Outcomes associated with isoniazid preventive therapy for tuberculosis prevention among human immunodeficiency virus positive patients attending antiretroviral therapy clinics in Mangalore

Introduction

Human immunodeficiency virus (HIV) infection is a major cause of infectious deaths worldwide. In 2017, the prevalence of HIV among adults in India was an estimated 0.22% (0.25% among males and 0.19% among females). In the same year, an estimated 69,110 people died of Acquired Immunodeficiency Syndrome (AIDS)-related causes nationally. India accounts for a high number of tuberculosis (TB) cases worldwide, with 20.2 lakh new cases annually. In addition, there is a concentrated HIV epidemic in India, focusing on a limited population due to risky behavior.¹ Globally, People living with HIV (PLHIV) are 18 times more likely to contract TB infection than those without HIV (as of 2020). Moreover, TB is the leading cause of death among PLHIV worldwide, accounting for nearly 6,90,000 deaths from HIV-associated TB in 2019.² In India, 25% of deaths among HIV patients are caused by TB.³

As part of the HIV-TB collaborative activities, the National AIDS Control Organization (NACO) and Central TB Division (CTD) jointly developed a national framework in 2008 and 2009. Interventions that decreased the morbidity and mortality due to TB in PLHIV were included, one of which is “The Three I’s. These include a) Intensified Case Finding (ICF), b) Infection Control (IC) to prevent TB in HIV care settings, and c) Isoniazid Preventive Therapy (IPT). The latter is due to the fact that Isoniazid protects against both the progression of latent TB infection (LTBI) to active disease (reactivation) as well as from reinfection when exposed to an active TB case. PLHIV who are unlikely to have active TB for a minimum of six months should be initiated on IPT as part of a comprehensive package of HIV care.³ There is substantial evidence from studies conducted all over the world to prove that the provision of IPT decreases the risk of getting infected by TB by 40% in PLHIV.⁴ Although the WHO has prioritized IPT recommendations for PLHIV for more than a decade, its uptake has been slow. In 2016, only approximately 42% of newly diagnosed HIV cases were initiated on IPT worldwide.⁵

In 2008, WHO recommended that Isoniazid Preventive therapy be administered to all PLHIV without TB disease, in localities where the prevalence of latent TB infection was more than 30%, and for all HIV patients with documented latent TB infection or those exposed to an infectious TB case, irrespective of their residence. The NACO introduced IPT as part of HIV care in 2016. It currently recommends screening all PLHIV for active TB using the WHO symptom-based screening algorithm (absence of current cough, fever > 2 weeks, night sweats, and weight loss of >3 kg within 4 weeks) and for isoniazid safety (history of chronic liver disease, seizures, use of heavy alcohol, and prior isoniazid resistance). For eligible persons, the IPT dose must be administered at 10 mg/kg/day (maximum 300 mg) for a six-month period along with pyridoxine (50 mg/day).³

Tuberculosis is a very common opportunistic infection among PLHIV that causes preventable acquired immune deficiency syndrome-related mortality and morbidity. In 2018, an estimated 10 million people were infected with TB worldwide, a number that has remained relatively stable in recent years. An estimate of 1.2 million deaths caused by TB among those people who were negative for HIV were reported in 2018 (a 27% reduction from 1.7 million in 2000), and an additional 2,51,000 deaths among PLHIV (a 60% reduction from 6,20,000 in 2000). India accounts for 27% of TB cases globally.⁶ Even PLHIV who are asymptomatic for TB need TB prophylaxis to lessen the risk of developing TB in the future, thus reducing TB/HIV death rates by approximately 40%. Approximately 49% of PLHIV and TB are unaware of their co-infection and are therefore not receiving adequate care.⁷

Despite India having a high burden of both HIV and TB, only a few studies have been conducted to test the benefits and risks of IPT for PLHIV in Indian settings. While IPT undoubtedly reduces the incidence of TB, it is still uncertain whether there is a reduction in all-cause mortality in HIV-infected patients.⁸ This necessitates further studies to evaluate the outcomes associated with IPT in a programmed setting in India. The Study was conducted with the objective of determining the incidence of TB among PLHIV after receiving 6 months of IPT, who were followed up over a minimum period of 2 years. Second, to determine the burden of adverse drug reactions due to isoniazid (as documented in the case file) and the incidence of persons living with HIV who did not receive IPT and developed TB, taking into consideration that this view of the matter will help improve patient care, reduce mortality and morbidity, and take a stand against two very detrimental infectious diseases in our country.

Methods

A retrospective cohort study was conducted in two ART centers: District Wenlock Hospital, Mangalore, and Kasturba Medical College Hospital in Attavar, Mangalore. The total study population (1014 patients) included patients who were attending ART centers for the treatment of HIV who had completed 6 months of IPT from January 2017 to May 2018 and were followed up until May 2020; patients in the comparison group consisted of those attending ART centers during the same period who did not receive IPT. Data were retrieved from the case files of these patients from June to November 2020.

The study was conducted after obtaining approval from the Institutional Ethics Committee of Kasturba Medical College, Mangalore on 24th June 2020 (approval number: IECKMCMLR-06/2020/206), the Medical Superintendent of Kasturba Medical College Hospital, Attavar, and the Project Director of the Karnataka State AIDS Prevention Society, Bangalore.

The study data included details of PLHIV from the two ART centers who received IPT for six months between January 2017 and May 2018. The data collected were based on demographic details, status of HIV disease and its treatment, past history of tuberculosis and any adverse drug reactions, comorbidities, and anthropometry. The information collected included the details of any patient who developed symptoms of TB disease until June 2020. In addition, CD4 counts and other relevant laboratory investigations were recorded in the data extraction sheet. The above specifics (collected from the case files at the ART centers) were entered into MS Excel and then transferred from the data extraction sheet to a statistical package for social science (SPSS) version 25. Analysis was performed using descriptive statistics and the chi-square test, where p<0.05 is considered as statistically significant. While comparing the BMI at the time of diagnosis of HIV with the latest BMI in the IPT exposed and IPT non-exposed groups, since the data were skewed, the reported median and interquartile range were calculated and non-parametric Mann–Whitney test was performed. The relative Risk was used to compare the incidence of patients who developed TB in the IPT and non-IPT groups. To compare CD4 counts at the start of IPT with the present CD4 counts and those that did not receive IPT, the CD4 count at HIV diagnosis was compared with the present CD4 counts, and the non-parametric Wilcoxon Signed Ranks Test was performed (P <0.05, considered significant).

Results

A total of 1014 patients were included in the analysis; the study population had a male preponderance of 597 (58.9%) compared to 417 (41.1%) females. Most patients (421, 41.5%) were > 50 years of age, followed by 418 (41.2%) patients in the 40-49 years age category. A significant number of patients, 207 (20.4%), were skilled workers, and 209 (20.6%) patients were housewives. Majority of the study population resided in an urban setting, which were 550 (54.2%) patients and 464 (45.8%) patients from rural areas (Table 1).

The baseline clinical information of the patients is shown in Table 2. Most PLHIV have been on ART for 5 – 10 years, that included 531 (52.4%) patients. 377 (39.1%) patients had a CD4 count of <200 cells/mm³ at the time of HIV diagnosis. Majority of PLHIV were diagnosed at Stage I, that is, 447 (44.1%). First line ART regimen was initiated in 994 (98%) patients, whereas 20 (2%) received second line ART regimen. 838 (82.6%) patients on ART had a compliance of ≥95%, compared to 176 (17.4%) patients who were <95% compliant. 524 (51.7%) patients had no history of opportunistic infection on the other hand, 490 (48.3%) patients had an AIDS defining illness. 58 (5.7%) PLHIV were diagnosed with diabetes, and 89 (8.8%) PLHIV were diagnosed with hypertension. 643 (63.4%) PLHIV had no history of alcohol consumption, whereas 105 (10.4%) PLHIV were social drinkers. 145 (14.2%) PLHIV had a history of smoking. 573 (56.5%) patients had a viral load <1000, and in 402 patients (39.6%), the viral load target was not detected.

The median BMI of the patients who received IPT was 21 kg/m² compared to the median BMI at the time of HIV diagnosis, which was 20 kg/m². In contrast, the median BMI of patients who did not receive IPT was 22 kg/m² compared to their median BMI at the time of HIV diagnosis, which was 20 kg/m² (Table 3). As depicted in Figure 1, 37% of the spouses of the patients were affected by HIV.

Figure 1. HIV in family members.

Out of 1014 patients, 525 (51.8%) patients received IPT and 489 (48.2%) did not. Among 525 PLHIV who received IPT, 8 (1.5%) patients developed TB, and among the 489 PLHIV who did not receive IPT, 32 (6.5%) patients developed TB after 2016. As per Table 4, there was a 77% reduction in the incidence of developing TB in those patients who received IPT as compared with those who haven’t received IPT (RR of 0.23, p value <0.0001). This difference was statistically significant.

Of the 525 patients who received IPT, 8 developed tuberculosis after completion of the IPT course. Of the eight patients, four developed tuberculosis within 12 to 24 months after completion of the IPT course, and two patients developed tuberculosis within 12 months after completion of the IPT course (Table 5). Of the eight patients who developed tuberculosis after completion of the IPT course, two patients had a history of one episode each of pulmonary tuberculosis, one patient had pulmonary TB 6 years before commencement of IPT, and the other had a history of pulmonary TB 20 years before initiation of the IPT course. All the 8 patients had an adherence of > 80% to the IPT course.

Out of 525 PLHIV who were initiated on IPT, total of 81 patients were non-adherent, out of which 4 patients finished their 6-month IPT course and had an adherence of <80%, and 77 patients discontinued their IPT course due to the side effects (Figure 2). 444 patients had an adherence to IPT treatment of more than 80%.

Figure 2. Side effect profile amongst those that stopped IPT.

Most patients completed the IPT course 24 to 29 months ago, which accounted for 176 patients; 150 patients completed the course 30 to 35 months back, and 122 patients completed the course ≥ 36 months back.

The median CD4 count of the 525 patients who received IPT significantly improved from 521 cells/mm³ at the initiation of IPT to 548 cells/mm³ (p <0.001) (Table 6). The median CD4 count of the 489 patients who did not receive IPT also showed a significant improvement from 276 cells/mm³ initially to 532 cells/mm³ (p <0.001). By looking into the median of the IPT Exposed patients with those that were not exposed (548 cells/mm³ and 532 cells/mm³ respectively), a minimal improvement in CD4 counts was observed in those who received IPT.

The total number of patients diagnosed with TB was 373, of which 165 developed pulmonary TB and 208 had extrapulmonary TB, of which 8 had both pulmonary and extrapulmonary TB (Figure 3).

Figure 3. Number of patients diagnosed with TB.

Of the 373 patients who were diagnosed with tuberculosis (irrespective of their IPT status), 348 had one episode of tuberculosis, 21 had two episodes of the disease, and four had three episodes.

A significant number of patients were diagnosed at the same time with HIV and tuberculosis infection, which included 152 patients, whereas 141 patients were infected with tuberculosis after being diagnosed with HIV, and a small number (80) of patients had a history of tuberculosis at diagnosis of HIV.

Of the 373 patients who were infected with tuberculosis in the past, the majority (190) had an initial CD4 count of ≥200 cells/mm³, 174 had an initial CD4 count of <200 cells/mm³, and CD4 was not documented in the remaining 9 patients (Table 7). A significant number of patients who developed tuberculosis were classified as WHO clinical HIV stages III and IV (Table 8).

This study also identified a few factors responsible for development of TB and inferred that Patients with a BMI ≥18.5 kg/m² were 0.705 times less likely to develop tuberculosis (OR= 0.705; 95% CI: 0.539-0.923). The PLHIV at WHO clinical stages III and IV had 17.287 times higher risk of acquiring TB than those who were at WHO clinical stages I and II (OR = 17.287; 95% CI: 12.534-23.844). Patients who had a baseline CD4 count of ≥ 200 cells/mm³ were 0.525 times less likely to develop TB than those with a CD4 count < 200 cells/mm³ (OR = 0.525; 95% CI: 0.402-0.686).

Discussion

In our study, A total of 1014 patients were included in the analysis, the study had a male predominance of 597 (58.9%) compared to 417 (41.1%) females. Out of 1014 PLHIV, 421 (41.5%) patients were above 50 years of age, 418 (41.2%) were in the age group 40–49 years, 115 (11.3%) were within 30–39 years, and 60 (5.9%) were below 30 years of age. Majority of the study population resided in an urban setting, which were 550 (54.2%) patients, compared to 464 (45.8%) patients from a rural area.

In this study, 525 (51.8%) PLHIV were subjected to IPT, 444 (84.5% of the 525 PLHIV) of the patients who received IPT had an adherence of more than 80% to IPT compared to 81 (15.42%) patients who were non-adherent, out of which 4 (0.76%) patients finished their 6-month IPT course and had an adherence of <80%, and 77 (14.6%) patients discontinued their IPT course. The four patients who completed IPT course and had an adherence to IPT less than 80% were all males, between the age group 40 and 50 years, three out of the four patients had a history of either smoking, alcohol, or tobacco consumption.

The reasons for stopping IPT were mostly medical reasons, including fatigue, which was detected in 28 patients, followed by gastritis in 16 patients, giddiness in 12 patients, peripheral neuropathy in 12 patients, myalgia in 3 patients, 2 had skin rashes, 1 patient each had other symptoms, including jaundice, vomiting, psychosis, or gynecomastia. Our study considered 80% as a cut off for good adherence, as recommended by WHO, the data for adherence to IPT were collected by going through the monthly documented data in the patients’ ART cards.

Adherence to a regimen plays a major role in its effectiveness. Infection with both TB and HIV causes compliance problems because of the high pill burden and adverse effects. In Addis Ababa, a cross-sectional study was conducted to explore the compliance of PLHIV to IPT. The adherence to IPT was found to be 89.5%. Patients who were on ART were more adherent than those who were on pre-ART. Patients who were counselled about IPT by their healthcare workers were more adherent than those who were uninformed about the reason for taking IPT. They concluded that adherence to IPT was high, suggesting that counselling should be strengthened for patients in the first two months of therapy.⁹

A Study conducted in Tanzania by Munseri, P.J.; Talbot, E.A. et al. to examine factors related to IPT completion among PLHIV in Tanzania, inferred that out of 565 subjects who were initiated on IPT, 87% among them completed the course and 13% did not. Non- compliance was physician-initiated in 33% of the patients, due to active TB or side effects, patient-initiated in 58%, as they were lost to follow-up or due to self-cessation, and 8% were due to patient demise (unrelated to IPT). After questioning patients, it was inferred that those who completed the course did so due to TB fear (44%), understood the importance of IPT (32%), and were counselled (22%). While those patients who did not complete the course were discouraged by stigma (58%), side effects (14%), and commute (1%), their study concluded that PLHIV who were counselled, had regular follow ups monthly and those who were provided with reimbursement for travel had high IPT completion rates with fewer adverse effects.¹⁰

In our Study, 525 patients received IPT, out of which 8 (1.5%) patients developed tuberculosis after completion of the IPT course as per the data that was retrieved from the case cards for over a minimum period of 2 years post IPT as compared to the other prospective pilot study.¹¹ Of the eight patients, four developed tuberculosis within 12 to 24 months after completion of the IPT course, two patients developed tuberculosis within 12 months after IPT completion, one patient contracted tuberculosis within 25 to 36 months from IPT completion, and one patient developed tuberculosis infection 36 months after IPT completion.

Of the eight patients who developed tuberculosis after completion of the IPT course, two patients had a history of one episode each of pulmonary tuberculosis previously and were treated for the same, one patient had pulmonary TB 6 years before commencement of IPT, and the other had a history of pulmonary TB 20 years before initiation of the IPT course; these patients could have developed isoniazid drug resistance that led to failure of IPT. All the 8 patients had an adherence of > 80% to the IPT course, 2 patients from the 8 patients, had an adherence of 85% and one patient had adherence of less than 95%, which could be the reason for failure of IPT; the remaining 5 patients had a CD4 count <500 cells/mm³ when IPT was initiated and a further decline in their CD4 count to <300 cells/mm³ was observed in the year 2020; the mean CD4 count of the 8 patients at the time of receiving IPT was 485.13 ± 194.161 and the mean CD4 count in the year 2020 reduced to 320.13 ± 182.060;, therefore a low CD4 count could be the reason for tuberculosis infection in these patients. Of 489 patients who were not initiated on IPT, 32 (6.5%) patients developed TB after 2016, of which 6 patients had a prior infection of TB before 2017. There was a 77% reduction in the TB incidence among those patients who received IPT as compared with those who did not receive IPT. (RR of 0.23, p value <0.0001)

A multicenter, prospective pilot study was conducted in seven ART centers in urban and semi-urban Indian cities. Monthly counselling and symptom review were conducted during the IPT course and for 6 months after completion of IPT. Their study reported that the TB incidence rate during 6 months after IPT completion was 0.64/100 p-y (95% CI 0.04-1.12) as compared to 2.42/100 p-y (95% CI 1.90-3.10) during the pre- IPT period. The IPT side effects were less than 5%, of which vomiting and skin rash were the most common.¹¹

In the Thibela TB study, they concluded that TB disease among PLHIV who had a prior exposure to IPT had typical outcomes of their treatment similar to their setting and a similar prevalence of isoniazid resistance to background, and that concerns of drug resistance should not prevent the implementation of IPT.¹²

In a Study conducted in Chennai to assess the efficacy of IPT in decreasing the incidence of TB in a cohort of PLHIV between 2012 and 2015 in four states of India, data were collected from nine ART centers. After counselling, eligible PLHIV commenced the IPT course for a duration of 6 months. This study concluded that IPT was effective in reducing incidence of TB in India by almost 50% under programmed conditions and setting the stage to enable and strengthen the IPT services along with ART, which will have a beneficial effect in lowering the TB burden among PLHIV.³

Another Study conducted in Western Ethiopia inferred that the incidence rate of TB in the IPT group was 1.98 per 100 p-y and in the non-IPT group was 4.52 per 100 years. This study identified various predictors of TB. Compared to patients who were at WHO clinical stage I or II, PLHIV at stage III or IV had 3.22 times greater risk of acquiring TB. They also reported that patients with a baseline CD4 count <200 cells/μL were 15 times more likely to develop TB than those with a CD4 count > 500 cells/μL. Their study reported that patients with BMI <18.5 kg/m² had a 1.85 greater chance of getting infected with TB than those with BMI > 18.5 kg/m².¹³

In our study, binary logistic regression was performed to compare the risk factors for TB infection with respect to baseline CD4 count, BMI value, and WHO clinical staging. 373 (36.7%) PLHIV had a past infection of TB, irrespective of their IPT status, of which 259 (43.3%) were male and 114 (27.3%) were female. 165 (44.2%) patients had pulmonary tuberculosis, compared to 208 (55.7%) patients who had extra- pulmonary tuberculosis, among which abdominal TB (2.8%) was most common, followed by TB meningitis (2.2%). Our study identified several risk factors for tuberculosis among PLHIV enrolled on ART irrespective of their IPT status, out of the 373 PLHIV who were diagnosed with tuberculosis, 174 (46.6%) patients had a CD4 count of less than 200 cells/mm³. Patients who had a baseline CD4 count of more than or equal to 200 cells/mm³ were 0.525 times less likely to develop TB than those patients whose baseline CD4 count were less than 200 cells/mm³ (OR = 0.525; 95% CI: 0.402, 0.686). Patients who had a BMI of more than 18.5 kg/m² were 0.705 times less likely to develop tuberculosis when compared to the underweight patients (OR= 0.705; 95% CI: 0.539, 0.923). A significant number of patients who developed tuberculosis were classified as WHO clinical HIV Stage III (152 patients) and Stage IV (145 patients). The PLHIV at WHO clinical stages III and IV had 17.287 times the risk of acquiring TB than those who were at WHO clinical stages I and II (OR = 17.287; 95% CI: 12.534, 23.844).

Conclusion

A Good adherence to IPT above 80% was observed in our study in 84.5% of the 525 PLHIV who were started on IPT. The proportion of PLHIV who contracted TB after six months of IPT was 1.5% of patients. The proportion of PLHIV who did not receive IPT and developed TB was 6.5% of patients. There was a 77% reduction in the incidence of TB in those patients who received IPT as compared with those who haven’t received IPT (RR of 0.23, p value <0.0001).

The failure of IPT could be attributed to an adherence of < 95% and a CD4 count of less than 500 cells/mm³ during IPT initiation. Further research is required to explore IPT resistance. The reasons for stopping IPT were mostly due to medical reasons that were the side effects of IPT, which were experienced by 14.6% of patients who were on IPT. The most common side effect is fatigue, followed by gastritis.

This study concluded that good adherence to IPT and completion of the IPT course significantly reduced the TB burden.

Ethics and consent

The study was conducted after obtaining approval from the Institutional Ethics Committee of Kasturba Medical College, Mangalore on 24th June 2020 (approval number: IECKMCMLR-06/2020/206), the Medical Superintendent of Kasturba Medical College Hospital, Attavar, and the Project Director of the Karnataka State AIDS Prevention Society, Bangalore.

Consent was waived off by the ethical approval committee. We obtained permission from the Medical Superintendent of Kasturba Medical College Hospital, Attavar, and the Project Director of Karnataka State AIDS Prevention Society, Bangalore, before commencement of the study. All the information collected from the patients’ case files was coded to maintain confidentiality.

Author contributions

Steffi Coelho: Data curation, investigation, methodology, project administration, writing-original draft preparation, writing-review and editing and visualization.

Vaishnavi Satish: Investigation, writing-review and editing, methodology and visualization.

Adail Lorainne Dsouza: Investigation, writing-review and editing, methodology and visualization.

Basavaprabhu Achappa: Conceptualization, methodology, supervision, validation, project administration.

Nikhil Victor Dsouza: Conceptualization, methodology, supervision.

Ramesh Holla: Formal analysis, methodology, software.

Himani Kotian: Formal analysis, methodology, software.

Pavan M R: Supervision, validation.