|
|
| |||||||||||||||||||||||||||||||||||||||||
Background The clinical and public health importance of recent reports of patients with CD4+ T-lymphocytopenia without human immunodeficiency virus (HIV) infection is unclear. We conducted investigations to determine the demographic, clinical, and immunologic features of patients with idiopathic CD4+ T-lymphocytopenia; whether the syndrome is epidemic or transmissible; and the possible causes.
Methods We reviewed 230,179 cases in the Centers for Disease Control and Prevention (CDC) AIDS Reporting System and performed interviews, medical-record reviews, and laboratory analyses of blood specimens from adults and adolescents who met the CDC case definition of idiopathic CD4+ T-lymphocytopenia (<300 CD4+ cells per cubic millimeter or a CD4+ cell count <20 percent of total T cells on two occasions and no evidence of infection on HIV testing), their sexual contacts, household contacts, and persons who had donated blood to them.
Results We interviewed 31 of the 47 patients identified with idiopathic CD4+ T-lymphocytopenia and 23 of their contacts. There were 29 male and 18 female patients, with a mean age of 43 years (range, 17 to 78); 39 were white, 4 were Asian, 2 were Hispanic, and 2 were black. Eighteen patients (38 percent) had one or more risk factors for HIV infection: seven had hemophilia, six had engaged in homosexual sex, six had received blood transfusions, and two had had heterosexual sex partners who were at risk for HIV infection. The other 29 patients (62 percent) had no identified risk factors for HIV infection. Nineteen persons (40 percent) had AIDS-defining illnesses (18 had opportunistic infections), 25 (53 percent) had conditions that were not AIDS-defining, and 3 (6 percent) were asymptomatic. We tested blood from 28 patients: 8 (29 percent) were found to have CD4+ T-lymphocyte counts of less than 300 cells per cubic millimeter, and 6 had CD8+ T-lymphocytopenia (<250 cells per cubic millimeter). Ten sex partners, three household contacts, and four children of the patients, as well as six persons who had donated blood to the patients, were immunologically and clinically normal.
Conclusions This investigation of patients with idiopathic CD4+ T-lymphocytopenia and unexplained opportunistic infections indicates that the disorder is rare and represents various clinical and immunologic states. The investigation of contacts revealed no evidence of a new transmissible agent that causes lymphocytopenia.
This report describes the results of a review of more than 230,000 reports in the AIDS Reporting System and of the investigations of the first 47 patients provisionally identified as having idiopathic CD4+ T-lymphocytopenia and of 23 sexual contacts, household contacts, and persons who gave blood to the patients.
Methods
Because the patients described in the medical literature had low levels of CD4+ T lymphocytes1,2,3,4,5,6,7,8,9,10,11,12 and because the first few patients reported to the CDC had CD4+ T-lymphocyte counts of less than 300 cells per cubic millimeter, the provisional case definition of idiopathic CD4+ T-lymphocytopenia used for national surveillance included the following: a documented absolute CD4+ T-lymphocyte count of less than 300 cells per cubic millimeter or of less than 20 percent of total T cells on more than one occasion, no evidence of infection on HIV testing, and the absence of any defined immunodeficiency or therapy associated with depressed levels of CD4+ T cells. The AIDS Reporting System was searched for all persons who had been given a diagnosis of AIDS and who had a reportedly negative HIV-antibody test to find any who met the provisional case definition. In collaboration with state and local health departments, the CDC developed a standardized surveillance system for collecting and reporting information on persons meeting this definition. The CDC then requested health care providers to report suspected cases through the acquired immunodeficiency syndrome (AIDS) surveillance sections of their local or state health departments13. In addition, letters were sent to investigators in charge of clinical trials and epidemiologic cohort studies sponsored by the Public Health Service, members of the Infectious Diseases Society of America and the National Hemophilia Foundation, laboratories participating in the Model Performance Evaluation Program of the CDC, and physicians and institutions who report cases of HIV infection. The CDC also contacted the investigators who reported cases at the Eighth International Conference on AIDS.
Medical epidemiologists were sent from the CDC to review medical records and to interview and draw blood from persons thought to have idiopathic CD4+ T-lymphocytopenia, their sexual contacts, persons who had donated blood to them, persons who had received blood from them, and household contacts. The questionnaire developed for the interview focused on demographics and possible risk factors for idiopathic CD4+ T-lymphocytopenia; information collected included sexual, occupational, residential, travel, and medical histories and any history of exposure to animals and chemicals. The blood samples were shipped by overnight mail to the CDC.
A complete blood count, a differential count, and lymphocyte immunophenotyping were performed on each blood sample. Serum samples were tested by enzyme immunoassay (Coulter, Hialeah, Fla.) for HIV type 1 (HIV-1) antigen, by enzyme immunoassay (Genetic Systems, Seattle) and Western blotting (Cambridge-Biotech, Rockville, Md.) for antibodies to HIV-1 and HIV type 2 (HIV-2), and by Western blotting (Diagnostic Technology, Singapore) for human T-cell lymphotropic viruses. The levels of IgM, IgG, and IgA in the serum samples were measured with an automated clinical analyzer (Dupont, Wilmington, Del.). Ongoing laboratory studies include amplification with the polymerase chain reaction (PCR) for retroviral DNA sequences14 and mycoplasma; cultivation of peripheral-blood mononuclear cells with normal peripheral-blood mononuclear cells, lymphoid cell lines, or both15; and evaluation of cell cultures for cytopathicity, formation of syncytia, and reverse transcriptase activity15. In addition, standard serologic assays are being performed for antibodies to a variety of agents, including hepatitis B and C viruses, human herpesvirus-6, cytomegalovirus, respiratory syncytial virus, measles virus, parainfluenza virus, enterovirus, adenovirus, parvovirus B19, coronavirus, mycoplasma, rickettsia, and Borrelia burgdorferi. Immunobiologic studies in progress include lymphokine assays, autoimmune profiles, measurements of 
2-microglobulin levels, and selective typing of human lymphocyte antigen.
Results
As of September 17, 1992, the CDC had identified 47 adults and adolescents who met the case definition for idiopathic CD4+ T-lymphocytopenia; 14 of these persons have been described previously10,12,13,16. These cases were reported from 20 states and most regions of the country. Two other patients initially met the idiopathic case definition and also were investigated, but were found to be HIV-1-seropositive on testing at the CDC; these patients are excluded from the results.
The survey of 230,179 cases of AIDS reported to the AIDS Reporting System initially revealed 299 HIV-seronegative patients. The evaluation of 172 of these patients (58 percent) is complete. Of these 172 patients, 131 have been reclassified as being HIV-positive, 4 have been reclassified as having some other type of defined immunodeficiency, and 8 HIV-exposed infants died before their serostatus could be determined. Twenty-six seronegative patients with AIDS had died, and 1 could not be located, but the race, age, reported risk factors for HIV infection, and history of AIDS-defining conditions of these 27 patients were comparable to those of HIV-seropositive patients with AIDS. Their mean age was 38 years; 89 percent were male, 74 percent were white, and 78 percent had risk factors for HIV infection (63 percent had engaged in homosexual sex, 11 percent were injection-drug users, and 4 percent had received transfusions). According to available information, none had had HIV cultures or experimental tests for HIV, such as PCR for retroviral DNA sequences. Two HIV-seronegative patients with AIDS in the AIDS Reporting System fulfilled the criteria for idiopathic CD4+ T-lymphocytopenia -- they had two recorded CD4+ cell counts of less than 300 cells per cubic millimeter -- and are included in the group of 47 patients described in this report.
The 47 patients with idiopathic CD4+ T-lymphocytopenia -- 45 living and 2 dead -- included 29 male (62 percent) and 18 female patients (38 percent) (Table 1). These 47 patients include some of those described by Spira et al17. (Patients 11, 12, 13, 35, and 36), Ho et al18. (Patients 7, 22, 25, and 34), and Duncan et al19. (Patients 3, 4, 17, and 18) elsewhere in this issue.
|
Clinically, 19 patients (40 percent) had AIDS-defining illnesses, 25 (53 percent) had other illnesses or conditions that were not AIDS-defining, and 3 (6 percent) were asymptomatic (Table 1). The 19 patients with AIDS-defining conditions did not differ significantly from the 28 other patients with respect to age, sex, race or ethnic group, or the likelihood of having risk factors for HIV infection.
Opportunistic infections at or before the time of the interview were the most common clinical illnesses, and determinations of CD4+ lymphocytes were not performed at the time of the infection in all patients. Three patients had extrapulmonary Mycobacterium avium complex, two had pulmonary M. avium complex, and one had atypical mycobacterial pneumonia. Three cases of pulmonary tuberculosis were identified. Four patients had Pneumocystis carinii pneumonia, and two had cerebral toxoplasmosis. Mycotic infections were also frequent, including two cases of extrapulmonary histoplasmosis, four cases of cryptococcal meningitis, and one case each of pulmonary and extrapulmonary cryptococcosis; four patients had oral candidiasis (thrush), two had recurrent vaginal candidiasis, and one had esophageal candidiasis. The viral infections reported included nine cases of herpes zoster and a few cases each of oral or genital herpes simplex, human papillomavirus infection, molluscum contagiosum, and cytomegalovirus infection (Table 1). Three cases of cancer were reported: one case of non-Hodgkin's lymphoma (occurring after CD4+ T-lymphocytopenia was documented), one case of invasive squamous-cell carcinoma, and one case of mixed squamous-cell and basal-cell carcinoma. Kaposi's sarcoma was not reported by study participants or their physicians or found on a review of hospital charts.
The reported absolute CD4+ T-cell levels among the 47 patients with idiopathic CD4+ T-lymphocytopenia ranged from 0 to 296 cells per cubic millimeter (0 to 37 percent of total T cells) (Table 2). The date on which the first CD4+ T-cell count of less than 300 cells per cubic millimeter was recorded was known for 44 patients: 3 had their first low count documented during 1983 through 1984, 3 during 1985 through 1986, 4 during 1987 through 1988, 11 during 1989 through 1990, and 23 during 1991 through 1992. The mean interval between the first reported low CD4+ T-lymphocyte count and the last was 15 months (range, 1 week to 102 months). The mean (±SD) value of the first reported low CD4+ T-cell count was 146 ±90 cells per cubic millimeter, and of the last reported low count, 149 ±100 cells per cubic millimeter (Table 2).
|
Immunoglobulin tests have been completed for blood specimens from 23 patients. Of the three immunoglobulin determinations for each patient, all but nine values (from five patients) were within normal ranges: IgA, 57 to 114 mg per deciliter; IgG, 568 to 1483 mg per deciliter; and IgM, 20 to 274 mg per deciliter. Three patients had values just outside the normal limits: Patient 5 (IgG, 1494 mg per deciliter), Patient 7 (IgG, 564 mg per deciliter), and Patient 9 (IgA, 53 mg per deciliter). Patient 3 had a low concentration of IgG (470 mg per deciliter), whereas Patient 16 had an elevated concentration of IgG (1653 mg per deciliter). Patients 6 and 27 had low levels of both IgA (41 and 32 mg per deciliter, respectively) and IgG (468 and 503 mg per deciliter, respectively).
The 47 patients with idiopathic CD4+ T-lymphocytopenia studied here had not been exposed to one another. Ten spouses or sex partners, three relatives or household contacts, and four children of the patients, as well as six persons who had given blood to the patients, were located and consented to participate in this investigation; all were clinically well and HIV-seronegative and had CD4+ T-lymphocyte counts in the normal range.
Discussion
Despite intensive surveillance efforts, few persons with opportunistic infections or unexplained lymphocytopenia without HIV infection have been identified over the past decade. Patients such as those we describe do not appear to be part of a recently emerging phenomenon, since cases of HIV-seronegative "AIDS" in the AIDS Reporting System or of idiopathic CD4+ T-lymphocytopenia have been reported since 1983. Such cases may well have occurred earlier without being identified because testing of T-lymphocyte subgroups was not widely available until recently. The increasing testing of persons with a variety of disorders, as well as large cohorts of HIV-seronegative homosexual men and men with hemophilia, may be contributing to the reported increase in the cases of idiopathic CD4+ T-lymphocytopenia in the past few years. Although an unknown infectious agent of immunodeficiency cannot be ruled out definitively, the epidemiologic data do not suggest that the condition is caused by a transmissible agent. To date, the cases have not been clustered, and the limited number of sexual contacts, household contacts, and persons who have donated blood to affected patients who have been studied were clinically well and had normal CD4+ T-lymphocyte counts.
At the meeting entitled "CD4+ T-Lymphocyte Depletion in Persons without Evident HIV Infection" (held in Atlanta, August 14, 1992), data on the frequency of idiopathic CD4+ T-lymphocytopenia in prospectively followed cohorts were presented. A review of the Multicenter AIDS Cohort Study of the NIAID (published elsewhere in this issue)20 showed that 2713 HIV-seronegative homosexual and bisexual men had had CD4+ T-lymphocyte counts determined 22,643 times as part of the study's routine protocol. Only one man had CD4+ T-cell counts persistently below 300 cells per cubic millimeter, and this patient was receiving immunosuppressive therapy. A similar review of a CDC-funded multi-city cohort study of homosexual and bisexual men found that none of the 2343 CD4+ T-cell counts determined in 756 HIV-seronegative men were persistently lowered in the absence of some other known cause of immunosuppression.
Also reported at the meeting was a review of 275 blood donors, 970 transfusion recipients, and 947 household contacts of transfusion recipients. The Transfusion Safety Study (reported on elsewhere in this issue)21 showed that 12 persons (0.5 percent) -- 2 recipients of blood transfusions, 5 men with hemophilia who received heat-treated coagulation factors and 2 who received untreated coagulation factors, and 3 household contacts -- had CD4+ T-lymphocyte counts below 300 cells per cubic millimeter on two or more occasions with no identifiable cause. All 12 persons had general lymphocytopenia, defined in that study as a total lymphocyte count of less than 1500 cells per cubic millimeter; these persons are undergoing further investigation by the CDC. A survey of healthy blood donors, including 304 at Irwin Memorial Blood Center in San Francisco, 235 at the Los Angeles Red Cross, and 262 at the Sacramento Medical Foundation Blood Center, found that none had a low CD4+ T-cell count on more than one occasion without having a known or likely reason for immunosuppression. Likewise, none of 970 blood donors at the New York Blood Center had a CD4+ T-lymphocyte count below 300 cells per cubic millimeter on even one occasion.
Studies of the populations mentioned above, similar selected populations (injection-drug users22), and healthy populations23 have indicated that a small percentage of HIV-seronegative persons have low CD4+ cell counts. Thus, although some of these healthy persons may meet the current case definition for idiopathic CD4+ T-lymphocytopenia, they probably represent a different group from those who have opportunistic infections.
The symptomatic patients appear to represent several different clinical and immunologic patterns. The temporal relation between the mycobacterial infections, P. carinii pneumonia, and extrapulmonary fungal and parasitic infections and the suppressed CD4+ T-lymphocyte counts in these patients and in others in the medical literature4,24,25,26,27 is unclear. Some conditions, such as P. carinii pneumonia, apparently result from antecedent immunosuppression; others, such as tuberculosis28 and many other conditions,26,27,29 may temporarily diminish lymphocyte subpopulations, producing the lymphocytopenia observed in some patients. For example, Patient 7, who recovered from cerebral toxoplasmosis, subsequently had CD4+ T-lymphocyte counts that were within the normal range and were consistent with counts he had before his illness. Seven additional patients who were recently tested at the CDC also had CD4+ T-lymphocyte counts of more than 300 cells per cubic millimeter.
Several of the 47 patients had multiple and different cellular abnormalities in addition to the depletion of CD4+ T lymphocytes. Of the 28 recently tested patients, 22 (79 percent) had total lymphocyte counts of less than 1500 cells per cubic millimeter, and 6 (21 percent) also had CD8+ T-lymphocytopenia (<250 cells per cubic millimeter). Moreover, two homosexual men who were thought to meet the case definition for idiopathic CD4+ T-lymphocytopenia were found to be HIV-seropositive on testing at the CDC, indicating that their lymphocytopenia may have been related to HIV infection. All patients with idiopathic CD4+ T-lymphocytopenia will need to be observed prospectively and tested after their acute opportunistic infections or after their first CD4+ cell count less than 400 cells per cubic millimeter to determine the natural history of their infections and lymphocytopenia.
CD4+ cell counts may be influenced by many factors, including age23 and even natural diurnal variation30,31. Laboratory variation may also account for some abnormal CD4+ T-lymphocyte counts32,33. For example, four blood samples from Patient 45 were split and sent to two testing laboratories in his home state, revealing absolute CD4+ T-lymphocyte counts that differed by 69 to 152 cells per cubic millimeter.
Almost all patients with idiopathic CD4+ T-lymphocytopenia tested to date had serum immunoglobulin levels within or very near the normal ranges. These findings are in contrast to the elevated immunoglobulin levels observed in HIV-infected persons34 and to the decreased levels typical of common variable immunodeficiency35.
Although nearly 40 percent of the identified patients with idiopathic CD4+ T-lymphocytopenia were at risk for HIV infection, this may reflect an ascertainment bias, in that such persons, particularly men with hemophilia or homosexual men, are being followed and are having regular determinations of lymphocyte subgroups as part of large, ongoing cohort studies of groups at risk for HIV infection. As such, they are more likely to be identified as meeting the case definition for idiopathic CD4+ T-lymphocytopenia. Because of this bias, it would be inappropriate to conclude that the risk factors for idiopathic CD4+ T-lymphocytopenia are necessarily the same as those for AIDS. In fact, 62 percent of the 47 patients investigated had no risk factors associated with HIV infection.
Although this report describes adults and adolescents with idiopathic CD4+ T-lymphocytopenia, 10 children with comparatively low CD4+ T-lymphocyte counts have also been described and are under investigation by the CDC. Because the CD4+ T-lymphocyte counts of healthy young children are considerably higher than the established values for adults,36 the definition of idiopathic CD4+ T-lymphocytopenia among children includes the following criteria: a CD4+ T-cell count of less than 1000 cells per cubic millimeter in children 0 to 23 months of age and of less than 300 cells per cubic millimeter in children 2 to 12 years of age, or a CD4+ T-lymphocyte count that is less than 20 percent of total lymphocytes, on at least two separate measurements; no serologic evidence of infection on HIV testing (even if the child's mother is HIV-seropositive); and the absence of any defined immunodeficiency or therapy associated with T-cell depletion. Persons fulfilling the case definition for adult1 or childhood idiopathic CD4+ T-lymphocytopenia should be reported to state and local health authorities.
Many laboratories are now working to identify possible etiologic agents of idiopathic CD4+ T-lymphocytopenia. Two recent preliminary reports suggest the presence of a retrovirus in affected patients,11,12 but conclusive evidence of a retrovirus is lacking. The relevance of these reports to the immunodeficiency detected in the patients described here is not known.
In summary, the investigations of cases of idiopathic depletion of CD4+ T lymphocytes indicate that they are rare, that they probably represent various disorders, and that in some cases, low CD4+ T-lymphocyte counts may reflect transient responses to infections or other conditions or even normal findings in asymptomatic patients. The absence of immunodeficiency in the contacts of patients with idiopathic CD4+ T-lymphocytopenia and the absence of significant clustering of cases provide no evidence of either a new infectious agent or an environmental cause of the disease. While laboratory investigations attempt to determine the underlying cause or causes of the opportunistic infections and lymphocytopenia, further epidemiologic investigations will need to focus on its natural history.
We are indebted to the following providers and presenters of epidemiologic and clinical information, cited in this article, at the meeting on CD4+ T-lymphocyte depletion in persons without evident HIV infection (held in Atlanta, August 14, 1992): Sten Vermund, M.D., Dr.P.H., NIAID, Bethesda, Md.; James Mosley, M.D., University of Southern California School of Medicine, Los Angeles; Michael Busch, M.D., Ph.D., Irwin Memorial Blood Center, San Francisco; Celso Bianco, M.D., New York Blood Center, New York; Charles Farthing, M.D., New York University Medical Center, New York; Eric Daar, M.D., Cedars-Sinai Medical Center, Los Angeles; Harry Prince, M.D., American Red Cross Blood Services, Los Angeles; and Paul Holland, M.D., Sacramento Medical Foundation Blood Center, Sacramento, Calif.; and to the following persons, who were instrumental in the field investigations of cases: Alan Greenberg, M.D., Pauline Thomas, M.D., Rosalyn Williams, Mary Anne Chaisson, Ph.D., and Isaac Weisfuse, M.D., New York City Department of Health; Dava Klirsfeld, M.D., Bellevue Hospital, New York; Bernard Nahlen, M.D., Los Angeles County Health Department; Allyson Sage, R.N., M.P.H., and Juan Ruiz, M.D., M.P.H., California Department of Health Services, Sacramento; Carl Lauter, M.D., William Beaumont Hospital, Royal Oak, Mich.; Jonas Shulman, M.D., Emory University School of Medicine, Atlanta; Roberta Monson, M.D., Little Rock, Ark.; Robert Hermann, M.D., Marietta, Ga.; Adam Asch, M.D., Cornell University Medical College, New York; Fabienne Laraque, M.D., M.P.H., New Jersey State Department of Health, Trenton; C. Ford von Reyn, M.D., Dartmouth-Hitchcock Medical Center, Lebanon, N.H.; Alfred DeMaria, Jr., M.D., and William Adams, M.D., Massachusetts Department of Public Health, Boston; Geoff Beckett, P.A., M.P.H., Maine Department of Health, Augusta; Christopher Grace, M.D., University of Vermont Medical School, Burlington; Sherwood Gorbach, M.D., Tufts University School of Medicine, Boston; Thomas Taylor, M.D., Dartmouth Medical School, Hanover, N.H.; Jordan Glaser, M.D., Staten Island University Hospital, New York; M. Geoffrey Smith, M.D., M.P.H., and Joanna Buffington, M.D., M.S., State Department of Health and Human Services, Concord, N.H.; Sara Hawk, P.A., Oklahoma Children's Hospital, Oklahoma City; Robert Schooley, M.D., University of Colorado Health Sciences Center, Denver; Indira Warrier, M.D., Children's Hospital, Detroit; Paul Zenker, M.D., M.P.H., Oklahoma State Department of Health, Oklahoma City; William Hall, M.D., M.P.H., Michigan Department of Public Health, Lansing; Nancy Madinger, M.D., University of Colorado Medical Center, Denver; Kenneth Gershman, M.D., M.P.H., and Nancy Enyart, M.A., Colorado Health Department, Denver; and Heidi Robb, R.N., M.P.H., and Sally Bidoll, R.N., M.P.H., Michigan Department of Public Health, Detroit.
Source Information
The members of the task force are as follows: Epidemiology and Surveillance Working Group: Martha F. Rogers, M.D. (chair); Epidemiologic Investigations: Scott D. Holmberg, M.D., M.P.H. (coordinator), Thomas J. Spira, M.D., Dawn K. Smith, M.D., M.P.H., M.S., Pascale M. Wortley, M.D., M.P.H., Lisa A. Jackson, M.D. (Division of Bacterial and Mycobacterial Diseases, National Center for Infectious Diseases [NCID]), Judith R. Rudnick, M.D. (Hospital Infections Program, NCID), Judith Falloon, M.D., and Doreen G. Chaitt, R.N., M.P.H. (Laboratory of Immunoregulation, National Institute of Allergy and Infectious Diseases); Surveillance Operations: I. Celine Hanson, M.D., and John W. Ward, M.D. (coordinators), Joyce J. Neal, Ph.D., M.P.H., Kenneth A. Bell, Laurence Slutsker, M.D., M.P.H., and C. Rene Jones; Other Members: Harold W. Jaffe, M.D., Mark N. Lobato, M.D., Kenneth G. Castro, M.D., Judith A. Hannan, Debra W. Jackson, Bruce L. Evatt, M.D., and Timothy J. Dondero, M.D.; and Laboratory Working Group: Gerald Schochetman, Ph.D. (chair), Charles A. Schable, M.S., J. Steve McDougal, M.D., J. Richard George, Ph.D., Thomas J. Spira, M.D., and Thomas M. Folks, Ph.D.
Address reprint requests to Dr. Smith at the Division of HIV/AIDS, Mailstop E-45, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, GA 30333.
References
| |||||||||||||||||||||||||||||||||||||||||
Related Letters:
CD4+ T-Lymphocytopenia without HIV Infection
Sheppard H., Winkelstein W., Lang W., Charlebois E., Heymann D. L., Belsey E., Esparza J. G., Laurence J., Corboy J. R., Stevens S. R., Griffiths T. W., Cooper K. D., Smith D. K., Neal J. J., Holmberg S. D.
Extract |
Full Text
N Engl J Med 1993;
328:1847-1850, Jun 24, 1993.
Correspondence
Idiopathic CD4+ T-Lymphocytopenia
DeHovitz J. A., Feldman J., Landesman S.
Extract |
Full Text
N Engl J Med 1993;
329:1045-1046, Sep 30, 1993.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
