Indian Pediatrics 2004; 41:595-599
Prevalence of Human Immunodeficiency Virus Infection in Children with Tuberculosis
T. Shahab, M.S. Zoha, M. Ashraf Malik, Abida Malik* and K. Afzal
From the Departments of Pediatrics and Microbiology*, Jawaharlal Nehru Medical College, AMU, Aligarh, UP 202 002, India.
Correspondence to: Dr. Kamran Afzal, Department of
Pediatrics, JN Medical College,
Manuscript received: July 1, 2003, Initial review
completed: August 14, 2003;
The scale of the AIDS crisis now outstrips even the worst case scenarios of a decade ago as dozens of countries are in the grip of the HIV/AIDS epidemic. Of the 14000 new HIV infections per day in 2002, 95% occurred in developing countries and 2000 of these were children(1). About a third of the 40 million people living with HIV/AIDS worldwide at the end of 2001 were co-infected with Mycobacterium tuberculosis, majority of these in resource constrained countries. Tuber-culosis accounts for up to a third of AIDS deaths worldwide(2). Escalating tuberculosis case rates over the past decade in many countries in sub-Saharan Africa and in parts of South-east Asia are largely attributable to the HIV epidemic.
The risk of active tuberculosis in children with HIV is 5-10 times higher than those without HIV(3). Not much data is available on the prevalence and association of HIV with tuberculosis in children in India. The present study was undertaken to estimate the prevalence of HIV infection in children with tuberculosis.
Subjects and Methods
This study was carried out between March 1999 and July 2000 after being cleared by the Institutional Ethics Committee. Consecutive cases of tuberculosis up to 12 year old both hospitalized as well as those attending the outdoor clinics were studied. For the diag-nosis of tuberculosis, suggestive clinical, radiological, cytological and histopatho-logical criteria were taken into considera-tion(4). Asymptomatic Mantoux test positive cases were excluded.
Pre-test and post-test counselling was given to the parents/guardians. After written consent from parents, detailed history, physical examination, relevant investigations and HIV screening of the children were done. About 4 mL venous blood was collected, serum separated and stored at 20Ί C. Samples were tested for HIV-1 and HIV-2 antibodies by INNOELISA (INNOGENETICS NV, Belgium). If tested negative, the child was labeled as HIV negative; positive result was confirmed by a second ELISA using a different antigen/principle. If the first ELISA was equivocal then a second ELISA was performed. If the second ELISA was also equivocal, then ELISA was repeated 2 weeks later. If this ELISA tested negative, then the child was labeled as HIV negative. A supplemental western blot test was also done in all ELISA positive/equivocal cases at National Institute of Communicable Diseases, New Delhi. Parents of all children with positive ELISA were also screened for HIV infection and tuberculosis.
Only five of 250 patients with tuberculosis were HIV positive giving a prevalence of 2%; most patients were in the 1-9 year age group (72.4%) and belonged to lower and middle socio-economic class (65% and 34% respectively). The male to female ratio was 2.3:1. Mean weight for age in the study population was 67.2% and height for age 87.6%. History of contact with tuberculosis was present in 47% of the total cases and in two-thirds of those in 0-1 year age group. Both the parents were illiterate in 61.5% cases, with illiteracy amongst the mothers in 203 (81.2%) cases.
Fever was the most frequent complaint (87.6%) followed by anorexia (48.5%), cough (41.2%), altered sensorium (33.2%), failure to thrive (32.4%), abnormal movement (28%) and lymphnode enlargement (23.6%). Pallor (82.4%) and lymphadenopathy (28.4%) were the most frequent general examination findings. Systemic examination revealed central nervous system involvement in 29% cases, while respiratory system findings and hepatosplenomegaly were present in 26.5% and 11.5% cases respectively. About 53.6% of the cases had tuberculosis involving one or more extrapulmonary sites (Table I). Nearly 78% of those with extra pulmonary involve-ment had coexisting pulmonary tuberculosis. Chest X-ray revealed pleuropulmonary lesions in 69.5% cases with miliary findings in 10.8% cases. BCG vaccination as assessed by the presence of BCG scar was seen in only 21% cases; 90% of those with tubercular meningitis and 77.3% of those with disseminated tuberculosis were unvaccinated. Overall Mantoux test positivity was 29% while BCG test was positive in 59.8% cases.
Table I Extrapulmonary Tuberculosis in Different Age Groups.
All the five HIV positive cases had disseminated/miliary tuberculosis, were males, and were HIV-1 positive. The clinical and radiological profile of the HIV positive cases is summarized in Table II. There was only one seropositive case of 9 months (in whom PCR could not be done) where ELISA repeated on follow up at 18 months of age was also found to be positive for HIV. HIV-seropositive parents were asymptomatic. None of the parents of seropositive cases had active tuberculosis.
Table II Clinical and Radiological Profile of HIV Seropositive Cases.
§ ± means present/absent, * based on NCHS standard, Not available for screening
HIV seroprevalence in the tubercular patients below 12 years of age in this study was 2%, which is much more than the prevalence of 0.7% reported among adults with tuberculosis from Delhi(5). This is also more than the prevalence of less than 1% reported in hospitalized children with high likelihood of HIV from this center(6). The high HIV seropositivity observed in tubercular adults in Mumbai (5.89%), and Pune (20.1%) is probably a reflection of the higher prevalence of HIV infection in general population in these regions(7,8). A prevalence rate of 18% is reported in children with miliary or neurotuberculosis(9). High seropositivity has been reported in both children and adults with tuberculosis in several regions in the grip of HIV epidemic in Central and East African countries(10-12). In contrast, we observed HIV positivity in 5/107 (4.7%) cases of disseminated and neurotuberculosis. If cases of disseminated tuberculosis alone are considered, then the seroprevalence increases to 18.50%, which is very high for a relatively low prevalence region.
In our study, none of the HIV positive cases were Mantoux positive which is in agreement with the observations of Dhurat, et al.(13). This makes it difficult to establish the diagnosis of tuberculosis in these HIV infected children. The clinical features of HIV infection in the seropositive cases were similar to the observations of other studies(13-16). We did not find any notable difference in clinical and radiological manifestations between HIV seropositive and negative tubercular patients confirming the observation of previous report(11); detailed clinico-radiological comparison was not possible due to the small number of seropositive children. Like other reports from children with HIV infection in India(9,15), we noted possible perinatal transmission in 3 of the 5 cases, in the other two cases the probable route of infection was parenteral. However, no conclusions can be drawn regarding the predominant route of transmission in pediatric HIV from such small number of seropositive cases.
HIV and tuberculosis are intersecting epidemics. While it is clear that HIV infection increases the risk of tuberculosis, immuno-logic and virologic evidence indicate that the reverse is also true(3,17). Moreover, both these infections have overlapping sympto-matology. Tuberculosis, which is common in this part of the world, may itself satisfy clinical criteria similar to those for AIDS. Hence, tuberculosis and HIV/AIDS programmes will need to collaborate to deliver a more effective response to TB/HIV. Recently a study from Mumbai has suggested clinically-directed selective HIV screening in the face of increasing HIV prevalence and limited financial resources in India(18). Based on our observations, it can be concluded that in areas where HIV infection is still not very high but prevalence of tuberculosis is much higher, regular screening of cases of disseminated and miliary tuberculosis in children may reduce the number of patients with undiagnosed HIV. However, larger and preferably multi-centric data on prevalence of HIV in disseminated/miliary tuberculosis is required to confirm this observation. Whether the prevalence, pattern and progression of tuberculosis in HIV infected children has any relation to CD4 counts also needs to be evaluated.
Contributors: MSZ, TS, MAM, KA and AM contributed to the concept, designing and analysis of data. MSZ, TS and MAM collected the clinical data; AM performed the ELISA test for HIV. KA and TS drafted the article; KA will act as the guarantor.
Competing interests: None stated.