|
|
| |||||||||||||||||||||||||||||||
Background Although patients with idiopathic CD4+ T-lymphocytopenia and serious opportunistic infections have been described previously, the clinical and immunologic features of this condition have not been well defined.
Methods We studied in detail five patients with idiopathic CD4+ T-lymphocytopenia. The studies included serologic testing, culture, and polymerase chain reaction for the human immunodeficiency virus (HIV) types 1 and 2, serologic testing for the human T-cell lymphotropic virus (HTLV) types I and II, lymphocyte phenotyping, immunoglobulin quantitation, and lymphocyte-transformation assays, as well as attempts to isolate a retroviral agent. The results were compared with those in HIV-infected persons matched for CD4+ T-cell counts and with those in normal controls. We also studied the spouses of patients and the blood donors for one patient.
Results In these five patients, there was no evidence of either HIV or HTLV infection. All the patients had both low percentages and low counts of CD4+ T cells, with relative increases in percentages, but not counts, of CD8+ cells. Numbers of B cells and natural killer cells were generally normal. As compared with HIV-infected persons, our patients had lower percentages and counts of CD8+ cells and more lymphopenia. CD4+ counts were relatively stable over time. Instead of the high immunoglobulin levels seen in HIV infection, these patients had normal or slightly low levels of immunoglobulins. The lymphocyte-transformation response to mitogens and antigens was depressed. Results in spouses and blood donors were normal.
Conclusions Idiopathic CD4+ T-lymphocytopenia differs from HIV infection in its immunologic characteristics and in its apparent lack of progression over time. Nothing about the immunologic or viral-culture studies performed in these patients or about their family members or blood donors suggests that a transmissible agent causes this condition.
Over the past two years, we have studied five persons with idiopathic CD4+ T-lymphocytopenia, which we define as CD4+ T-lymphocyte depletion (absolute CD4+ count, <300 per cubic millimeter, or <20 percent, on more than one occasion), no serologic evidence of HIV infection, and no defined immunodeficiency disease or therapy associated with T-cell depletion. This report describes the preliminary clinical and laboratory findings for these five patients (who are also described in the accompanying article by Smith et al16).
Methods
Study Population
All the patients were referred for the evaluation of unexplained low CD4+ T-lymphocyte counts, and they each consented to participate in a study to evaluate the nature and cause of this abnormality. Laboratory data from patients with HIV infection and comparable CD4+ T-lymphocyte counts, as well as data from normal blood donors matched to the patients for age, race, and sex, were used in the evaluation of the patients.
HIV and HTLV Testing
Serum samples were tested for antibodies to HIV types 1 and 2 with an enzyme immunoassay (Genetic Systems, Seattle). The Western blot assay for HIV-1 was also performed (Biotech/Dupont, Rockville, Md.) and was interpreted according to the criteria of the Association of State and Territorial Public Health Laboratory Directors and the Centers for Disease Control and Prevention (CDC)17. HIV-1 antigen in serum was tested by the p24 antigen assay (Coulter, Hialeah, Fla.). Culture for HIV was performed as described elsewhere18. Cells from two patients (Patients 1 and 3) were also cocultivated with H9 cells and CEM.ss cells (obtained from the AIDS Research and Reference Reagents Program, Division of AIDS, National Institute of Allergy and Infectious Diseases, National Institutes of Health). The cultures were monitored for the formation of syncytia and for cytotoxicity, and the supernatants were collected for testing for HIV p24 antigen19 and reverse transcriptase activity20. Electron-microscopical analysis of cultured cells was also performed,21 and sections were examined extensively for evidence of virus-like particles. The polymerase chain reaction (PCR) was also used to detect HIV proviral DNA in the patients' mononuclear cells22,23. Serologic testing for the human T-cell lymphotropic virus (HTLV) was performed with a variety of assays, including a Western blot assay containing recombinant envelope proteins (Diagnostic Biotechnology HTLV Blot 2.3, Singapore), a recombinant HTLV type I tax Western blot assay (CDC), and a synthetic-peptide assay for HTLV types I and II (Coulter/IAF Biochem International, Montreal).
Immunologic Studies
Lymphocyte-subgroup analysis was performed as described elsewhere24. Serum levels of IgG, IgA, and IgM were measured by a turbidimetric technique with reagents from Dupont Diagnostics (Wilmington, Del.) and the Discrete Automated Clinical Analyzer IV (Dupont Diagnostics). Lymphocyte-transformation responses were quantitated as described elsewhere25.
Statistical Analysis
The test results for patients with idiopathic CD4+ T-lymphocytopenia, HIV-infected persons matched for CD4+ T-cell counts, and normal blood donors were compared by the Wilcoxon rank-sum test,26 since not all the test indexes were normally distributed. A P value of 0.05 or less was considered to indicate statistical significance, and Bonferroni's correction was applied in order to correct for multiple comparisons.
Case Reports
Patient 1
Patient 1 (Patient 12 in the accompanying report by Smith et al.16) was a 70-year-old white man who presented with Pneumocystis carinii pneumonia in March 1991. He responded well to intravenous treatment with trimethoprim-sulfamethoxazole for two weeks. Serologic testing for HIV antibodies was negative, but a CD4+ lymphocyte count was 19 per cubic millimeter. He had no risk factors for HIV infection other than the receipt of three units of whole blood during duodenal-ulcer surgery in 1987. The blood had been tested for HIV antibodies and was negative. His history was otherwise unremarkable. He had not received any immunosuppressive therapy, except for a three-month course of steroids in 1987.
His family had no history of immunodeficiency disease or of recurrent or unusual infections. His wife and two adult sons were in good health. His travel history included a cruise in the Caribbean in 1980 and a trip to Japan, Hong Kong, and China in 1983.
After his recovery from P. carinii pneumonia, the patient was given oral trimethoprim-sulfamethoxazole daily as prophylaxis against P. carinii pneumonia. His subsequent course was complicated by a fungal infection in the groin, which was treated both topically and with oral ketoconazole and which responded well. In April 1992 a grade II papillary transitional-cell carcinoma of the bladder was diagnosed. The patient had had a workup for hematuria in August 1990 and was said to have had a bladder carcinoma, but this diagnosis was not confirmed at another center in both October 1990 and February 1991. He was treated with four weekly infusions of the bladder with thiotepa. He is asymptomatic at this writing.
Patient 2
Patient 2 (Patient 13 in the report by Smith et al.16) was a 38-year-old white male health care worker who presented in October 1984 with headache and was found to have cryptococcal meningitis. He was treated with amphotericin B. In January 1985, shingles developed in a single dermatome. In July 1985, he had a recurrence of his headaches and was found to have a cryptococcoma with a capsule-deficient Cryptococcus neoformans. The cryptococcoma was excised, and he was treated with a full course of amphotericin B and flucytosine. He has been in good health since, except for a nonspecific maculopapular rash over his trunk and upper extremities in late 1985 and early 1986 and a grand mal seizure in January 1988, for which he has been taking phenytoin. He is asymptomatic as of this writing. His medical history revealed two false positive tests for syphilis in November 1984 and June 1985, but this test was negative on other occasions. He had no history of sexually transmitted diseases.
He has no family history of immunodeficiency or serious infections. His wife and two children are in good health.
Patient 3
Patient 3 (Patient 36 in the report by Smith et al.16) was a 56-year-old white woman who sought treatment for jaundice and fever and was found to have a common-duct stone. She underwent cholecystectomy with removal of the stone but had a prolonged postoperative course. Pneumonia and vaginal candidiasis developed. HIV serologic testing was positive, with confirmation by the Western blot assay. She was treated with ciprofloxacin without antifungal therapy, and her pneumonia and vaginitis resolved. She was referred to a tertiary-care center in December 1989, where repeat HIV serologic tests were negative and the Western blot assay showed reactivity only with p24 antigen. Serologic testing for the HIV p24 antigen was negative. A CD4+ cell count was 86 per cubic millimeter. She was temporarily treated with zidovudine and pyrimethamine-sulfadoxine until the results of the repeat serologic testing became known. In January 1991 she was reevaluated. HIV serologic testing was repeated twice and was negative, although her CD4+ count was still low. She had no risk factors for HIV infection other than a history of multiple blood transfusions for hemorrhage during pregnancies, beginning in 1951, and later for menorrhagia in the late 1970s and early 1980s. She had a hysterectomy in 1985 for cervical carcinoma in situ. Her medical history provided no relevant information.
There was no family history of immunodeficiency or serious infections except for a history of tuberculosis in her father's family. Her husband and two children are in good health at this writing.
Patient 4
Patient 4 (Patient 35 in the report by Smith et al.16) was a 45-year-old white man who sought treatment for molluscum contagiosum on his back, legs, and ankles. The lesions disseminated and were treated by curettage or freezing. In 1989 a chest film revealed a mass, and the findings of a lung biopsy were consistent with the diagnosis of a plaque due to asbestosis. He had no known history of exposure to asbestos. In February 1989, he was tested for HIV antibodies by enzyme immunoassay and was negative, but his CD4+ cell count was 96 per cubic millimeter. HIV serologic testing was repeated six months later and was again negative, but the CD4+ cell count remained low. During the six months before the writing of this account he had fatigue and a 5-lb (2.3 kg) weight loss. His medical history provided no relevant information. He had traveled to Mexico and Japan.
His family has no history of immunodeficiency or serious infections. His wife and son are in good health.
Patient 5
Patient 5 (Patient 11 in the report by Smith et al.16) was a 70-year-old white woman who sought treatment in 1983 for disseminated varicella zoster infection. In 1988 she had fever, cough, and pleuritic chest pain and was found to have extensive mediastinal lymphadenopathy. An open biopsy of a mediastinal mass revealed histoplasmosis. Treatment with a combination of amphotericin B and ketoconazole resulted in the resolution of her lung mass and clinical improvement. A CD4+ count during her hospitalization was 275 per cubic millimeter. Two months later it was 499 per cubic millimeter. She was well until April 1990, when a cough developed; she responded well to treatment for presumptive mycoplasma pneumonia. In April 1991 she had cough, fever, vomiting, and dehydration. Acid-fast bacilli were found in her sputum, which eventually grew Mycobacterium avium complex. At that time her CD4+ count was 199 per cubic millimeter. She responded to treatment with ciprofloxacin, which was stopped after her pulmonary infiltrate cleared. In September 1991 a cough again developed, and although she was afebrile, she had a new pulmonary infiltrate. She was again given ciprofloxacin but did not respond to treatment. Bronchial washings again grew M. avium, and she was given ciprofloxacin, rifampin, and ethambutol. Her cough resolved, but her malaise continued. In January 1992 epigastric distress developed, and she was found to be anemic. A gastrointestinal evaluation found a mass lesion in her stomach with a small ulcer. Biopsy of the mass showed histoplasmosis, and Helicobacter pylori was cultured from the area of the ulcer. She was given fluconazole beginning in March and was also treated for the helicobacter infection. She is asymptomatic at this writing.
She has no family history of immunodeficiency or serious infections. She is a widow and has two adult children who are in good health.
Results
The results of HIV serologic testing, culture, and PCR are shown in Table 1. HIV-1 enzyme immunoassay results were weakly reactive in two patients and negative in three. All the HIV-1 Western blot results were either negative or indeterminate. Reverse transcriptase activity was measured with both manganese and magnesium in cultures from three patients (Patients 2, 3, and 4), and the results were negative in all three. Electron-microscopical examination of cultured cells demonstrated no evidence of viral particles. PCR for HIV-1 was negative in all the patients. None of the patients with idiopathic CD4+ T-lymphocytopenia had antibodies to either HTLV-I or HTLV-II.
|
|
|
|
|
Investigation of the patients' spouses and the blood donors to Patient 1 revealed no abnormalities, either of clinical condition or in the phenotype of the lymphocytes.
Discussion
The most frequent cause of CD4+ T-cell depletion unassociated with immunosuppressive therapy in the United States today is HIV infection. Except for a weakly reactive enzyme immunoassay for HIV in two patients (who had negative or indeterminate Western blot assays), serologic testing, culture, and PCR produced no evidence of infection with either HIV or HTLV in the patients we studied. Moreover, measurements of reverse transcriptase activity (a generic marker for retroviruses) after coculture with normal phytohemagglutinin blasts or cell lines revealed no evidence of any other retroviral agent in these patients. There was no cytopathic effect, and no virus-like particles were detected to suggest the presence of another agent that might be capable of killing CD4+ cells.
The changes in lymphocyte populations in these patients differ in nature from those seen in patients with HIV infection who have comparable percentages or counts of CD4+ T cells. In HIV infection lymphocyte levels are maintained at a higher level by the early increase in CD8+ cells in response to the infection27; the patients with idiopathic CD4+ T-lymphocytopenia, who do not have this increase, tend more often to have lymphopenia than comparable HIV-infected persons matched according to percentage of CD4+ T cells. Thus, without a compensatory increase in another cell population, the decrease in the count of CD4+ T cells, which make up the largest single fraction of peripheral-blood lymphocytes, is sufficient to explain the greater lymphocytopenia seen in idiopathic CD4+ T-lymphocytopenia than in HIV infection. The increase in the percentage of CD8+ cells in this syndrome is probably only a reciprocal increase due to the drop in the percentage of CD4+ T cells and similar to the reciprocal increases in the percentages of B cells (CD19+) and natural killer cells (CD3-CD16/56+).
Immunoglobulin levels are usually elevated in patients with HIV infection28. Our patients with idiopathic CD4+ T-lymphocytopenia had either normal or slightly low levels, which, at least in the case of IgG, were lower than those of normal controls. This may also be due to a lack of help from CD4+ T cells in immunoglobulin production in idiopathic CD4+ T-lymphocytopenia without the direct stimulation of such production seen in HIV infection29.
The lack of progressive changes in the CD4+ cell count over time in some of these patients also differs from the invariably progressive drop in counts in HIV infection. Although three of the patients we studied have had low counts for several years, the natural history of this condition is still unknown.
The functional effect of the low CD4+ T-cell counts is evident in the lack of significant response to both nonspecific mitogens and specific soluble antigens in the lymphocyte-transformation assay. In this respect, these patients resemble those with low CD4+ T-cell counts due to HIV infection30.
If HIV is excluded as a cause of the immunodeficiency in these patients, we are left with the challenge of determining the cause. None of these patients have had recent immunosuppressive therapy or disease associated with immunosuppression or immunodeficiency. It has been suggested that this condition might fit into the diagnosis of common variable immunodeficiency31. Although that immunodeficiency is most commonly characterized by recurrent sinopulmonary infections, often with chronic lung disease, and by defective antibody production or hypogammaglobulinemia, it has been diagnosed in a heterogeneous group of imprecisely classified patients32. Patients with common variable immunodeficiency generally have defects in the de-velopment of B cells into immunoglobulin-secreting plasma cells. Some also have depressed lymphocyte responses to mitogen or low CD4+ cell counts33. None of our patients with idiopathic CD4+ T-lymphocytopenia had clinical symptoms suggestive of common variable immunodeficiency, although two had low IgG levels and one had low IgA levels in addition. At least one of these two patients had normal levels of immunoglobulins after the development of opportunistic infection and low CD4+ T-cell counts. Hypogammaglobulinemia may develop due to the loss of CD4+ T cells that assist in B-cell immunoglobulin production34.
The identification of additional patients with idiopathic CD4+ T-lymphocytopenia may reveal that it too is a heterogeneous entity, in which some patients' symptoms or laboratory abnormalities overlap with those seen in common variable immunodeficiency and others belong to other subgroups of idiopathic CD4+ T-lymphocytopenia. At a recent CDC meeting on this subject, three groups described evidence of possible retroviral agents in patients with idiopathic CD4+ T-lymphocytopenia. Since evidence of a novel retrovirus in these patients is not conclusive, nonviral mechanisms affecting CD4+ T-cell counts should also be considered15.
Although three of our patients came from a single state and two were described by a single physician, extensive questioning revealed no apparent connection among these patients. This geographic clustering probably reflects a reporting bias. Studies of the spouses of four of our patients and of the blood donors to one patient revealed no abnormalities in lymphocyte subpopulations. Our search for a possible retroviral agent in this condition was also unrevealing. Together, these findings lead us away from the hypothesis that idiopathic CD4+ T-lymphocytopenia is caused by a transmissible agent.
Persistently low CD4+ T-cell counts are highly unusual24,35. Transiently low counts have been noted in a number of cohort studies of HIV-negative persons: the Multicenter AIDS Cohort Study (Vermund S: personal communication), the CDC San Francisco Cohort Study (Holmberg S: personal communication), the Transfusion Safety Study,36 the New York City Intravenous Drug User Study (DesJarlais D: personal communication), and the New Jersey Intravenous Drug User Study37. Counts have also been reported as being low in several acute infections38,39 and in single cases of cryptococcal meningoencephalitis and disseminated histoplasmosis40,41. Patients 2 and 5, who had cryptococcal meningitis and mediastinal histoplasmosis, respectively, had normal CD4+ T-cell percentages at the time of their initial disease, although the counts were low. In Patient 5 the lymphopenia resolved, but the percentages also subsequently became low. Thus, in some patients, idiopathic CD4+ T-lymphocytopenia may develop consequent to an acute infection, whereas in others the infection is a consequence of the low CD4+ T-cell count.
We have demonstrated major differences in clinical, virologic, and immunologic findings between five patients with idiopathic CD4+ T-lymphocytopenia and patients with HIV infection. Further epidemiologic, immunologic, and virologic studies will aid in the characterization of this newly recognized condition.
We are indebted to the members of the Epidemic Response Laboratory for processing specimens; to Robert C. Hermann, M.D., Gerald Blackburn, D.O., and Neldagae Chisa, M.D., for referring patients; to Marjorie Hubbard, Marianne Simon, Dolly Sinha, Donna Warfield, Cynthia Goldsmith, Judy Petzelt, Charles Schable, Jennifer Rapier, and Martha Monroe for technical assistance; to Ruthie Odum and Fred Ingram for data analysis; to Linda Nealy for assistance in the preparation of the manuscript; to Phyllis Moir for editorial review; and to Steve McDougal, Thomas Hodge, Thomas Folks, Chin-Yih Ou, Harold Jaffe, James Curran, and James Hughes for reviewing the manuscript.
Source Information
From the Immunology Branch, Division of HIV/AIDS (T.J.S., B.M.J., J.K.A.N.), and the Retrovirus (R.B.L., T.R.) and Herpesvirus (A.C.M.) Branches, Division of Viral and Rickettsial Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta; the Divisions of Allergy, Immunology, and Infectious Diseases, William Beaumont Hospital, Royal Oak, Mich. (C.B.L.); the Division of Infectious Diseases, Department of Medicine, Emory University School of Medicine, Atlanta (J.A.S.); and the University of Arkansas for Medical Sciences, Little Rock (R.A.M.). Presented in part at the Keystone Symposia on Molecular and Cellular Biology, Keystone, Colo., March 27-April 4, 1992, and the Eighth International Congress of Immunology, Budapest, Hungary, August 23-28, 1992.The use of trade names is for identification only and does not imply endorsement by the Public Health Service or the Department of Health and Human Services.
Address reprint requests to Dr. Spira at the Division of HIV/AIDS, Mailstop D-08, National Center for Infectious Diseases, Centers for Disease Control and Prevention, 1600 Clifton Rd., Atlanta, GA 30333.
References
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
