*Elizabeth Glaser Pediatric AIDS Foundation, Maseru, Lesotho;
†Stellenbosch University, Cape Town, South Africa;
‡New York Department of Health;
§Ministry of Health Lesotho, Maseru, Lesotho;
‖Center for Tuberculosis Research, Johns Hopkins University, Baltimore, MD;
¶Elizabeth Glaser Pediatric AIDS Foundation Washington, DC;
#Department of Epidemiology/Biostatistics, Milken Institute School of Public Health, The George Washington University, Washington, DC; and
**Georgetown University, Washington, DC.
Correspondence to: Appolinaire Tiam, MBChB, MMed, Elizabeth Glaser Pediatric AIDS Foundation, First Floor Sechaba House, 4, Bowker Road, Maseru 100, Lesotho (e-mail: firstname.lastname@example.org).
Supported by funds from the Bill and Melinda Gates Foundation through an award to the Johns Hopkins University, and the Elizabeth Glaser Pediatric AIDS Foundation.
Parts of the data were presented at the 43rd Union World Conference on Lung Health, and Abstract Oral Presentation OP-157-16, November 13–17, 2012, Kuala Lumpur, Malaysia.
The authors have no conflicts of interest to disclose.
A.T. designed the study, analyzed the data, and wrote the article. S.K. contributed to the design and analysis, and helped with the editing. R.M. analyzed the data and contributed to editing. C.R.G., L.B.M.M., K.N-S, M.S., O.O., A.A., G.L.B., R.E.C., and L.G. contributed to study design and editing of article.
Received May 05, 2014
Accepted May 05, 2014
Sub-Saharan Africa suffers a double epidemic: there is a high rate of HIV and tuberculosis (TB) coinfection leading to high morbidity and mortality.1–4 Southern Africa has high rates of TB among HIV-infected pregnant women, with the combination of TB and HIV contributing significantly to mortality among pregnant women.5,6 TB has also been associated with increased risk of HIV transmission from mother to child.7
In a meta-analysis of TB symptom screening algorithms coordinated by World Health Organization (WHO), the best performing rule was a 4-symptom screen using one of any of the following: cough in the last 24 hours, fever, night sweats, and weight loss. This 4-symptom screen rule had a sensitivity of 90.1% in a clinic setting and negative predictive value of 98.3%, assuming TB prevalence of 5% among people living with HIV. CD4 cell count (greater or less than 200 cells/mL) had little impact on the sensitivity or specificity of this symptom screen.7–10 The WHO guidelines thus recommend screening for TB using the presence of cough of any duration, fever, night sweats, and weight loss, and collecting sputum for testing from symptomatic persons with any of the symptoms above.8
Several studies have demonstrated the efficacy of active case finding (ACF) for identification of active TB enabling initiation of appropriate medical therapy, and provision of isoniazid preventive therapy (IPT) to prevent TB.8–11 In 2011, the WHO released updated guidelines for intensified case finding (ICF), IPT, and infection control (IC), and known as the 3Is, which included a conditional recommendation to provide IPT for pregnant women.8,12,13 By definition, ICF is provider-initiated TB screening among HIV-infected persons, whereas ACF screens all individuals regardless of their known risks for TB.5,8,14 Researchers in Tanzania and Brazil reported that a large proportion of adult patients offered IPT accepted it with 85%–87% completing the recommended 6-month treatment.14,15 Research in Thailand, India, and South Africa found that IPT reduces the risk of TB significantly among HIV-positive individuals despite various reported operational challenges.16,17
Lesotho adopted the 2010 WHO guidelines for ICF/IPT/IC and designed an implementation process that included the introduction of ACF and IPT within maternal and child health (MCH) settings.18 We undertook an evaluation of the uptake of ACF examining the rate of TB screening and diagnosis between HIV negative and positive, and uptake and completion of IPT among eligible HIV-positive pregnant women during the first year of implementation of the national 3Is guidelines in 2 secondary level health facilities in Lesotho. To the best of our knowledge, this is the first evaluation on the rollout of IPT among HIV-positive pregnant women in Africa.
This was a descriptive prospective study using data abstracted from patients' file after the participants were seen by a health care provider. The aim of the study was to evaluate the implementation of guidelines for ACF among HIV-positive and HIV-negative pregnant women, and provision of IPT for prevention of TB among eligible HIV-positive pregnant women in Lesotho, to inform the Ministry of Health (MOH). We abstracted data from selected women who were identified at their first visit for routine antenatal care (ANC) services. Study-specific data were abstracted from the women's initial and subsequent standardized forms and registers completed during receipt of routine services. Follow-up data were extracted until women on IPT reached the 6-month end-of-treatment mark.
Pregnant women were eligible to be included in the study if they presented for their first ANC visit for pregnancy during the study period, irrespective of their gestational age. Women were excluded if they had active TB, were on TB treatment, or were already receiving IPT. The minimum age for the women enrolled was 14 years. The study population was identified by a research nurse through review of the medical records and ANC register. There was no direct interaction between study participants and the research team. All study data were collected through review and abstraction of data from existing medical records.
Study sites were 2 hospital-based MCH clinics, Berea hospital in Berea District and Saint Joseph's hospital in Maseru District. These sites were purposively selected by MOH as the pilot sites for the rollout of ACF/IPT guidelines because of their relatively high volume of patients and their proximity to the capital city. Implementation of the program began in June of 2011, in Berea hospital, and in September 2011 in Saint Joseph's hospital. The research was started at the same time to accompany the implementation.
Lesotho Procedures on ACF and IPT for Pregnant Women Attending ANC Clinics
Pregnant women seeking ANC services received information about TB during the routine educational session that was provided to all ANC attendees. All pregnant women received standard medical care within the ANC, in line with the MOH guidelines, including HIV testing and antiretroviral (ARV) drugs for treatment and/or the prevention of mother-to-child transmission (PMTCT) of HIV. Pregnant women were assessed for signs and symptoms of TB by trained health care workers using the National TB screening tool that screened for the presence of fever, night sweats, cough, or weight loss. If the woman had any of these symptoms, she would be given 3 sputum collection containers, provide 1 sample immediately and bring the 2 remaining sputum samples to the laboratory for testing the following day. Sputum samples are processed daily, and AFB smear microscopy results are given to the patients the same day. Women with positive sputum tests for Acid Fast Bacilli (AFB) are referred to the TB clinic for initiation of treatment. Women with AFB-negative sputa are referred to a physician for further investigations and management within 1–2 days for consideration of empiric antibacterial therapy with amoxicillin or erythromycin and/or chest x-ray imaging and GenXpert testing that became available during the course of the study according to National guidelines. The study sites both had x-ray facilities on site, whereas the GenXpert facility was located about 30 km from Berea hospital and 40 km from St Joseph's hospital with samples transported centrally through the existing MOH sample transport system and results available within 2 days when the study is requested.
HIV-positive pregnant women with a negative symptom screen were eligible for IPT within ANC. They were prescribed 300 mg isoniazid (INH) tablets (1 daily) with vitamin B6 25 mg once a day and instructed to return to the facility monthly for clinical and laboratory monitoring, as well as to obtain refills of INH over the period of 6 months. Clinical monitoring included active assessment for side effects and toxicities related to INH, including gastrointestinal and central nervous system side effects, and peripheral neuropathy. Other routine testing included baseline alanine aminotransferase (ALT) measurement, and follow up ALT among patients with elevated ALT at baseline, or clinical symptoms suggestive of hepatitis.
A unique study identification number (ID) was assigned to all women enrolled in the study. The ID number was used on all study case report forms to ensure anonymity. The study ID was linked to the woman's routine medical record numbers on a locator form to allow research nurses to identify and extract data collected during subsequent ANC visits. Clinical data that were extracted included demographic information of the patient, HIV status, findings of symptoms screening, and concurrent medications. Extracted laboratory data that were included in this analysis were CD4 count, ALT, and sputum microscopy; radiology studies were infrequently ordered, and results were not available for analysis. Sources of information from which data were abstracted included TB treatment and IPT registers, the antiretroviral therapy (ART) card, ANC register, and the infant delivery register and postpartum register. All abstracted data were entered into a customized study database in duplicate by 2 separate individuals. All data queries were resolved before final data analysis.
Rates of TB diagnosis through ACF and subsequent laboratory testing, initiation of eligible HIV-infected pregnant women on IPT, incidence of side effects during IPT, and completion rates of IPT were determined using data abstracted from standardized forms and registers. Completion of IPT was based on the National guidelines definition, which calls for 6 consecutive months of daily INH. Among those who default during this period, IPT therapy completion is defined as reinitiation within a month of the first default to complete a total of 6 months of INH. Adherence was not assessed as part of this research.
This study's protocol was reviewed and approved by the Lesotho Ministry of Health ethics committee, the Baylor College of Medicine, Bristol Myers-Squibb Children's Clinical Centre of Excellence Lesotho, and the Johns Hopkins Medicine institutional review boards. We requested and obtained a waiver of consent to abstract existing data from routine clinic registers.
Sample Size and Statistical Consideration
A study sample size of 160 (20%) HIV-positive pregnant women and 640 (80%) HIV-negative pregnant women was required to detect a 10% or greater difference in the presence of TB symptoms among the 2 groups, with an alpha of 0.05% and 80% power, in a population with approximately 20% HIV prevalence. The sample size was inflated to accommodate an approximately 20% chart incompletion rate. This sample size was also sufficient to detect at least a 10% difference in TB screening rates among HIV-positive and HIV-negative women, and provide precision within 5% of the point estimate of the percentage of eligible HIV-positive women who were initiated on IPT.
Frequencies and medians were used to summarize the rollout of active TB case–finding and IPT. Uptake of active TB case–finding was defined as the proportion of pregnant women presenting for their first ANC visit who were screened for TB, as per MOH guidelines. Uptake of IPT was defined as the proportion of HIV-infected pregnant women presenting at 1 or more antenatal clinic visit who were eligible for IPT and started on IPT.
Differences in characteristics between HIV-positive and HIV-negative women were assessed using the χ2 test or Wilcoxon rank-sum test for categorical or continuous variables, respectively. Among HIV-positive women, the bivariate relationships between IPT initiation and factors such as maternal age, gestational age at the first ANC visit, maternal CD4 count, concurrent ART, and known HIV-positive status at the first ANC visit were estimated using logistic regression. All statistical analyses were conducted using STATA (version 11.0).
Between June 2011 and December 2012, records from 800 of 1763 eligible pregnant women with their first ANC visit during the enrollment period were consecutively reviewed until the enrollment targets of 160 HIV-positive and 640 HIV-negative pregnant women were met, as defined by sample size calculation (Fig. 1). Among study participants, 56 (35.0%) of HIV-positive and 50 (7.8%) of HIV-negative pregnant women had documented knowledge of their HIV status at their first ANC visit (Table 1). Among 694 pregnant women presenting without a documented HIV-negative result of less than 3 months or HIV-positive test, 104 (14.9%) tested HIV positive.
Table 1 summarizes characteristics of the women. When compared with HIV-negative women, HIV-positive pregnant women tended to be older (P = 0.001) and had higher numbers of previous pregnancies and deliveries (P < 0.001). Gestational age at the first ANC visit was slightly later in HIV-positive women (P = 0.076). About 99.8% of women were screened for TB using the clinical symptoms screen, with HIV-positive women more likely to report TB symptoms compared with HIV-negative women (11.3% versus 2.3%; P < 0.001). Two (1.3) HIV-positive women had confirmed TB compared with 1 (0.15%) HIV-negative pregnant women (P = 0.191). One of the women was AFB smear negative, but diagnosed on the basis of chest x-ray findings, second was diagnosed by GenXpert testing, and the third diagnosed with detection of AFB on sputum microscopy. Among women with symptoms of TB, with the predominant symptom being the presence of cough, 1200 per 100,000 cases of HIV-positive women compared with 200 patients per 100,000 cases of HIV-negative women were diagnosed with active TB disease.
Among 128 (80.0%) HIV-positive pregnant women who had a CD4+ T-lymphocyte cell count, 61.7% had a CD4+ T-lymphocyte cell count greater than 350 cells per cubic millimeter and 88.1% were WHO clinical stage 1 (Table 2). More than a third (36.2%) of these women received ART for their own health; the rest were on Zidovudine (AZT) prophylaxis to prevent mother-to-child transmission of HIV.
Among HIV-positive women, 158 (98.7%) were eligible for IPT, per clinical guidelines (Figs. 1, 2). Of these, 140 (88.6%) were asymptomatic and IPT was initiated in 111 (79.3%) asymptomatic women. Twenty-nine women with a negative clinical symptom screen who were eligible for IPT were not initiated on prophylaxis. Among these women, 15 were not initiated at the earliest clinical opportunity due to lack of availability of INH at the time of their visit and were unable to be tracked to be offered IPT when the medication became available; 5 women were referred to another program or facility; 3 women declined IPT; 2 women left the program without informing HCWs; 2 women moved to South Africa. The reason for lack of IPT initiation is unknown for 2 of the women.
In the symptomatic women, 13/18 (72.2%) received IPT after active TB were excluded. Of the 5 symptomatic women who did not receive IPT, 2 had active TB and were initiated on TB treatment. Among patients initiated on IPT, 80 (64.5%) completed the recommended 6-month course of therapy, and 3 (2.4%) were transferred out. Overall, a high loss to follow-up was observed, 39 (31.5%) and 2 (1.6%) died of causes unrelated to IPT or TB; 1 patient died from septic abortion and another from puerperal sepsis.
Of the 124 patients who were initiated on IPT, none reported any clinical side effects of IPT. Among 99 women who initiated IPT and had baseline ALT performed, 3 women had elevated ALT of 45, 43, 43 IU/L, respectively, at a mean gestational age of 22.2 weeks. The overall median ALT was 19 IU/L (interquartile range, 13–24) among women who had baseline ALT. Fewer women (N = 20) had a repeat ALT performed among whom 6 women had elevated ALT level of 43, 50, 53.2, 56, 60, and 81, IU/L, respectively. Overall, no significant elevation in ALT level was noted among women who had subsequent ALT testing, with a median of 28 IU/L (interquartile range, 20–38).
Among HIV-positive women, gestational age at the first ANC (unadjusted odds ratio, 0.93; 95% confidence interval: 0.88 to 0.98) and ARV use for treatment versus PMTCT prophylaxis (odds ratio, 4.59; 95% confidence interval: 1.32 to 15.93) were associated with initiation of IPT (Table 3). Late presentation to ANC was associated with a lower likelihood of initiating IPT; a 1-week increase in gestational age was associated with an approximately 7% decrease in the odds of IPT initiation. CD4 cell count above or below 350 cells per cubic millimeter was not associated with IPT initiation.
This study demonstrates that implementation of ACF of TB and IPT in the MCH setting is feasible in a resource-constrained setting. Despite the high volume of patients, health care workers within the MCH setting achieved almost universal screening of all pregnant women for clinical symptoms and signs of TB. Of note, the yield of ACF for the detection of TB primarily was low. Improved diagnostics with greater sensitivity than smear microscopy such as GeneXpert testing may identify more cases of TB, especially among HIV-positive patients with compatible clinical symptoms. Even in the context of pregnancy, most HIV-positive women agreed to initiate IPT to prevent the development of active TB. Given the nature of this study, which relied on existing clinical records, we were unable to ascertain the reason for lack of uptake of IPT among the approximately 20.0% of women who were eligible for IPT but did not receive it. Although it is possible that women did not accept the offer to initiate IPT, it is also feasible that this subset of women was not offered IPT by the health workers, especially during stock out of drugs as reported in previous studies among general adult population.19–21 Further studies including process analyses are warranted to identify reasons for not providing IPT to eligible HIV-positive pregnant women, and to identify interventions that could address structural and behavioral impediments to successful implementation of IPT.
Of those receiving IPT, two-thirds completed the recommended 6-month course of therapy, higher than rates reported in routine care in the United States and Canada.22 Although the completion rate was higher than 59.0% 6-month completion rate reported by studies in South Africa, it is concerning that a significant proportion of women did not complete the 6-month course of IPT prophylaxis.17 The high attrition identified in this evaluation demonstrates the need to critically evaluate processes to support IPT completion in this particular patient group. The suboptimal IPT completion in this sample population further highlights the importance of identifying alternative chemotherapeutic strategies for effective TB prophylaxis with a shorter duration of therapy in pregnancy.
This was a descriptive prospective study that used existing clinical data without interaction between the research staff and the participants. The quality of the routinely collected data was sufficient to conduct this evaluation because of timing of the release of the ICF/IPT/IC guidelines in Lesotho, which paralleled the revisions to the ANC and other clinical registers to support the implementation of the updated PMTCT and ART guidelines. As such, clinical forms were updated in such a way that it allowed inclusion of indicators needed in this evaluation. Given the study design, however, we were unable to implement interventions to optimize IPT management. Future program evaluations should be designed to address attrition on an ongoing basis, by examining and optimizing processes iteratively to ensure improvement in access and completion of IPT during pregnancy and postpartum. Some studies reported isoniazid (INH) side effects, health care workers' attitude, stock out of supplies, and migration as some of the factors that were associated with patients who failed to complete their 6-month course of IPT.21,23 Our findings suggest that side effects with IPT during pregnancy are not commonly similar to reports from other studies in Botswana and Brazil, further strengthening the case for use during pregnancy.14,24–27 However, further studies are needed to assess the safety of this intervention, when administered in the context of a routine program in places with high HIV prevalence and concurrent use of ART, as this study was not specifically designed or powered to assess the safety of IPT in this context. In studies among the general adult population receiving isoniazid, less than 2% experience significant side effects.23
Interestingly, women who were on ART for their own health were more likely to receive IPT compared with women receiving ARV prophylaxis for PMTCT. Reasons for this may be due to the more regular interaction with the health care system, and/or a perception (by either the woman or the health care worker) that IPT is important for their health. It may also be due to the fact that this subset of women may be more informed through posttest support and ART adherence counseling.
To address issues of mobility and challenges of retention in care, Lesotho rolled out a program to optimize PMTCT that involved copackaging ARV PMTCT prophylaxis together with iron, folic acid, vitamin B complex, and vitamin A that was given to pregnant women for the whole duration of pregnancy.28 Adding IPT to this pack may be an innovative way to improve the completion of IPT 6-month course in pregnancy and beyond; however, the safety of such an intervention would have to be independently investigated, especially in light of the high loss to follow-up that was demonstrated in this feasibility study.
The authors acknowledge that this research was made possible by Bill and Melinda Gates Foundation Grant 19790.01 (Gates) for the Consortium to Response Effectively to the AIDS-Tuberculosis Epidemic (CREATE), through a subcontract from Johns Hopkins University (JHU). The authors thank the staff from Berea and St Joseph's hospitals that made the study possible. They specifically thank Makhohlisa Matela, Mathabo Mareka, Mothabeng David, Masuoane Mamello, Mohale Sesomo, and Lekhosa Thabiso. They thank Drs Glenda Gray, Jonathan Fuchs, William McFarland, and Jeff Mandel for their assistance as mentors to AT through a Manuscript Writing Workshop organized by the NIH Office of AIDS Research-supported OCTAVE Project, NIMH-supported UCSF ITAPS Program (R25MH064712-08), and the Fogarty International Center AITRP Program (D43TW00003).
1. World Health Organization. TB/HIV Facts 2012–2013. 2012.
2. World Health Organization. Tuberculosis Women and TB. 2009.
3. Mnyani CN, McIntyre JA. Tuberculosis in pregnancy. BJOG. 2011;118:226–231.
4. Bassett IV, Wang B, Chetty S, et al.. Intensive tuberculosis screening for HIV-infected patients starting antiretroviral therapy in Durban, South Africa. Clin Infect Dis. 2010;51:823–829.
5. Gounder CR, Wada NI, Kensler C, et al.. Active tuberculosis case-finding among pregnant women presenting to antenatal clinics in Soweto, South Africa. J Acquir Immune Defic Syndr. 2011;57:e77–e84.
6. Khan M, Pillay T, Moodley JM, et al.. Maternal mortality associated with tuberculosis-HIV-1 co-infection in Durban, South Africa. AIDS. 2001;15:1857–1863.
7. Gupta A, Bhosale R, Kinikar A, et al.. Maternal tuberculosis: a risk factor for mother-to-child transmission of human immunodeficiency virus. J Infect Dis. 2011;203:358–363.
8. World Health Organization. Guidelines for Intensified Case Finding and Isoniazid Preventive Therapy for People Living with HIV in Resource-constrained Settings. 2011.
9. Kali PB, Gray GE, Violari A, et al.. Combining PMTCT with active case finding for tuberculosis. J Acquir Immune Defic Syndr. 2006;42:379–381.
10. Thorson A, Hoa NP, Long NH, et al.. Do women with tuberculosis have a lower likelihood of getting diagnosed? Prevalence and case detection of sputum smear positive pulmonary TB, a population-based study from Vietnam. J Clinical Epidemiology. 2004;57:398–402.
11. Sheriff FG, Manji KP, Manji MP, et al.. Latent tuberculosis among pregnant mothers in a resource poor setting in Northern Tanzania: a cross-sectional study. BMC Infect Dis. 2010;10:52.
12. Keskin N, Yilmaz S. Pregnancy and tuberculosis: to assess tuberculosis cases in pregnancy in a developing region retrospectively and two case reports. Arch Gynecol Obstet. 2008;278:451–455.
13. Getahun H, Sculier D, Sismanidis C, et al.. Prevention, diagnosis, and treatment of tuberculosis in children and mothers: evidence for action for maternal, neonatal, and child health services. J Infect Dis. 2012;205(suppl 2):S216–S227.
14. Durovni B, Cavalcante SC, Saraceni V, et al.. The implementation of isoniazid preventive therapy in HIV clinics: the experience from the TB/HIV in Rio (THRio) study. AIDS. 2010;24(suppl 5):S49–S56.
15. Munseri PJ, Talbot EA, Mtei L, et al.. Completion of isoniazid preventive therapy among HIV-infected patients in Tanzania. Int J Tuberc Lung Dis. 2008;12:1037–1041.
16. Khawcharoenporn T, Apisarnthanarak A, Manosuthi W, et al.. Isoniazid preventive therapy and 4-year incidence of pulmonary tuberculosis among HIV-infected Thai patients. Int J Tuberc Lung Dis. 2012;16:336–341.
17. Golub JE, Pronyk P, Mohapi L, et al.. Isoniazid preventive therapy, HAART and tuberculosis risk in HIV-infected adults in South Africa: a prospective cohort. AIDS. 2009;23:631–636.
18. Ministry of Health Lesotho. National Guidelines for 3Is, IPT/ICF/IC. 1st ed. 2011.
19. Bristow CC, Larson E, Vilakazi-Nhlapo AK, et al.. Scale-up of isoniazid preventive therapy in PEPFAR-assisted clinical sites in South Africa. Int J Tuberc Lung Dis. 2012;16:1020–1022.
20. Chehab JC, Vilakazi-Nhlapo K, Vranken P, et al.. Survey of isoniazid preventive therapy in South Africa, 2011. Int J Tuberc Lung Dis. 2012;16:903–907.
21. Getahun H, Granich R, Sculier D, et al.. Implementation of isoniazid preventive therapy for people living with HIV worldwide: barriers and solutions. AIDS. 2010;24(suppl 5):S57–S65.
22. Horsburgh CR Jr, Goldberg S, Bethel J, et al.. Latent TB infection treatment acceptance and completion in the United States and Canada. Chest. 2010;137:401–409.
23. Kwara A, Herold JS, Machan JT, et al.. Factors associated with failure to complete isoniazid treatment for latent tuberculosis infection in Rhode Island. Chest. 2008;133:862–868.
24. Taylor AW, Mosimaneotsile B, Mathebula U, et al.. Pregnancy outcomes in HIV-infected women receiving long-term isoniazid prophylaxis for tuberculosis and antiretroviral therapy. Infect Dis Obstet Gynecol. 2013;2013:195637.
25. Ouyang DW, Shapiro DE, Lu M, et al.. Increased risk of hepatotoxicity in HIV-infected pregnant women receiving antiretroviral therapy independent of nevirapine exposure. AIDS. 2009;23:2425–2430.
26. Lyons F, Hopkins S, Kelleher B, et al.. Maternal hepatotoxicity with nevirapine as part of combination antiretroviral therapy in pregnancy. HIV Med. 2006;7:255–260.
27. Rangaka MX, Wilkinson RJ, Boulle A, et al.. Randomized controlled trial of isoniazid preventive therapy in HIV-infected persons on antiretroviral therapy. 19th International AIDS Conference; 2012.
28. McDougal L, Moteetee MM, Mohai F, et al.. Lesotho's minimum PMTCT package: lessons learned for combating vertical HIV transmission using co-packaged medicines. J Int AIDS Soc. 2012;15:17326.