Published in the Bulletin of Experimental Treatments for AIDS December 1996 issue, by the San Francisco AIDS Foundation.

December 1996 Table of Contents

Main Page

beta@sfaf.org

Blood Tests

by Mark Bowers

Blood tests are crucially important in diagnosing HIV disease, predicting disease progression, assessing response to therapy and identifying opportunistic pathogens. Most people have only a basic idea about the nature and purpose of HIV tests, complete blood counts (CBC), chemistry (chem) panels, CD4 and CD8 cell counts and percentages, and viral load tests. Some clinicians and people with HIV share some confusion and uncertainty about the exact benefits of viral load testing and its relationship to CD4 cell counts. Staff of AIDS service organizations report that their clients frequently ask them to interpret their blood work. The following is a short explanation of selected blood tests, and a description of current controversies where they apply to HIV disease.

HIV Serology

Establishing HIV serology, or evidence of HIV infection in the blood, is recommended for several categories of people, including:

• people with other sexually transmitted diseases (STD) LI> people in "high risk" categories, including men who have sex with men, injection drug users, and men and women who have had unprotected sex outside of a mutually monogamous relationship, as well as sexual partners of individuals in any of these groups
  
  
• hemophiliacs and persons who received blood products or transfusions or tissue transplants between 1977 and May 1985, as well as their sexual partners
  
  
• people who have or have had sex with multiple partners
  
  
• persons who think they are at risk
  
  
• pregnant women (counseling and voluntary testing are recommended by the Centers for Disease Control and Prevention [CDC], the American Academy of Pediatrics and the San Francisco AIDS Foundation)
  
  
• people with active tuberculosis
  
  
• people with occupational or other exposure to body fluids from an HIV positive person, including blood, semen, vaginal secretions, cerebrospinal fluid, synovial (joint) fluid, peritoneal (abdominal cavity) fluid, pleural (lung) fluid, pericardial (around the heart) fluid and amniotic (within the placental sac enclosing a fetus) fluid
  
  
• persons aged 15-54 who are admitted to a hospital where the seroprevalence rate exceeds 1% or where AIDS case rates exceed 1/1,000 discharges
  
  
• healthcare workers who perform invasive procedures that could expose them to HIV
  
  
• blood, semen, bone marrow and organ donors (mandatory in all states)
  
  
• persons whose medical condition is consistent with HIV disease.

In the United States, 41 states require informed consent before an HIV test can be performed. Twenty-two states allow HIV tests without informed consent for patients who are the source of occupational exposures to healthcare workers. All states require tests for semen, blood, bone marrow and organ donors. Anonymous testing is available in some areas, including major California cities.

HIV-1 ELISA (enzyme-linked immunosorbent assay) is usually the first test performed on potentially HIV-infected blood. Results may be positive or negative. ELISA is a sensitive but non-specific test; therefore, a positive ELISA must be confirmed by a positive Western Blot test, which is more specific but less sensitive. Both ELISA and Western Blot tests measure the presence of antibodies against HIV, not the virus itself. After exposure to HIV, a period of time (known as the "window period") elapses before a measureable amount of antibody is produced (seroconversion); this period rarely lasts longer than 6 months. False-negative results may occur during the window period. HIV-1 tests will detect antibodies to most variants of HIV-1 except subtype O, a few cases of which have been recently seen in the U.S. ELISA will detect 80% of people with HIV-2 infection, but Western Blot test results will be indeterminate. HIV-2 tests and a combined HIV-1/HIV-2 test are available in the U.S. Blood centers in the U.S. currently use the combined HIV-1/HIV-2 antibody test to screen all donated blood.

False-positive tests are most frequent among volunteers in vaccine studies. These volunteers develop antibodies that are similar to those produced by people with natural HIV infection, although they do not have HIV in their bodies. Because ELISA detects these antibodies, it cannot distinguish between people with HIV infection and HIV vaccine volunteers. Otherwise, the rate of false-positive results is 1/135,000. Indeterminate results usually mean that an ELISA test was positive and the Western Blot confirmatory test showed only one band. An indeterminate result may occur when a person is tested around the time of seroconversion (indicating a recent exposure), when antibodies are decreased due to advanced HIV infection, when the person has had a recent blood transfusion or organ transplant, or when the person has collagen-vascular disease, autoimmune disease or cancer. People with HIV-2 and participants in HIV vaccine studies also may have indeterminate Western Blot test results.

Anonymous home testing kits are now approved by the Food and Drug Administration (FDA). The first approved kit, Confide, includes an over-the-counter home blood collection kit, HIV-1 antibody testing at a certified laboratory and a test result center that provides both results and counseling over the phone. Results are obtained in 7 days by contacting a toll-free number (1-800-THE-TEST). Two HIV testing and counseling systems are available from Home Access Health Corporation. Home Access Express and Home Access allow people to collect blood samples, access pre-test and post-test counseling, and receive test results in either 3 days or 7 days, depending on the system purchased (call 1-800-HIV-TEST).

A salivary test for HIV-1, OraSure, was FDA-approved in December 1994. The sensitivity of the test mirrors that of ELISA. The test sample is easier to collect, cheaper to process and better received by patients than the standard blood test. A urine test for HIV antibodies made by CalType Biomedical was FDA-approved in August 1996. The false-positive and false-negative rates for this test are about 1 or 2 in every 100 tests. This test may only be administered by physicians, and positive test results must be confirmed by a Western Blot test. Documentation of pre-test counseling is needed before any sample will be evaluated.

Initial or Baseline Laboratory Studies

The following are recommendations for baseline blood testing of HIV positive individuals. Recommendations are drawn from the AIDS Care Program of the Johns Hopkins Medical Institutions.

Baseline Blood Tests Used by the AIDS Care Program of the Johns Hopkins Medical Institutions

• CBC with differential
• Chemistry panel (SMA12, 14 or 20)
• VDRL (syphilis serology)
• Hepatitis B virus (HBV) serology
• Hepatitis C virus (HCV) serology
• G6PD (enzyme)
• CD4/CD8 lymphocyte subsets
• HIV viral load
• Toxoplasma serology (IgG antibody)

Johns Hopkins Researchers Recommend against:

• neopterin
• beta-2 microglobulin
• p24 antigen
• erythrocyte sedimentation rate

because "their use in individual patients is disputed, and there are no specific guidelines for management decisions as there are with CD4 cell counts..."

• Herpes simplex virus (HSV) antibodies
• Cytomegalovirus (CMV) antibodies

since "50-90% of healthy adults have serologic evidence of prior infection and these diagnoses require clinical observation plus microbial detection."

• Cryptococcal antigen assay

because it "is not useful for routine screening, but is often advocated when cryptococcal meningitis is a diagnostic consideration."

Mary Romeyn, MD, a San Francisco physician and author of Nutrition and HIV: A New Model for Treatment, also annually measures levels of vitamin B12, folate, magnesium and testosterone (in males and females), and recommends supplements if deficiencies are found. Romeyn also emphasizes the importance of monitoring triglycerides (fats) to detect underlying infections or the beginning of the wasting process. Triglycerides are part of most routine chemistry panels.

CBC with Differential

Complete blood counts are taken for almost everyone with a major illness because the CBC reveals anemia (low red blood cell count), leukopenia (low white cell count) and thrombocytopenia (low platelet count). As many as 30-40% of people with AIDS may have one or more low counts. The CBC should be repeated at 6-month intervals and after the initiation of any drug known to lower red blood cell (RBC) counts, white blood cell (WBC) counts or platelet counts. The CBC is used to determine CD4 and CD8 cell counts and ratios. The CBC also includes hematocrit, hemoglobin and RBC morphology.

Hematocrit (Hct) is the percentage of RBC relative to plasma volume, and is an indirect measure of the oxygen-carrying capacity of the blood. Normal values are 0.42-0.54 for adult males and 0.37-0.47 for adult females, meaning that 37-54% of the whole blood is RBC.

Hemoglobin (Hgb) is a protein molecule in RBC that helps them transport oxygen from the lungs to tissues and organs. This protein also helps the RBC carry off excess carbon dioxide. When Hgb measures are low, tissues may not be receiving enough oxygen, leading to poor healing and less efficient organ function. Normal adult Hgb values are 13.5-18.0 grams/deciliter (g/dL) for males and 12.0-16.0 g/dL for females. Normal Hct values are about 3 times Hgb values, so there is little need to remember or follow both.

Red blood cell morphology relates to the shape of the cells. Normal RBC are flattened, with concave surfaces on 2 sides. Other shapes may have difficulty transporting oxygen or travelling through capillary vessels, which are only slightly wider than the average RBC.

WBC counts include CD4 and CD8 cell counts, the percentages of each compared to the total number of lymphocytes, and values for neutrophils, lymphocytes, monocytes, eosinophils and basophils, types of WBC. All of these cells are part of the immune system. They must be able to move through pores in capillaries to reach their target antigens (e.g., foreign organisms, toxins, allergens). The normal adult white cell count is 4.5-11x109 cells/liter. During an infection, this number may increase.

Eosinophils, neutrophils and basophils are collectively called granulocytes. They react to allergens and infectious agents, cause inflammation, and bind to antigens or engulf and destroy (phagocytose) them. Eosinophil counts are useful in detecting allergic reactions or the presence of intestinal parasites.

Neutropenia is a reduction in the number of neutrophils, and is often life-threatening because it may lead to bacterial and fungal infections. Mild neutropenia is characterized by a neutrophil count of 1,000-2,000 cells/microliter (mcrL); moderate neutropenia, 500-1,000 cells/mcrL; and severe neutropenia, less than 500 cells/mcrL. Symptoms include fever, chills and ulcers in the mouth or on other mucous membranes. A common cause of neutropenia is prescription drug use, including penicillins, sulfonamides, AZT, ganciclovir, hydroxyurea and some cancer drugs. Another cause may be an underlying infection that invades or damages the bone marrow. The most common treatment for neutropenia is the recombinant growth factor granulocyte colony-stimulating factor (G-CSF, filgrastim, Neupogen). G-CSF stimulates production of new neutrophils, helping to control bacterial infections.

Bands are immature neutrophils that are released into the blood when the body needs more infection control than mature neutrophils can provide. An elevated number of bands suggests an acute infection.

Absolute neutrophil counts (ANC) are calculated by adding the percentage of neutrophils to the percentage of bands and multiplying by the total number of WBC. Recent research by Mark Jacobson, MD, at San Francisco General Hospital showed a rapid increase in secondary infections when absolute neutrophil counts fall below 999 cells/mcrL. However, many insurance companies still believe 500-700 cells/mcrL is sufficient protection, and will not reimburse for G-CSF until the ANC falls below 500 cells/mcrL.

Monocytes are immature macrophages, and are a target for HIV infection. Monocytes and macrophages are antigen-presenting cells (APC) that show pieces of antigen to CD4 and CD8 cells for recognition and destruction. Macrophages are present in most tissues and can cross physiologic barriers, including the blood-brain barrier, that prevent the entry of most other cells. HIV-infected macrophages, like the Trojan horse, can help seed HIV throughout the body. The recombinant growth factor granulocyte macrophage colony-stimulating factor (GM-CSF, sargramostim, Leukine) is used to increase the number of granulocytes and macrophages in the body.

Lymphocytes include T-cells and B-cells. T-cells are divided into CD4 and CD8 cells, both of which have learned to recognize antigens and distinguish them from tissues that belong to the body. They recognize an antigen only when it is shown to them by an APC. This is why the immunity they provide is called cell-mediated immunity. B-cells recognize antigens directly or with the help of T-cells. They produce and release antibodies. Antibody production is called humoral immunity.

A low platelet (thrombocyte) count -- less than 150x109/liter -- is referred to as thrombocytopenia. People with this condition bleed more than others because their blood cannot form clots. Some drugs may cause thrombocytopenia, including trimethoprim-sulfamethoxazole (TMP-SMX, Bactrim or Septra), amphotericin B, pyrimethamine and ketoconazole. A platelet transfusion is the usual treatment for thrombocytopenia.

CD4/CD8

T-cells are marked with specific molecules on their surface (called cluster of differentiation, or CD) that identify their immune function. All T-cells are marked with CD3, while those marked with CD4 are considered helper-inducer cells and those marked with CD8 are cytotoxic T-lymphocytes (CTL, or killer T-cells) or suppressor cells. CD4 cells are infected with HIV when HIV binds both to the CD4 molecule and a molecule called fusin. CD8 cells also may be infected with HIV.

CD4 cell counts have been the basis for staging HIV disease, initiating antiretroviral therapy (below 500 cells/mm³), establishing an AIDS diagnosis (below 200 cells/mm³), deciding on prophylaxis against selected opportunistic infections (e.g., start TMP-SMX below 200 cells/mm³) and making treatment changes. Normal adult values for CD4 cell counts range from 800 to 1,050 cells/mm³ (values may vary from laboratory to laboratory). CD4 cells are targets of HIV infection, and a continuing decline in their number is known to correlate with increased susceptibility to infections that are normally controlled by the immune system. Sufficient destruction of CD4 cells over time removes an important immune element from the body's repertoire of responses, and eventually results in ineffective management of infection.

With the development and increasing acceptance of viral load measurements to count viral particles in the blood, the International AIDS Society-USA, the San Francisco Department of Public Health and many other health care organizations include consideration of viral load test results alongside CD4 cell counts in initiating and changing antiretroviral therapy. Under the new paradigm, the CD4 cell count remains useful for deciding when to initiate prophylaxis for opportunistic infections. Low CD4 cell counts have predicted poor clinical outcomes.

A current controversy about CD4 cell counts was sparked by the recent finding that an HIV-uninfected person who received 4 weeks of antiretroviral treatment with AZT had a 30% increase in CD4 cells within 2 days of beginning treatment. In an editorial in the Journal of the American Medical Association, Jay Levy, MD, of the University of California at San Francisco, speculated that the increase represented a resdistribution or shift of CD4 cells from the lymph system to the peripheral blood, rather than an overall increase in CD4 cell number. He thus called into question the immediate rise in CD4 cell counts after antiretroviral therapy, stating that they may "reflect a response to the drug itself and not a drug-induced protection from CD4 cell death." In response, Sabine Kinloch-de-Loes, MD, observed that while CD4 cell counts in AZT-treated HIV positive patients generally rise, the CD4 cell counts of untreated matched controls receiving placebo fall. The difference is statistically and probably clinically significant. Both agree that a clinical regimen should be judged by its clinical effectiveness, rather than by complete reliance on surrogate markers such as CD4 cell count.

In recent experiments, the administration of the recombinant cytokine protein interleukin 2 (IL-2, or T-cell growth factor) to individuals with HIV disease has resulted in increases in CD4 cells. At this time, it is not known whether the new cells augment the repertoire of the cellular immune system (its ability to seek out and destroy a larger number of antigens) or whether the increases only reflect more copies of the CD4 cells that were already present when IL-2 was given. Therefore, people whose CD4 cell counts have fallen to levels that signal the initiation of prophylaxis for opportunistic infections should continue to receive prophylaxis, even if their CD4 cells counts later return to higher levels. Early research on IL-2 showed deleterious effects on people with fewer than 200 CD4 cells/mm³.

CD4 Cell Count and Viral Load

CD4 cell counts by themselves are less predictive of progression to AIDS or death than are HIV RNA viral load test results. Combining results from both tests allows greater precision in making treatment decisions than does one test alone. Numerous analogies have attempted to provide perspective on the relationship between CD4 cell counts and viral load test results. Thomas O'Brien, MD, at the National Cancer Institute suggests the martial analogy that CD4 cell counts estimate the number of casualties, but HIV RNA estimates the strength of the enemy. This analogy underscores a fundamental difference in the 2 tests: viral load tests gauge viral strength while CD4 cell counts provide a reflection of immune system strength. New guidelines for initiating or changing antiretroviral therapy stress the importance of both tests.

Chemistry Panel

An initial evaluation of an HIV positive patient should include an SMA12, SMA 14 or SMA 20 chemistry panel. The numeral reflects the number of different elements that are measured. These tests will reveal underlying hepatitis (inflammation of the liver) and help to assess liver function when starting a drug regimen. Specific liver function tests help determine if there is an adverse reaction to a drug or if the dose of the drug should be decreased.

A series of tests detects changes in the way the liver produces new substances and breaks down and excretes old substances, and assesses whether liver cells are healthy. One liver function test is designed to detect hepatitis A, another detects hepatitis B, a third detects hepatitis C, and another distinguishes between chronic and acute hepatitis or other liver disorders.

Several chemistry tests measure proteins in the serum, the clear fluid part of the blood. Serum contains no blood cells, but does contain salts, glucose and proteins, including antibodies. Serum bilirubin measures a yellow breakdown product of RBC that is filtered by the liver and spleen. High levels of bilirubin may indicate a defect in processing bilirubin or obstruction in the flow of bile from the gall bladder. The serum albumin test measures a major protein synthesized in the liver. Low albumin levels are found in chronic liver disorders, or may indicate poor nutrition. Serum alkaline phosphatase is an enzyme in the bile. Levels are increased in cholestasis (obstruction of bile flow) or when there is bone disease or chronic kidney failure. Serum aminotransferases are abbreviated SGOT (serum oxaloacetic transaminase) and SGPT (serum glutamic pyruvic transaminase). These liver enzymes are released when liver cells are damaged because of chronic or acute hepatitis.

Glucose (blood sugar) level is measured to detect endocrine disorders such as diabetes. Blood urea nitrogen (BUN) and creatinine clearance are among the tests done to investigate urinary symptoms and kidney disorders. Creatinine level is an indirect measure of the breakdown of muscle tissue, and may indicate myopathy (muscle wasting).

G6PD (glucose 6 phosphate dehydrogenase) is an enzyme that, when deficient, may lead to the destruction of RBC, resulting in anemia. Many physicians only test G6PD when a patient has a history of allergy to TMP-SMX and is a candidate for an alternative prophylactic drug. Dapsone, a drug used to treat or prevent Pneumocystis carinii pneumonia, is metabolized by G6PD; if G6PD is absent, dapsone cannot be tolerated.

Prothrombin time is a test of blood clotting that depends on vitamin K. Abnormal results of this test suggest either liver cell damage or cholestasis. Amylase is an enzyme made in the pancreas and salivary glands that may be elevated due to inflammation of the pancreas or salivary glands, or as a side effect of many drugs used in treating HIV, including ddI, ddC and d4T.

Tests for Specific Infectious Diseases

VDRL (Venereal Disease Research Laboratories) and RPR (rapid plasma reagin) are screening tests for syphilis, a sexually transmitted disease. In addition to the liver function tests described above, tests are also done to detect antibodies against heptatis B virus (HBV) and hepatitis C virus (HCV), which are inflammatory diseases of the liver. Both HBV and HCV are transmitted via sexual contact, through sharing contaminated needles for injection and, rarely, by blood transfusion.

Antibody tests that detect herpes simplex virus (HSV) and cytomegalovirus (CMV) are not routinely recommended, since most adults have antibodies to them, and clinical signs and symptoms are needed to diagnosis active disease. A Toxoplasma serology may be done to test for the presence of antibodies to this protozoan that can cause brain inflammation in people with suppressed immune systems. Its use in asymptomatic patients is controversial, but it is useful in individuals with low CD4 cell counts.

Viral Load Tests

Three tests are able to detect and quantify HIV in the peripheral blood. The Amplicor HIV-1 Monitor Test (reverse transcriptase polymerase chain reaction, or RT-PCR test) made by Roche Molecular Systems is FDA-approved for establishing HIV disease prognosis. The Quantiplex (branched-chain DNA, or bDNA) test made by Chiron, and the NASBA (nucleic acid sequence-based assay) made by Organon Teknika also quantify HIV RNA in the blood plasma. NASBA is more commonly used in Europe than in the U.S. Each test has a limit of detection below which it cannot quantify HIV. Current limits of detection for clinically available tests are 400 copies/mL for the RT-PCR test and 10,000 copies/mL for the bDNA test. Second generation tests are being developed that measure as few as 25 copies (RT-PCR) or 400 copies (bDNA). Test results are reported in copies per milliliter (copies/mL), and are variable depending on which assay is used, the laboratory that processes the sample and the individual laboratory technician who runs the assay. Changes in viral load are referred to in logs (powers of 10), and significant changes are greater than 0.5 log.

In addition to disease prognosis, viral load tests are used to detect HIV in the blood of newborns (who still carry their mothers' antibodies) and in the blood of individuals who were recently exposed to HIV but have not yet manufactured antibodies.

The International AIDS Society-USA and the San Francisco Department of Public Health recommend the use of viral load test results to decide when to initiate or change antiretroviral therapy. Specific recommendations continue to evolve with improvements in limits of detection of individual tests and with increased clinical experience. BETA reports on these advances as developments occur. Current debate on the use of viral load tests questions whether a viral load that is decreased as a result of antiretroviral drug therapy is similar to the low viral load seen in long-term nonprogressors, and whether it predicts an equally favorable clinical outcome. A survival benefit of lower viral load has been seen in some individuals taking combination antiretroviral therapy compared to matched controls in clinical studies. These data support the hypothesis that induced low viral load levels are similar to naturally occurring low levels.

Clinical Guidelines for Using Viral Load Assays

According to John Mellors, MD, from the University of Pittsburgh:

• measure HIV RNA twice (2-4 weeks apart) when first seeing an individual to establish a baseline
  
  
• to estimate prognosis and make treatment decisions, baseline viral load measures should be made in conjunction with a medical history, physical exam and CD4 cell count
  
  
• strongly consider antiretroviral therapy if the level of HIV RNA is greater than 30,000 copies/mL, regardless of CD4 count
  
  
• measure HIV RNA 2-4 weeks after starting or changing therapy and discontinue regimens that fail to reduce HIV RNA level by at least 3- to 10-fold (0.5-1 log) or, preferably, to an undetectable level
  
  
• monitor every 3-4 months for evidence of drug failure, which is indicated by a rise in HIV RNA to within 0.5 log of baseline
  
  
• do not overreact to changes in HIV RNA level that are less than 3-fold (0.5 log), because that degree of variation could be due to test variation alone
  
  
• use the same kind of collection tube, the same assay and the same laboratory to minimize variation in results.

After Viral Load, What's Next?

Viral load testing has advanced the field of HIV/AIDS care greatly, and serves to enhance an individual's understanding of his or her own infection. HIV RNA tests tell how much virus is circulating freely in the blood, and give a reflection of overall viral production in the body. Once the viral load has been suppressed to unmeasurable levels in the peripheral blood, what should be measured next to gauge the effects of antiretroviral therapy? PCR-based DNA tests and branched-chain DNA tests are currently in development to count the number of HIV-infected cells in the body. Once the number of free-floating infectious HIV particles has been effectively reduced, doctors and patients will want to know how many infected cells there are, and whether that number is decreasing over time. Infected cells are an important source of newly infected cells through cell-to-cell interactions, and represent a reservoir of infection that must be eliminated if HIV is ever to be successfully eradicated from the body. PCR-based DNA tests of blood and lymph tissue represent the future of HIV testing.

Mark Bowers is Managing Editor of Treatment Publications at the San Francisco AIDS Foundation.

References

Bartlett JG. The Johns Hopkins Guide to Medical Care of Patients with HIV Infection. Williams and Wilkins, Baltimore, 1994.

Bartlett JG. Medical Management of HIV Infection. Physicians and Scientists Publishing Co., Inc. 1996.

Dailey JF. Dailey's Notes on Blood. Medical Consulting Group, Arlingtion, VA, 1996.

Deyton L. Importance of surrogate markers in evaluation of antiviral therapy for HIV infection. Journal of the American Medical Association 276:159-160. July 10, 1996.

Kinloch-de-Loes S and others. Controversies: does CD4 cell count reflect clinical efficacy? Journal of the American Medical Association 276:1220. October 16, 1996.

Levy JA and others. Plasma viral load, CD4 cell counts and HIV-1 production by cells. Science 271:670-671. February 2, 1996.

Levy JA. Surrogate markers in AIDS research: is there truth in numbers? Journal of the American Medical Association 276:161-162. July 10, 1996.

Mellors J W and others. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 272:1167-1170. May 24, 1996.

Romeyn M. Nutrition and HIV: A New Model for Treatment. Jossey Bass, San Francisco, 1995.

Saag M. HIV viral load markers in clinical practice. Nature Medicine 2(6):625-629. June 1996. Sanford JP. The Sanford Guide to HIV/AIDS Therapy. Antimicrobial Therapy, Inc., Dallas, 1994. Wallace MR and others. Early clinical markers and CD4 percentage in subjects with human immunodeficiency virus infection. Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology 12:358-362. 1996.

Page last updated 20 December 1996