Incidence of tuberculosis in human immunodeficiency virus-infected children in India: Is there a role of isoniazid preventive therapy?

AIM: The aim of this study is to determine the role of isoniazid preventive therapy (IPT) in human immunodeficiency virus (HIV)-infected children in India.
MATERIALS AND METHODS: Factors associated with the development of tuberculosis (TB) in 81 HIV-infected children were analyzed.
RESULTS: The mean age of presentation was 6.36 ± 3.67 years. According to the CDC classification, 4.9% of patients were in Class N, 11.1% were in Class A, 56.8% were in Class B, and 27.2% were in Class C at presentation. TB at presentation was more common in children in CDC Class B and C (P = 0.026). Gender, CD4 count, TB contact, prior TB, and ART status did not have any effect on the development of TB. Children up to 3 years of age developed TB after 6.23 ± 14.07 months after presentation, those between 3 and 6 years developed TB after 14.6 ± 23.27 months, those between 6 and 9 years developed TB after 6.54 ± 21.23 months, those between 9 and 12 years developed TB 40.2 ± 35.98 months after presentation (P = 0.042). Eight patients (16.7%) had multidrug-resistant (MDR) TB and 1 patient (2.08%) had extensively drug-resistant TB.
CONCLUSION: Younger children are more likely to develop TB within a year of presentation whereas those who were near the adolescent age group were more likely to develop TB after 3 years of diagnosis of HIV. Thus, role of IPT in adolescents for 3 years at the time of diagnosis may not be useful as they tend to develop TB later on whereas IPT may be useful in the younger age group. With high incidence of MDR-TB, role of IPT in HIV-infected children in India needs to be re-assessed.

INTRODUCTION

Human immunodeficiency virus (HIV) infection increases the susceptibility to new tuberculosis (TB) or activation of latent TB. There is a 30-fold greater risk of developing TB in HIV-infected patients as compared to those who are not infected with HIV.[1] The incidence of TB and HIV co-infection in India is 8.6/100,000 while the mortality in such patients is 2.8/100,000.[2] The World Health Organization (WHO) recommends isoniazid preventative therapy (IPT) for 6 months in HIV-infected adults and children who are on antiretroviral therapy (ART) and have previously completed antituberculous therapy (ATT).[3] However, the 2015 updated WHO guideline recommends 36 months of IPT in HIV-infected adults and children living in an area of high prevalence of TB irrespective of a positive tuberculin skin test.[4] Even though the initiation of ART is associated with the reduction in the incidence of TB, the co-infection still remains high in these patients.[5] This study determines the incidence of TB in HIV-infected children before and after the diagnosis of HIV and their response to ATT which helps us ascertain if IPT is necessary in children in India, especially in a high endemic area of TB.

MATERIALS AND METHODS

This retrospective study was carried out in the pediatric HIV clinic at B. J Wadia Hospital for Children in Mumbai from April 2002 to May 2013. A total of 81 HIV infected children, who had a regular follow-up for at least 3 years, were included in this study. Both children who were on ART and children who have not received ART were a part of this study. The route of HIV infection was not considered. The exclusion criterion was any patient who had not followed up regularly for a minimum period of 3 years. The diagnosis of HIV infection was made based on positive ELISA by two different kits. Laboratory tests such as viral load, CD4 T-cell count and CD8 T-cell count were done as and when required. Patients were classified clinically in Class N, A, B, and C according to the CDC guidelines.[6] The ART schedule, the duration after presentation when ART was initiated, changes in the ART schedule and the number of months from initiation after which the schedule was changed were noted. Records were evaluated for age at the time of diagnosis of HIV, gender, and previous TB treatment. If a history of previous TB was present, details such as the type of TB, duration of ATT, and the number of months between the diagnosis of HIV and the development of TB were noted. History of TB contact and Mantoux test (MT) was also recorded. In patients who developed TB after being diagnosed HIV infected, the type of TB, the number of months between the diagnosis of HIV and TB, drug resistance status of the infection, CD4 T-cell count at the time of diagnosis of TB and the outcome of the ATT were recorded. Diagnosis of TB was based on either a bacteriologically confirmed case of TB or a clinically diagnosed case of TB. Bacteriologically confirmed case of TB was diagnosed when a patient from whom a biological specimen was positive by smear microscopy (presence of acid-fast bacillus) or culture (growth of Mycobacterium TB) or by a positive geneXpert. A patient who does not fulfill the criteria for bacteriologically confirmed TB but has been diagnosed with active TB on the basis of X-ray abnormalities or suggestive histology and extrapulmonary cases without bacteriological confirmation was labeled as clinically diagnosed TB based on the WHO criteria.[7] Patients who were given MT by 5 TU units of purified protein derivative-S were considered to have a positive result if there was an induration >5 mm at the end of 48 h.

The incidence of TB on follow-up was noted and influence of factors such as age, gender, CD4 count at the time of development of TB, contact history and previous TB treatment and the presence or absence of ART were analyzed.

Statistical analysis

SPSS Version 20.0 (IBM) was used for analysis. Data are represented as frequency (percentage). Categorical variables were analyzed using the Chi-square test or Fisher's exact test. Numerical data were analyzed by t-test. Values of P < 0.05 are considered statistically significant.

RESULTS

Out of 81 children included male: female ratio was 1.07:1. The age of the children ranged from 4 months to 15 years with a mean age of presentation being 6.36 ± 3.67 years whereas the median age was 6 years. According to the CDC classification, 4 patients (4.9%) was in Class N, 9 patients (11.1%) were in Class A, 46 patients (56.8%) were in Class B, and 22 patients (27.2%) were in Class C at presentation. The CD4 T-lymphocyte count at presentation ranged from 16 to 7455 cells/cumm with a mean CD4 count of 838.1 ± 1020.5 cells/cumm. Thirty-four (42%) patients had contact with a patient having TB. Forty-six (56.8%) patients underwent MT at presentation of which 16 (19.8%) patients tested positive. Of the 16 MT positive patients, 10 patients presented with active TB and were started on ATT whereas 3 patients developed TB on follow-up. 5 of the 16 patients had prior TB infection. Twenty-seven (33.3%) patients had a history of previous TB at a mean duration before presentation of 23.5 ± 26.3 months. Of the 27 patients, 1 patient (3.7%) had tubercular meningitis, 2 patients (7.4%) had tuberculous infection of the lymph nodes, and 24 patients (88.89%) had pulmonary TB. The patients received ATT for a mean duration of 10.6 ± 7.5 months. Only 1 patient had two episodes of TB. Both the infections in this patient were pulmonary TB. Forty-eight (59.3%) patients developed TB after a mean duration of 12.2 ± 23.4 months from presentation. Of these 48 patients, 41 patients (85.42%) had pulmonary TB, 2 patients (4.17%) had lymph node TB, 2 patients (4.17%) had abdominal TB, 1 patient (2.08%) has tubercular meningitis, 1 patient (2.08%) had mediastinal TB, and 1 patient (2.08%) had tubercular sacroiliitis. Eight patients (16.7%) had multidrug-resistant (MDR) TB and 1 patient (2.08%) had extensively drug-resistant TB. At the time of being diagnosed with TB, the mean CD4 T-lymphocyte count was 787.3 ± 1214.2 cells/cumm and ranged from 36 to 7455 cells/cumm. All 48 patients received ATT of which 43 patients (89.5%) completed treatment, 3 patients (6.3%) were lost to follow-up, 1 patient (2.08%) did not complete treatment, and 1 patient (2.08%) was undergoing therapy currently. Six patients had repeat TB after a mean duration of 59.5 ± 32.6 months from the diagnosis of TB after presentation, of which 1 patient had MDR-TB, 5 patients had pulmonary TB whereas 1 patient had lymph node TB. The mean CD4 T-lymphocyte count at the diagnosis of repeat TB was 433.8 ± 199.6 cells/cumm and ranged from 153 to 710 cells/cumm. Five patients completed ATT whereas 1 patient did not. Seventy-two patients (88.9%) were started on ART after a mean duration of 21.7 ± 29.7 months since the time of presentation. The ART schedules are shown in Table 1 and change in ART in depicted in Table 2.

Table 1

Initial anti-retroviral therapy schedule in 72 patients

ART schedule	Number of patients (%)
No ART	9 (11.1)
ABC + 3TC + EFV	1 (1.2)
AZT + 3TC + EFV	9 (11.1)
AZT + 3TC + NVP	42 (51.9)
d4T + 3TC + EFV	3 (3.7)
d4T + 3TC + NVP	12 (14.8)
FTC + NVP	1 (1.2)
TDF + 3TC + EFV	2 (2.5)
TDF + FTC + EFV	2 (2.5)

ABC=Abacavir; AZT=Zidovudine; d4T=Stavudine; EFV=Efavirenz; FTC=Emtricitabine; NVP=Nevirapine; TDF=Tenofovir disproxil fumarate; 3TC=Lamivudine; ART=Anti-retroviral therapy

Table 2

Change of anti-retroviral therapy schedule in 31 patients

ART schedule	Number of patients (%)
No change	50 (61.7)
ABC + 3TC + LPV	3 (3.7)
ABC + 3TC + NVP	2 (2.5)
AZT + 3TC + EFV	2 (2.5)
AZT + 3TC + NVP	4 (4.9)
AZT + d4T + NVP	1 (1.2)
AZT + ddI + LPV	1 (1.2)
d4T + 3TC + EFV	2 (2.5)
d4T + 3TC + NVP	6 (7.4)
ddI + d4T + EFV	1 (1.2)
ddI + d4T + NLF	2 (2.5)
TDF + 3TC + ATZ	5 (6.2)
TDF + 3TC + NVP	1 (1.2)
TDF + FTC + EFV	1 (1.2)

ABC=Abacavir; ATZ=Atazanavir; AZT=Zidovudine; ddI=Didanosine; d4T=Stavudine; EFV=Efavirenz; FTC=Emtricitabine; LPV=Lopinavir; NLF=Nelfinavir; NVP=Nevirapine; TDF=Tenofovir disproxil fumarate; 3TC=Lamivudine

The development of TB after presentation was not associated with age, gender, previous TB, contact with TB, CD4 T-lymphocyte count as shown in Table 3. Patients in CDC Class B and C had a greater chance of developing TB [Table 3].

Table 3

Factors associated with development of tuberculosis

Factors	Tuberculosis (n=48), n (%)	P
Gender
Male	26 (54.17)	0.615
Female	22 (45.83)
Previous TB
Yes	16 (33.33)	0.778
No	32 (66.67)
TB contact
Yes	20 (41.67)	0.946
No	28 (58.33)
CD4 count <500
Yes	21 (43.75)	0.906
No	27 (56.25)
CDC class
N	1 (2.08)	0.026
A	3 (6.25)
B	26 (54.17)
C	18 (37.5)

TB=Tuberculosis; CDC=Centers for Disease Control and Prevention

There was no significant association between the ART status and the time of diagnosis of TB after presentation. However, younger children developed TB much earlier as compared to the adolescent age group which developed TB after 3 years of presentation (P = 0.042) [Table 4].

Table 4

Association of anti-retroviral therapy status and the time of diagnosis of tuberculosis after presentation

Factors	Time of diagnosis of tuberculosis after presentation (months)	P
ART status
On ART	13.32±24.15	0.2804 (CI=−11.22–37.86)
Not on ART	0
Age groups (years)
0–3 (n=13)	6.23±14.07	0.042
3–6 (n=15)	14.6±23.27
6–9 (n=13)	6.54±21.23
9–12 (n=5)	40.2±35.98
12–15 (n=2)	0

ART=Anti-retroviral therapy; CI=Confidence interval

DISCUSSION

In non HIV-infected individuals, isoniazid provides protection against TB by increasing the number of TH1 helper cells which produce interferon γ. This effect was seen within a month of therapy and provided long-lasting protection in individuals receiving isoniazid prophylaxis.[8] However, the ability of isoniazid to provoke such a response in HIV-infected individuals is unknown and the mechanism of IPT in such individuals is unclear. TB in HIV-infected individuals is known to accelerate the course of the disease by causing rapid worsening of the immune system which allows the virus to proliferate.[9] Due to the insufficiency of the immune response, the incidence of other severe infections increases. Although the mechanism of IPT is unknown in HIV-infected patients, the efficacy of isoniazid against other possible pathogens could play a role in its use for prophylaxis. Isoniazid is known to be effective in leishmaniasis (in combination with rifampicin), inhibit the development of the malarial parasite and also inhibit Gram-negative bacteria indirectly by inhibiting fatty acid metabolism.[10111213]

Studies regarding the efficacy of IPT have yielded mixed results. A study conducted by Zar et al. involving 263 children of median age 24.7 months in South Africa showed that there was 54% decrease in mortality and 72% decrease in the incidence of TB in the children receiving IPT as compared to the placebo group due which the placebo group of the trial was terminated.[14] However, a study conducted by Madhi et al. showed no benefit in preventing infection in both, HIV-infected children and HIV-uninfected children. The study included 548 HIV-infected and 806 HIV-uninfected infants of the median age of 96 days. The incidence of TB in the isoniazid group and the placebo group of the HIV-infected cohort was 3.7% and 4%, respectively, and was insignificant. A similar result was obtained in the HIV-uninfected cohort.[15]

Long-term usage of isoniazid can lead to complications such as hepatotoxicity and peripheral neuritis. Isoniazid resistance in mycobacterium is also a feared consequence of IPT. Hepatotoxicity in adults on IPT occurred only in 0.15% of individuals completing therapy and the risk of hepatotoxicity is greater in women and older patients.[16] A 5 years follow-up study in patients with fibrotic pulmonary lesions receiving prophylactic isoniazid therapy showed 52 weeks of therapy was the most successful in preventing TB, but it was also associated with the increased incidence of hepatitis. A 24-week regimen had the highest rate of prevention of TB per case of hepatitis caused.[17]

It has been reported that IPT is safe in children infected with HIV, including children on ART, and they tolerate isoniazid well. Although there are concerns about the development of isoniazid resistance in Mycobacterium with IPT, a large number of clinical trials have not shown any such risk.[141518]

CONCLUSION

The study showed that there was no significant association of the development of TB and gender, previous TB, TB contact, CD4 T-lymphocyte count, CD4 T-lymphocyte count <500 and hence, we cannot predict the risk of development of TB in HIV patients. The association of time of diagnosis of TB with age was significant with children at a younger age more likely to develop TB within a year of presentation whereas those who were near the adolescent age group were more likely to develop TB after 3 years of diagnosis of HIV. Thus, the role of IPT in these patients for 3 years may not be useful as they tend to develop TB later on whereas IPT may be useful in the younger age group. The incidence of MDR TB was 18.78% in our study group, and usefulness of IPT may not be there in these patients.

Furthermore, our data also show that 59.3% of patients developed TB of which 89.5% patients were treated successfully. Even 70% of patients with MDR-TB completed treatment, whereas 20% were lost to follow up and 1 patient is still undergoing treatment. Therefore, given the low efficacy and complications such as adverse drug reactions and increase in isoniazid resistance in Mycobacterium of IPT and successful treatment of TB infections, the role of IPT in Indian children needs to be assessed in randomized controlled trials to determine its efficacy and also whether it leads to more incidence of INH resistance in these children when they develop TB.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.