Reports on HIV/AIDS: 1990

SEPTEMBER 21, 1990, MMWR, Vol. 39, No. 37, pp. 638-639, 645-646
Tuberculin Reactions in Apparently Healthy
HIV-Seropositive and HIV-Seronegative Women - Uganda
Persons latently infected with Mycobacterium tuberculosis are at substantially
increased risk for developing clinically apparent tuberculosis (TB) if they become
infected with human immunodeficiency virus (HIV) (1,2). Although skin testing with
purified protein derivative (PPD) by the Mantoux method is a standard method of
screening for tuberculous infection, this method may be hampered by nonreactivity
to skin tests of persons who become immunosuppressed because of progressive HIV
infection. In Uganda, a continuing study of HIV infection in postpartum women,
conducted by the Ministry of Health in collaboration with Case Western Reserve
University, provided an opportunity to study the tuberculin reactivity of apparently
healthy women of known HIV serologic status. This report presents data from the
Uganda study.
In 1988-89, approximately 95% of 2000 pregnant women presenting to Mulago
Hospital in Kampala for uncomplicated delivery volunteered to participate in a
prospective study of HIV infection. Serum specimens obtained from these participants were tested for HIV antibody by enzyme-linked immunosorbent assay (ELISA)
using Recombigen-HIV EIA Kits* (Cambridge BioScience, Worcester, Massachusetts).
All seropositive women and a random sample of seronegative women were then
enrolled in the study.
During the postpartum period, women were tuberculin tested by the Mantoux
technique using Old Tuberculin (OT) 1:2000 (equivalent to 5 tuberculin units [TU] of
PPD) with Tuberculin "GT"* (Behringwerke AG, Marburg, Federal Republic of
Germany) (this preparation is used by the Tuberculosis Control Program of Uganda).
All tuberculin tests were applied and read by the same trained technician who did not
know the HIV status of participants. All reactions were measured at 48 hours with a
millimeter rule and recorded as the mean of two perpendicularly intersecting
diameters of induration. Results were available for analysis for 94 women (33
HIV-seronegative and 61 HIV-seropositive), all of whom appeared healthy and had no
signs or symptoms attributable to HIV infection or opportunistic infection.
Of the 33 HIV-seronegative women, 27 (82%) had tuberculin skin test reaction sizes
>3 mm (the diameter the Ministry of Health selected as a cutpoint), and the median
reaction size for this group was 10.6 mm (Figure 1). Of the 61 HIV-seropositive
women, 29 (48%) had reactions >3 mm, and the median reaction size was 7.5 mm
(p<0.05 for frequency of reactions >3 mm, chi-square test; p<0.01 for difference in
medians, Mann-Whitney U test) (Figure 1).
All but one patient were examined for a BCG (Bacillus of Calmette and Gu6rin)
vaccination scar. Of 32 HIV-seronegative women, 18 (56%) had a BCG scar; of the 61
HIV-seropositive women, 28 (46%) had a BCG scar. For both HIV-seronegative and
HIV-seropositive women, tuberculin nonreactivity was more likely among those
without a BCG scar. Among the HIV-seronegative women, two (11%) of 18 with a BCG
scar had no detectable tuberculin reaction, compared with four (29%) of 14 without a
BCG scar (p=0.17, Fisher's exact test). Among the HIV-seropositive women, seven
*Use of trade names is for identification only and does not imply endorsement by the Public
Health Service or the U.S. Department of Health and Human Services.

103