Worldwide, the largest population of persons with HIV infection live in sub-Saharan Africa (1). Because of substantial differences in available technology and resources, the management of HIV-infected persons in these regions differs from management of such persons in industrialized countries. Technology-dependent laboratory tests commonly used in the United States, such as CD4+ lymphocyte counts, neopterin levels, and quantitative determination of viral burden (2,3), are unavailable in most African hospitals. However, other less technology-dependent markers of immune competence are available and have prognostic value in HIV disease; e.g., anergy to delayed-type hypersensitivity (DTH) skin testing independently predicts HIV disease progression (4-7). DTH testing is a useful component of HIV infection staging systems (8,9). In the United States, the Centers for Disease Control (CDC) recommends DTH testing for detection of tuberculosis in HIV-infected persons (10). Despite its apparent utility in developed countries, few studies have examined the use of DTH testing in African hospitals (11,12).
Unlike CD4+ lymphocyte counts, total lymphocyte counts may be obtained with simple technology available in many African hospitals. When CD4 counts are unavailable, total lymphocyte counts may be used as a proxy for CD4 counts (13,14). The World Health Organization (WHO) has proposed that total lymphocyte counts be used in developing countries without access to lymphocyte subset analysis (15). However, because CD4+ lymphocyte counts are superior to total lymphocyte counts as markers in HIV disease (13), data regarding the utility of total lymphocyte counts are limited. Despite numerous clinical studies of HIV infection in Africa, including some reporting CD4 counts (14,16,17), few have reported total lymphocyte counts (12,14,18). In a recent prospective study, total lymphocyte counts <1,000 × 106/L were shown to be associated with increased risk of mortality (18).
In the present study, we examined DTH testing and total lymphocyte counts in medical inpatients with and without HIV infection in an urban East African hospital. We also evaluated these tests for prediction of HIV infection in these patients.
Muhimbili Medical Centre in Dar es Salaam (population ≈2 million) is the national academic medical center of Tanzania. The present study was conducted on two of the seven general medical wards at Muhimbili from February to April, 1992.
Patients aged ≥16 years were eligible for the study. The only exclusion criterion was an inability to provide a medical history. Informed consent was obtained. Interviews, conducted in Kiswahili by native Tanzanians, and physical examination were performed without knowledge of HIV serologic test results. HIV-seropositive patients were provided outpatient follow-up care by members of the research team.
DTH testing was performed with the Multitest® CMI (Institut Merieux) device, a self-contained multiple-puncture device delivering seven antigens and a glycerol control. The seven antigens are tuberculin, candidin, diphtheria, tetanus, trichophytin, proteus, and streptococcus. The device was applied to the volar aspect of the forearm, and induration was measured with calipers 48 h after its application. In accordance with the manufacturer's recommendations, a positive response was defined as induration ≥2 mm at the puncture site.
Total Lymphocyte Counts
Total lymphocyte counts were calculated by multiplying the total leukocyte count by the percentage of lymphocytes in the peripheral smear (manual differential). Smears were examined by a trained medical technologist.
Serum samples were assayed for HIV infection by enzymelinked immunosorbent assay (ELISA) (Genetic Systems). ELISA testing (Organon Teknika) was repeated on all initially positive samples. Repeatedly positive samples were confirmed by Western blot (Organon Teknika) and considered positive if two or more of the gp120, gp160, gp41, and p24 antigen bands were present. Indeterminate blots were considered negative.
HIV Clinical Stage Criteria
Patients were categorized according to the proposed WHO staging classification scheme with clinical criteria only (15). This staging system comprises four clinical stages based on severity of disease or performance status. Details of the staging system have been reported previously (15,18,19). We did not use performance status for staging. Because of limited numbers, we combined stages 1 (asymptomatic or generalized lymphadenopathy) and 2 (“mild disease”). For the tuberculosis component of stage 3, we included only history of tuberculosis in the past year or current pulmonary tuberculosis diagnosed by smear and/or culture.
We also used the proposed Kigali clinical staging criteria, a modified version of the WHO staging scheme (18). According to these criteria, significant changes include the incorporation of pulmonary tuberculosis, oral candidiasis, and chronic genital or oral ulcer in stage 4. We could not assess body mass index as suggested in the Kigali system because of unreliable weight and height measurements in severely ill patients.
The distributions for age, summed response induration, number of positive antigens, and total lymphocyte counts were highly skewed. Therefore, results are presented as the median and 5th and 95th percentiles.
Comparisons between groups were made by the chi-square test for dichotomous or categorical variables and the Wilcoxon rank-sum test for ordinal or continuous variables. For comparisons between stages of HIV infection, the Kruskal-Wallis test was used for ordinal or continuous variables. The critical value for statistical significance was α ≤ 0.05.
We performed multivariable logistic regression to identify predictors of HIV infection (20). The relation of the responses of the seven DTH antigens (dichotomized as positive or negative) to HIV infection was evaluated. In a separate logistic regression model, we assessed lymphocyte count, triceps skinfold thickness, number of positive antigens, and summed response induration of DTH reactions as predictors of HIV infection. Skinfold thickness, number of positive antigens, and summed response induration were modeled as continuous variables. Lymphocyte count was categorized into three categories (cutpoints at 20th and 60th percentiles) with respect to predictive value for HIV infection using a quantile-quantile plot (21).
Two hundred-forty inpatients (120 men and 120 women) were invited to participate. HIV serology and DTH testing was completed in 201 (84%) patients, including 99 men and 102 women. Of the 39 patients who did not complete the study, 11 were discharged before clinical evaluation or serum collection, 4 refused HIV serologic testing, and 4 died before reading of the anergy panel. Twenty patients discharged before reading the anergy panel did not return to the hospital for reading. Total lymphocyte counts were available for 148 patients.
Of the 201 patients, 90 (45%) were HIV positive, including 40 men and 50 women. HIV-seropositive patients were younger (median [5, 95th percentile] = 30 [20,45] years) than HIV-seronegative patients (40 [19,66] years, p < 0.0001). Among the HIV-seropositive group, women were younger than men (27 [19,39] vs. 33 [24,47] years of age, p < 0.001).
According to the WHO staging scheme, 17 (19%) HIV-seropositive patients were stage 1-2, 24 (27%) patients were stage 3, and 49 (54%) patients were stage 4. According to the Kigali scheme, 17 (19%) patients were stage 1-2, 10 (11%) were stage 3, and 63 (70%) were stage 4.
The frequencies of positive responses for individual antigens are shown in Table 1. Tuberculin and candidin induced positive responses most commonly.
To identify antigen subsets for general use in African settings, we examined responses to selected antigen combinations and compared these responses with the full antigen panel. The predictive value of anergy to these antigen combinations as compared with anergy to the full panel was determined (Table 2). Anergy to tuberculin and candidin (predictive value = 81%) predicted anergy to the full panel more consistently than anergy to any other combination of two antigens (predictive values 55-69%). Combinations of three antigens, including tuberculin and candidin plus one additional antigen, had predictive values of 85-90%. Generally, the predictive values for antigen combinations were slightly higher for the HIV-seropositive group.
Immune Function and Clinical Stage of HIV Infection
Anergy was observed in 43% (39 of 90) of the HIV-seropositive persons as compared with 15% (17 of 111) of the HIV-seronegative patients (p < 0.0005). Among HIV-infected persons, anergy occurred progressively more frequently in persons with more advanced stages of disease (Table 3), according to either the WHO or Kigali staging schemes (p < 0.05).
The number of positive antigens and the summed response indurations are potentially useful measures of intermediate immunologic status. Patients with HIV infection responded to fewer antigens and had lower summed response indurations as compared with HIV-seronegative patients (Table 3). More important from a clinical and prognostic perspective, the number of positive antigens and summed response indurations were related directly to the clinical stage of HIV infection.
Total lymphocyte counts were significantly lower among HIV-seropositive persons (1,130 [260, 3,180] × 106/L) than among HIV seronegative persons (1680 [460, 3,720] × 106/L, p < 0.005). Median total lymphocyte counts were progressively lower for each stage of HIV disease according to either the WHO staging system (stage 1/2, 2,200 [280, 3,590] × 106/L; stage 3, 1,240 [380, 4,370] × 106/L; stage 4, 920 [210, 2,240] × 106/L; p < 0.005) or the Kigali scheme (stage 1/2, 2,200 [280, 3,590] × 106/L; stage 3, 1,885 [380, 4,370] × 106/L; stage 4, 970 [210, 2,150] × 106/L; p < 0.005).
Total lymphocyte counts were related to anergy status among HIV-seropositive individuals. Among HIV-seropositive patients with total lymphocyte counts <1,000 × 106/L, anergy to DTH testing was present in 56% (18 of 32). For total lymphocyte counts of 1,100-2,000 × 106/L, and >2,000 × 106/L, the proportion of HIV-seropositive patients with anergy to DTH testing was 33% (7 of 21) and 14% (3 of 17), respectively.
Multivariable Analysis: Prediction of HIV Infection
Anergy to all seven antigens, reflected by the intercept in the full logistic regression model, was a highly significant predictor of HIV infection [OR 2.9 95% CI (2.2, 3.8), p < 0.0001). Of the seven antigens, only responses to diphtheria [OR 0.2 (0.1, 0.6), p = 0.001], tuberculin [OR 0.3 (0.2, 0.7), p = 0.002] and trichophytin [OR 0.2 (0.1, 0.3), p = 0.002] were statistically significant predictors of HIV infection.
In a second logistic regression model examining the relation of lymphocyte counts, triceps skinfold thickness, number of positive antigens, and summed response induration to HIV infection, a total lymphocyte count <600 × 106/L [OR 3.5 (1.0, 12.2), p = 0.05] and number of reactive antigens [OR 0.4 per each additional antigen (0.3, 0.7), p = 0.001] were statistically significant predictors of HIV infection.
Many tests routinely used in the United States and Europe are unavailable for patient care in developing countries. Consequently clinicians practicing in developing countries have limited tools for management of patients with HIV infection. In an African patient population, we examined two tests, DTH testing and total lymphocyte counts, which are potentially useful for the clinical management of HIV-infected persons. Anergy to DTH testing was highly predictive of HIV infection and occurred with increased frequency among patients with more advanced stages of HIV disease. Measurement of the total summed response induration and number of positive antigens, which allow assessment of intermediate levels of immune dysfunction (4,7), decreased significantly with progressively severe clinical stages of HIV infection. Total lymphocyte counts also decreased with progressive stages of HIV disease.
In one previous study, the Multitest CMI device was used in an African patient population (11); the investigators reported an apparent relation between anergy and severity of disease, as estimated by degree of weight loss or opportunistic infection. Responsiveness to PPD and candidin was also associated with lower CD4+ lymphocyte counts in another Zairian study (12). Similarly, total lymphocyte counts have been shown in both African and U.S. settings to correlate with CD4+ lymphocyte counts (13,14). In particular, total lymphocyte counts <1,000-1,500 lymphocytes × 106/L correlate with CD4+ lymphocyte counts <200 × 106/L.
The assessment of relatively simple, technologically appropriate tests is an important step toward improved care for persons with known or suspected HIV infection in Africa. The present study and previous studies examining DTH testing or total lymphocyte counts support the use of these tests in areas with limited resources; e.g., serological tests for HIV infection may be available only after several days, if at all, in many African hospitals. In this situation, DTH testing and total lymphocyte counts provide information regarding the likelihood of HIV seropositivity. Specifically, anergy and marked lymphopenia, used in conjunction with clinical predictors of HIV infection (22), may be considered indicative of a high probability of HIV infection.
DTH testing and total lymphocyte counts may also be used to assess the status of immune function in known HIV-infected persons; e.g., knowledge that an HIV-infected person has anergy and lymphopenia should signal the possibility of more significant opportunistic infections. Decisions regarding empirical therapy should also incorporate this information. It is unfortunate that our cross-sectional study design prohibits assessment of the relation between prognosis and DTH testing or total lymphocyte counts. However, serial application of DTH testing and lymphocyte counts is likely to provide information useful for clinical decision making.
We chose the Multitest CMI device for DTH testing because it is easy to use, provides reproducible administration of antigens, and has been used in several previous studies (6,9,11,23). It also provided simultaneous inoculations of seven antigens, which would have been less practical using the Mantoux technique. It is unfortunate that the cost of the Multitest device restricts its usefulness in areas with few resources. Our data suggest that a limited panel of three antigens (e.g., tuberculin, candidin, tetanus), which costs ≈$1.50 for intradermal injection, may be an acceptable alternative.
Assessment of immunologic function among HIV-seropositive patients in Africa is limited by the expense and technological requirements of the commonly used immunologic markers. Technologically appropriate markers, such as DTH testing, total lymphocyte counts, and erythrocyte sedimentation rates, should be investigated more fully in longitudinal studies to provide inexpensive tools for clinicians caring for HIV-infected persons in Africa. It is hoped that inexpensive, technologically appropriate tests can be used to improve the care of HIV-seropositive patients in Africa and other regions of the developing world.
Acknowledgment: The Multitest CMI device used in this study was a gift from Institut Merieux, Lyon, France. We thank the faculty, residents, staff, and patients of firm V of the Muhimbili Medical Service for support, patience, and cooperation during this study. We thank Drs. H. Rwiza, F. Mugusi, E. Bearer, S. Saha, D. Weaver, and D. Durack for contributions to this study. We also thank Lori Nicholson and Denise Paulsen for technical assistance.
1. World Health Organization. The current global situation of the HIV/AIDS pandemic. Weekly Epidemiol Rec
2. Fahey JL, Taylor JMG, Detels R, et al. The prognostic value of cellular and serologic markers in infection with human immunodeficiency virus type 1. N Engl J Med
3. Mellors JW, Kingsley LA, Rinaldo CR, et al. Quantitation of HIV-1 RNA in plasma predicts outcome after seroconversion. Ann Intern Med
4. Blatt SP, Hendrix CW, Butzin CA, et al. Delayed-type hypersensitivity skin testing predicts progression to AIDS in HIV-infected patients. Ann Intern Med
5. Birx DL, Brundage J, Larson K, et al. The prognostic utility of delayed-type hypersensitivity skin testing in the evaluation of HIV-infected patients. J Acquir Immune Defic Syndr
6. Gordin FM, Hartigan PM, Klimas NG, Zolla-Pazner SB, Simberkoff MS, Hamilton JD. Delayed-type hypersensitivity skin tests are an independent predictor of human immunodeficiency virus disease progression. J Infect Dis
7. Zeballos RS, Cavalcante N, Freire CA, et al. Delayed hypersensitivity skin tests in prognosis of human immunodeficiency virus infection. J Clin Lab Anal
8. Fernandez-Cruz E, Desco M, Montes MG, Longo N, Gonzalez B, Zabay JM. Immunological and serological markers predictive of progression to AIDS in a cohort of HIV-infected drug users. AIDS
9. Royce RA, Luckmann RS, Fusaro RE, Winkelstein W Jr. The natural history of HIV-1 infection: staging classifications of disease. AIDS
10. Centers for Disease Control and Prevention. Purified protein derivative (PPD)-tuberculin anergy and HIV infection: guidelines for anergy testing and management of anergic persons at risk of tuberculosis. MMWR
11. Colebunders RL, Lebughe I, Nzila N, et al. Cutaneous delayed-type hypersensitivity in patients with human immunodeficiency virus infection in Zaire. J Acquir Immune Defic Syndr
12. Mukadi Y, Perriens JH, St-Louis ME, et al. Spectrum of immunodeficiency in HIV-1-infected patients with pulmonary tuberculosis in Zaire. Lancet
13. Blatt SP, Lucey CR, Butzin CA, Hendrix CW, Lucey DR. Total lymphocyte count as a predictor of absolute CD4+
count and CD4+
percentage in HIV-infected persons. JAMA
14. Martin DJ, Sim JGM, Sole GJ, et al. CD4+
lymphocyte count in African patients co-infected with HIV and tuberculosis. J Acquir Immune Defic Syndr Hum Retrovirol
15. World Health Organization. Acquired immunodeficiency syndrome (AIDS): interim proposal for a WHO staging system for HIV infection and disease. Weekly Epidemiol Rec
16. Kivihya-Ndugga LE, Ochola JJ, Otieno G, Muthami LN, Gathua S. Clinical and immunological markers in Kenyan pulmonary tuberculosis patients with and without HIV-1. East Afr Med J
17. Ackah AN, Coulibaly D, Digbeu H, et al. Response to treatment, mortality, and CD4 lymphocyte counts in HIV infected persons with tuberculosis in Abidjan, Côte d'Ivoire. Lancet
18. Lifson AR, Allen S, Wolf W, et al. Classification of HIV infection and disease in women from Rwanda: evaluation of the World Health Organization HIV staging system and recommended modifications. Ann Intern Med
19. The WHO International Collaborating Group for the Study of the WHO Staging System. Proposed “World Health Organization staging system for HIV infection and disease”: preliminary testing by an international collaborative crosssectional study. AIDS
20. SAS Institute Inc. SAS/STAT user's guide, version 6
. Cary, NC: SAS Institute, 1989.
21. Wartenberg D, Northridge M. Defining exposure in casecontrol studies: a new approach. Am J Epidemiol
22. Miller WC, Thielman NM, Swai N, et al. Diagnosis and screening of HIV/AIDS using clinical criteria in Tanzanian adults. JAIDS
23. Guelar A, Gatell JM, Verdejo J, et al. A prospective study of the risk of tuberculosis among HIV-infected patients. AIDS