth of the incision, the skill of the technologist, and also the skin characteristics of the patient. For example, it may be abnormal in patients with Ehler Danlos disease, osteogenesis imperfecta, and scurvy.33Therefore, its reproducibility is a problem; nor has it been shown to be useful as a general screen for hemorrhagic risk.
In the last decade, automated instruments have been developed that may be more reliably used as a screening test for platelet function abnormalities.33 Many of these instruments appear to be most useful in detecting aspirin use and some have been designed specifically for monitoring antiplatelet therapy, including aspirin and clopidogrel.39 The PFA-100, the earliest instrument and the subject of most studies, measures the time that platelets aspirated at high shear rate through an aperture of a membrane coated with collagen and either epinephrine or ADP aggregate and obstruct further flow (the “closure” time). The PFA-100 is useful in screening for moderate and severe von Willebrand disease (VWD) and monitoring the effect of desmopressin therapy,40 although its sensitivity and specificity in screening for milder VWD and rarer platelet secretion and storage function defects has been questioned.41,42 Abnormal findings need to be confirmed by specific testing for VWD disease (VWD antigen, ristocetin cofactor assay, and multimer analysis) or by platelet aggregometry or other specific assays.41,42 Of note, once medications have been excluded as the cause of closure time prolongation, 90% of abnormalities will be due to VWD and 10% due to other platelet defects.
The bleeding time or the automated instruments may also be of use in the initial screening of patients requiring an invasive procedure who have acquired disorders that affect platelet function such as severe renal insufficiency, myeloproliferative disorders, or drug-induced platelet dysfunction.33
TESTS FOR HYPERCOAGULABILITY
Antithrombin, Protein C, and Protein S
Functional assays of these proteins are more sensitive to deficiency than are antigen-based assays. Patients heterozygous for an inherited defect in one of these factors typically have only modest reductions from the lower limit of normal in factor levels. Nevertheless, they may be at significant risk of venous hypercoagulability (particularly in patients with antithrombin deficiency). Acquired deficiency of antithrombin occurs in disseminated intravascular coagulation, liver disease, heparin therapy, and extensive thrombosis. Acquired deficiency of proteins C and S occurs with vitamin K deficiency, warfarin therapy, and extensive thrombosis. Free protein S deficiency occurs in patients with increased C4b binding protein secondary to inflammation. These assays should not be performed in the setting of acute venous thromboembolism.
Activated Protein C Resistance
Abnormality of this aPTT-based assay is mainly associated with the inherited polymorphism of factor V Leiden. This assay has been replaced by molecular assays for factor V Leiden.
Lupus Anticoagulants
These acquired antibodies to β2-glycoprotein-1, a protein with a high affinity for phospholipid, can be associated with both arterial and venous hypercoagulability. On the other hand, positive assays are frequently seen in otherwise healthy individuals and may be transient. The functional assays, the dilute Russell viper venom test (DRVTT) and low phospholipid aPTT-based tests (with confirmation by high phospholipid or platelet neutralization tests), may be more sensitive to thrombosis risk than are the serologic tests for anticardiolipin antibodies. Russell viper venom directly activates factor X, bypassing factor VII and the earlier factors in the intrinsic pathway (see Chapters 20 and 22).
Factor V Leiden and Prothrombin G 20210A
The molecular assays for these common mutations, essentially polymorphisms, are not affected by the presence of acute venous thromboembolism.
TESTS FOR EVALUATION OF PATIENTS WITH HEMATOLOGIC MALIGNANCIES
Serum Protein Electrophoresis
Electrophoresis separates proteins mainly on the basis of electric charge, which can be performed by agarose gel electrophoresis or the automated capillary zone electrophoresis.19 When normal plasma proteins are electrophoresed on agarose films and stained with a protein-binding dye, six zones appear. Albumin, α1, α2, and two β globulins appear as 5 discrete bands, and γ globulins migrate as a more diffuse, electrophoretically heterogeneous band(Fig. 27.2 and Table 27.3). Monoclonal immunoglobulins (M paraproteins) are recognized because they form discrete bands in the β or γ globulin regions. The concentration of the proteins in normal or the paraprotein bands (“spikes”) can be determined by densitometry. When hyperglobulinemia is detected by routine total protein and albumin measurements, protein electrophoresis is essential to distinguish between reactive polyclonal from monoclonal processes. In order to confirm monoclonality, immunofixation studies are required to demonstrate that the suspected monoclonal band contains only a single heavy chain and a single light chain (Fig. 27.3). The capillary zone electrophoresis is usually more sensitive than agarose gel electrophoresis, particularly for small paraproteins, but false negative results have been reported.43 Poor correlation with nephelometric immunoglobulin quantitation tests may alert to false negative results in capillary or agar electrophoresis when monoclonal proteins have atypical mobility due to temperature, or pH, or unknown factors. Falsepositive artifacts resembling spikes are more frequent in agarose electrophoresis, which will require immunofixation to be recognized.
The presence of a monoclonal immunoglobulin is consistent with, but not diagnostic of, the presence of a plasma cell or lymphoid malignancy or light-chain amyloidosis. Most monoclonal proteins with serum concentrations below 3 g/dL are not associated with clinical or pathologic evidence of malignancy, and they are referred to as monoclonal gammopathy of unknown significance (MGUS), because of the potential for future malignant transformation of the plasma cell clone. Occasional individuals exhibit two monoclonal proteins (“diclonal” gammopathy), which may represent the products of two separate clones. If both bands contain the same heavy and light chain, the two immunoglobulins may originate from a single clone despite differing electrophoretic mobilities, possibly due to multimer formation. Patients with marked polyclonal gammopathy, for example due to HIV infection or liver disease, may have multiple small discrete bands termed oligoclonal gammopathy. The majority of the γ globulin in these individuals is polyclonal.
FIGURE 27.2 Serum protein electrophoresis.
M paraproteins of greater than 3.0 g/dL usually reflect the presence of IgG or IgA multiple myeloma or Waldenström macroglobulinemia. The concentration of the M protein in the plasma (or urine) is a marker of tumor burden, and serial monitoring by electrophoresis is extremely important in assessing response to therapy. For this determination to be reliable, the M paraprotein must be quantitated separately from polyclonal immunoglobulins.
Table 27.3 Plasma Protein Migration Patterns on Standard Protein Electrophoresis
Albumin Zone
Albumin
Alpha1 Zone
Alpha1-antitrypsin
Alpha1-lipoproteins (high density lipoproteins HDL)
Alpha2 Zone
Alpha2-macroglobulin
Haptoglobin
Ceruloplasmin
Beta Zone
β-Lipoprotein (low density lipoprotein)
Transferrin
C3 (complement)
Gamma Zone
Fibrinogen (in incompletely clotted specimens)
IgA
IgM
IgG
FIGURE 27.3 Immunofixation illustrating an IgG kappa monoclonal protein.The arrowhead points to the putative monoclonal band in the serum protein electrophoresis (SP lane) stained with a protein-binding dye. Staining in the remaining lanes indicates the presence of products reacting with specific antibodies directed against IgG (G), IgA (A), and IgM (M) heavy chains, and Ig kappa (κ) and lambda (λ) light chains. Small arrow on the right indicates origin.
Urine Protein Electrophoresis
Urinary light-chain excretion must be evaluated in patients with hypogammaglobulinemia or other findings suspicious of a plasma cell dyscrasia. Approximately 15% of myeloma patients excrete monoclonal light chains in the urine (Bence Jones proteinuria) in the absence of any detectable M protein in the serum. For screening, a random urine specimen is concentrated before electrophoresis. Discrete bands are then assayed by immunofixation to confirm whether they represent intact monoclonal immunoglobulins (due to “overflow” of the serum M protein) or free light chains. If a paraprotein is present, serial 24-hour urine collections are useful in monitoring tumor burden and response to therapy.
Immunofixation
Immunofixation replaced immunoelectrophoresis to confirm monoclonality of discrete bands noted in the β or γ globulin regions of the protein electrophoresis patterns. Antibodies against the γ, α, μ heavy chains, and κ and λ light chains are layered separately on membranes containing the electrophoresed samples. A monoclonal immunoglobulin will form immunofixation bands with antibodies against one heavy chain class and/or one light-chain type (Fig. 27.3). IgM and IgA proteins are more likely to be found close to the β globulin region; IgG proteins may be located in any area of the β and γ globulin zones. Free light chains are seen in the serum protein electrophoresis only in myeloma patients with severe renal failure or in instances where the light chains form spontaneous tetramers too large for renal clearance. The uncommon IgD and very rare IgE myelomas should be suspected when a serum paraprotein binds only an anti-light-chain antibody.
Serum Free Light Chains
This nephelometric automated immunoassay depends on the use of antibodies to κ and λ light chains that react with epitopes that are exposed in free light chains but hidden in the intact monoclonal immunoglobulins. Because of its superior convenience and comparable sensitivity to urine electrophoresis, it is being used frequently to monitor response to treatment of light-chain myeloma and amyloidosis.44,45 The assay also has diagnostic and prognostic utility in amyloidosis, monoclonal gammopathy, and smoldering multiple myeloma.46 The concentration of serum free light chain over time have been noted to have a greater variability than M paraproteins do for monitoring of stable monoclonal gammopathy patients and therefore are less useful in detecting progression in these patients.47
Quantitative Serum Immunoglobulins
Quantitative serum immunoglobulins are measured with automated nephelometric or turbidometric immunoassays; these assays are most useful for quantitating normal immunoglobulins rather than to detect M paraproteins. For example, IgA M proteins that form multimers and monoclonal IgM pentamers with a propensity to disassociate into smaller molecular weight species can produce erroneous results using this assay, since the immunoglobulin quantitation can be affected by changes in the molecular weight.
Serum Cryoglobulins
The most critical step in this test is the treatment of the specimen before it reaches the laboratory. The blood should be drawn into a preheated syringe or warm tube, transported to the laboratory at 37°C and kept at this temperature until the serum is separated from the clot. The serum is then refrigerated at 4°C and examined after 24 hours. A precipitate that dissolves when the tube is rewarmed to 37°C indicates the presence of a serum cryoglobulin. Electrophoresis and immunofixation of the separated and redissolved cryoprecipitate will reveal the immunoglobulins involved in the cryoglobulin formation. Cryoglobulinemia may due to (i) a monoclonal immunoglobulin, usually IgM, (ii) a monoclonal IgM with rheumatoid activity binding to polyclonal IgG (“mixed cryoglobulinemia”), or (iii) polyclonal IgM bound to polyclonal IgG. Mixed cryoglobulinemia (ii and iii) are associated with a variety of lymphoproliferative or autoimmune disorders and infections, particularly hepatitis C.48
Serum Viscosity
The Ostwald viscosimeter measures viscosity by comparing the time required for serum and water to flow through a capillary tube at 37°C. Normal serum is 1.4 to 1.8 times more viscous than water. Symptoms resulting from hyperviscosity generally occur when the relative viscosity exceeds 6 but may occur with a relative viscosity as low as 3 or 4.
Serum β2-Microglobulin
Serum β2-microglobulin is a small protein noncovalently linked with class I human leukocyte antigen (HLA) molecules. Serum levels are measured by nephelometric immunoassay. Elevated levels may be found in inflammatory conditions, renal failure (due to failure of excretion) and in lymphoid and plasma cell malignancies. High serum β2-microglobulin is an important prognostic feature indicating advanced disease and poor response to therapy in myeloma and certain lymphomas. In myeloma, elevated β2-microglobulin levels predict early relapse following autologous transplantation. Serum levels are not useful however, to follow chemotherapy responsiveness in myeloma because of the lack of specificity.
Serum Lactic Dehydrogenase
This protein is measured by an automated enzyme activity-based assay. The normal range varies depending on the type of assay. Elevated serum levels reflect necrosis of cells rich in the enzyme. The most marked elevations (more than 5 times normal) usually are noted with severe megaloblastic anemia, intravascular hemolytic anemia, or hemophagocytic syndromes. Similar levels can be seen in acute leukemia and lymphomas often as harbingers of tumor lysis syndrome. An elevated serum level is a risk factor in the International Prognostic Index for non-Hodgkin lymphoma. Distinguishing the five isoenzymes of lactic dehydrogenase is seldom useful in diagnosing hematologic disorders.
Serum Uric Acid
Serum uric acid is measured by an automated enzyme assay. Elevated serum levels particularly over 10 mg/dL in patients with hematologic malignancies should raise concern for increased cell turnover (tumor lysis syndrome) and incipient renal compromise due to uric acid precipitation. Hypouricemia may occur in Hodgkin lymphoma. Bence Jones (light chain) proteinuria rarely can cause an interstitial nephritis with a proximal tubular defect resulting in hypouricemia due to marked hyperuricosuria along with other features including polyuria, glycosuria, and aminoaciduria, known as the Fanconi syndrome.
NEUTROPHIL EVALUATION
Marrow Granulocyte Reserve
Marrow granulocyte reserve may be assessed by the hydrocortisone stimulation test49 or by the response to filgrastim.50 In the hydrocortisone stimulation test, the absolute neutrophil concentration is measured before administering 200 mg hydrocortisone intravenously and again 3, 4, and 5 hours later. A failure of the neutrophil concentration to increase by at least 1,600 neutrophils per microliter indicates poor marrow granulocyte reserve. Similarly, failure of the absolute granulocyte level to rise above 5,000/μL 24 hours after a subcutaneous injection of 5 mcg/kg filgrastim indicates increased risk for chemotherapy-induced febrile neutropenia.
Leukocyte Alkaline Phosphatase
This assay is an inexpensive cytochemical test of peripheral blood neutrophils used to screen patients with leukocytosis suggesting a diagnosis of chronic myelogenous leukemia. The assay depends on the enzyme’s ability to cleave a dye, which then stains the cells. Individual neutrophils are scored by the intensity of staining as 0 to 4+, and the sum of the scores of 100 cells are tallied. Neutrophils from patients with chronic myelogenous leukemia or paroxysmal nocturnal hemoglobinuria have low levels of leukocyte alkaline phosphatase (<10 Kaplow units). Neutrophils from patients with reactive leukocytosis and polycythemia vera have elevated scores (>80 Kaplow units). Patients with other myeloproliferative disorders may have normal, low, or high levels. This assay has been largely replaced by the fluorescence in situ hybridization (FISH) assay for BCR-ABL in peripheral blood neutrophils or cytogenetic analysis of the bone marrow. The availability of highly effective therapy for chronic myelogenous leukemia warrants the use of the more specific, albeit expensive FISH assay51 (see Chapter 13).
References