Skip Navigation

Postpartum Tuberculosis Incidence and Mortality among HIV-Infected Women and Their Infants in Pune, India, 2002–2005

  1. Amita Gupta1,
  2. Uma Nayak1,
  3. Malathi Ram1,
  4. Ramesh Bhosale3,
  5. Sandesh Patil3,
  6. Anita Basavraj3,
  7. Arjun Kakrani3,
  8. Sheeja Philip4,
  9. Dipali Desai3,
  10. Jayagowri Sastry4,
  11. Robert C. Bollinger1,2, and
  12. for the Byramjee Jeejeebhoy Medical College–Johns Hopkins University Study Group
  1. 1Infectious Diseases, Johns Hopkins University School of Medicine, Baltimore, Maryland
  2. 2Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland
  3. 3Byramjee Jeejeebhoy Medical College, Pune, India
  4. 4Byramjee Jeejeebhoy Medical College–Johns Hopkins University Maternal Infant Transmission Study, Pune, India
  1. Reprints or correspondence: Dr. Amita Gupta, Centre for Clinical Global Health Education, Division of Infectious Diseases, 600 N. Wolfe St., Phipps 540B, Baltimore, MD 21287 (agupta25{at}jhmi.edu).
 
Next Section

Abstract

Background. In contrast with many other countries, isoniazid preventative therapy is not recommended in clinical care guidelines for human immunodeficiency virus (HIV)—infected persons with latent tuberculosis (TB) in India.

Methods. Seven hundred fifteen HIV-infected mothers and their infants were prospectively followed up for 1 year after delivery at a public hospital in Pune, India. Women were evaluated for active TB during regular clinic visits, and tuberculin skin tests were performed. World Health Organization definitions for confirmed, probable, and presumed TB were used. Poisson regression was performed to determine correlates of incident TB, and adjusted probabilities of mortality were calculated.

Results. Twenty-four of 715 HIV-infected women who were followed up for 480 postpartum person-years developed TB, yielding a TB incidence of 5.0 cases per 100 person-years (95% confidence interval [CI], 3.2–7.4 cases per 100 person-years). Predictors of incident TB included a baseline CD4 cell count <200 cells/mm3 (adjusted incident rate ratio [IRR], 7.58; 95% CI, 3.07–18.71), an HIV load <50,000 copies/mL (adjusted IRR, 3.92; 95% CI, 1.69–9.11), and a positive tuberculin skin test result (adjusted IRR, 3.08; 95% CI, 1.27–7.47). Three (12.5%) of 24 women with TB died, compared with 7 (1.0%) of 691 women without TB (IRR, 12.2; 95% CI, 2.03–53.33). Among 23 viable infants with mothers with TB, 2 received a diagnosis of TB. Four infants with mothers with TB died, compared with 28 infants with mothers without TB (IRR, 4.71; 95% CI, 1.19–13.57). Women with incident TB and their infants had a 2.2- and 3.4-fold increased probability of death, respectively, compared with women without active TB and their infants, controlling for factors independently associated with mortality (adjusted IRR, 2.2 [95% CI, 0.6–3.8] and 3.4 [95% CI, 1.22–10.59], respectively).

Conclusions. Among Indian HIV-infected women, we found a high incidence of postpartum TB and associated postpartum maternal and infant death. Active screening and targeted use of isoniazid preventative therapy among HIV-infected women in India should be considered to prevent postpartum maternal TB and associated mother-to-child morbidity and mortality.

Tuberculosis (TB) is the most common HIV-related opportunistic infection and the most important cause of morbidity and mortality in HIV-infected adults in the developing world [1]. In areas with a high prevalence of HIV infection and TB, active case finding for TB has revealed that TB is also a concern during pregnancy and at the time of delivery of a newborn [2, 3]. Data suggest that both TB and HIV infection have been identified as independent factors for maternal mortality and perinatal outcomes; in combination, TB and HIV infection have a greater detrimental impact than their individual effects [4, 5]. In the pre-HIV era, <6% of maternal mortality was shown to be caused by TB [6]. In the HIV era, this statistic has dramatically changed. Coinfection with TB and HIV has been shown to account for ∼15% of maternal deaths in teaching hospitals in southern Africa [5, 7,8–9]. Furthermore, newborns exposed to an adult with active TB may be at high risk for TB morbidity and mortality, but the risk has been poorly quantified [10, 11].

India has the highest TB burden globally, accounting for <20% of the 8.8 million global cases each year, and up to 50% of India's population is estimated to have latent TB [12, 13]. TB kills ∼370,000 Indian adults annually; most of these persons are aged 15–54 years, and TB is a major cause of death in women of reproductive age [12]. India is also estimated to be one of the countries with the largest number of HIV-infected persons in the world (5.2–5.7 million) [14], and TB is by far the most important cause of morbidity and mortality in Indian HIV-infected persons, accounting for 50% of deaths.

Although there is a high burden of TB and high risk for active TB among HIV-infected persons in India, India's national HIV and TB guidelines do not promote isoniazid preventive therapy (IPT) for HIV-infected persons [15]. There is legitimate concern that widespread and inappropriate use of IPT could increase the community burden of isoniazid-resistant TB infection and limit the success of the national TB-control program. The objective of our study was to determine the incidence of active TB infections among postpartum Indian women and to assess the impact of maternal HIV and TB coinfection on maternal and infant outcomes in the absence of IPT. Such data are needed, because delivery and the postpartum period represent important entry points for women and their infants into the health care system and, therefore, ideal times to consider an intervention that includes TB prevention and management.

Previous SectionNext Section

Methods

From 16 August 2002 through 31 December 2005, HIV-infected mother and infant pairs were enrolled and prospectively followed up as part of an ongoing National Institutes of Health—funded phase III randomized clinical trial to assess the role of nevirapine therapy administered to infants during breastfeeding. Although the primary outcome of HIV transmission is pending completion of the trial, this study is an analysis of the data specific to TB that were collected as part of the secondary outcome of maternal and infant morbidity and mortality.

HIV-infected women at a large urban, public teaching hospital in Pune, Maharashtra, India, were enrolled during their third trimester, at delivery, or within 1 week after delivery. Mother and infant pairs were followed up prospectively for up to 12 months after delivery. At each scheduled visit (at weeks 1, 2, 3, 4, 5, 6, 10, and 14 and at 6, 9, and 12 months), women and their infants underwent clinical examination and select laboratory investigations. All infants were given bacille Calmette-Guérin vaccination at birth, according to Indian pediatric guidelines. HIV infection in infants was confirmed by HIV DNA PCR testing and quantitative HIV load measurements. The maternal baseline hemoglobin concentration was obtained at enrollment in the study, and the CD4 cell count and HIV load measurement that were obtained closest to delivery were used in this analysis.

Informed consent was obtained from all participants in the study. Human experimentation guidelines of the US Department of Health and Human Services and the participating institutions were observed in the conduct of this research. Informed consent procedures and this research were reviewed and approved by independent ethical committees in Pune, India, and in the United States.

TB screening and diagnosis. At delivery, women were offered tuberculin skin tests (TSTs; also known as Mantoux tests), according to standard methods [16], and were evaluated during regular clinic visits for active TB. If the TST results were positive (defined as an induration ⩾5 mm in diameter for HIV-infected persons) [17] or symptoms suggested active TB (e.g., persistent cough, fever, and weight loss), chest radiographs were performed and clinical and laboratory information were obtained using a standardized data collection form. TB was “confirmed” when Mycobacterium tuberculosis was cultured from the mother, and TB was “probable” when (1) acid-fast bacilli were detected in maternal sputum smears by microscopic examination, (2) histological features suggested TB, or (3) congenital TB was confirmed after the culture of M. tuberculosis from the neonate. TB was “suspected” when the mother had only clinical and radiological evidence suggesting TB and showed a response to anti-TB therapy. Some microbiological data were missing. Culture is not a routine practice in India and, according to national TB guidelines, is only recommended when there is treatment failure or lack of response to treatment; for 7 cases, either the clinician did not send appropriate samples for culture or appropriate media was not available, and for 4 cases, sputum samples for acid-fast bacilli testing were not adequately obtained. A diagnosis of TB in infants was determined according to standard methods [18]. Because of the difficulty of confirming cases, most infant TB cases were determined using clinical and radiological criteria (including response to treatment), as well as history of close contact to persons with active TB.

Analyses. Data were analyzed using Stata statistical software, version 9.0 (Stata). Student's t test for means were used when continuous variables were normally distributed. Nonparametric Mann-Whitney U tests or Fisher exact tests were used when continuous variables were not normally distributed or when cell size was ⩽5 observations, respectively; χ2 tests were used for discrete variables. Log-linear (Poisson) regression analyses were used with TB incidence as the outcome variable. Continuous variables were categorized into clinically meaningful groups (i.e., CD4 cell count <200 cells/mm3 or ⩾200 cells/mm3). Age, marital status, employment status, family type, education, TST status, CD4 cell count, HIV load, and maternal hemoglobin concentration were covariates. Collinearity was checked by calculating the variance inflation factors, which were all <2.5. Variables that were significantly associated in the univariate analysis were analyzed using multivariate Poisson regression. Forward and backward stepwise selection of covariates was performed using a significance level of P ⩽ .05. Interactions between age and other covariates were explored. Model fit was assessed by using the Pearson's goodness-of-fit test, as well as Akaike's information criterion [19]. Overdisperson was not evident in our model. Both forward and backward selection arrived at the same model as did Akaike's information criterion, with the exception of 1 variable (maternal hemoglobin concentration). This variable was forced into the model, because it has been identified as an independent predictor in published studies. Lastly, adjusted probabilities for maternal and infant mortality by maternal TB status were calculated, controlling for all other statistically significant covariates using Poisson regression. All continuous variables were centered. To calculate infant mortality probabilities, we added key available covariates for infant mortality, such as birth weight (low birth weight, <2500 g), HIV PCR status, gestational age (preterm, <38 weeks), and infant mode of feeding.

Previous SectionNext Section

Results

A total of 715 HIV-infected women were followed up for 480.6 person-years (PY; median duration of follow-up, 0.96 PY; interquartile range [IQR], 0.27–1.0 PY). The characteristics of the cohort are shown in table 1. of 715 women, 688 (96.2%) had a TST administered around the time of delivery and appropriately read within 72 h of delivery; 145 of these women (21.1%) had positive TST results. Twenty-four postpartum incident TB infections were detected, yielding an incidence estimate of 5.0 cases per 100 PY (95% CI, 3.2-7.4 cases per 100 PY), with a median time after delivery to incident TB of 90 days (range, 4–333 days). Seven cases were confirmed within the first 2 weeks after delivery, suggesting that the women may have had subclinical TB at the time of delivery. Fifteen women (62.5%) had a baseline CD4 cell count <200 cells/mm3, 15 (62.5%) had a baseline HIV load <50,000 copies/mL, and 9 (37.5%) had positive TST results at the delivery screening. Sixteen women received a diagnosis of pulmonary TB, 6 received a diagnosis of extra-pulmonary TB, and 2 had both of these types of TB. Six cases were “confirmed”, 12 were “probable”, and 6 were “suspected” (12 [60%] of 20 women were smear-positive for TB, and 6 [35.3%] of 17 were culture-positive for TB). All but 3 women initiated TB treatment (1 was lost to follow-up, and the other 2 died before therapy was initiated). Seven women completed treatment and were cured, 10 had treatment ongoing at the time of analysis, and 4 discontinued treatment. Three women (11.5%) died at 5, 6, and 11 months after delivery; 2 of these deaths were due to TB, and 1 was due to bacterial sepsis.

Figure 1
View larger version:
Figure 1

Adjusted maternal and infant mortality incidence rate ratios (IRRs) within the first year after delivery, by maternal tuberculosis (TB) status, among a cohort of HIV-infected women and their infants in Pune, India (2002–2005). The unadjusted IRR for mothers was 12.2 (95% CI, 2.0–53.3; P = .004), and the unadjusted IRR for infants was 4.7 (95% CI, 1.2–13.6; P = .017). The maternal IRR was adjusted for CD4 cell count, log HIV load, hemoglobin concentration, age, and educational status at the time of delivery of the infant. Infant mortality was adjusted for HIV PCR status during the first year of life, preterm birth (<38 weeks), low birth weight (<2500 g), infant mode of feeding, and the aforementioned maternal factors. Mortality IRRs and 95% CIs were calculated using Poisson regression. aIRR, adjusted incidence rate ratio; PY, person-years.

Table 1
View larger version:
Table 1

Characteristics of patients and unadjusted and adjusted incidence rate ratios (IRRs) of maternal tuberculosis (TB) in a cohort of HIV-infected women in Pune, India, 2002–2005.

Compared with women who did not develop TB during the study period, women with incident TB were more likely to be unemployed (41.7% vs. 19.8 %; P = .009, by χ2 test), to have a lower median baseline CD4 cell count (333 cells/mm3 vs. 472 cells/mm3; P = .001), and to have a higher median baseline HIV load (69,208 copies/mL vs. 11,430 copies/mL) (table 1). Women who developed TB were also more likely than women who did not develop TB to have positive TST results (37.5% vs. 20.5%; P = .04, by χ2 test) and to have at least moderate anemia (hemoglobin concentration, ⩽9 g/dL; 37.5% vs. 19.1%; P = .03, by χ2 test).

Correlates for incident maternal TB. In univariate Poisson regression, the following baseline characteristics were associated with TB incidence: older age, being employed, a CD4 cell count ⩽200 cells/mm3, an HIV load <50,000 copies/mL, positive TST results, and a hemoglobin concentration ⩽9 g/dL (table 1).

In multivariate Poisson regression, a CD4 cell count ⩽200 cells/mm3 was the strongest independent predictor and was associated with a 7.58-fold increased incidence of TB (adjusted incidence rate ratio [IRR], 7.58; 95% CI, 3.07–18.71). An HIV load <50,000 copies/mL (adjusted IRR, 3.92; 95% CI, 1.69–9.11), being employed (IRR, 3.00; 95% CI, 1.32–6.79), and positive TST results (adjusted IRR, 3.08; 95% CI, 1.27–7.47) were also independently associated with incident maternal TB infection. Within 12 months after delivery of an infant, 3 (12.5%) of 24 mothers with TB died, compared with 7 (1.0%) of 691 mothers without TB (IRR, 12.2; 95% CI, 2.03–53.33). Adjusting for CD4 cell count, HIV load, hemoglobin concentration, education, and age, the probability of maternal mortality within the first year after delivery was 3-fold higher among women with postpartum incident TB than among those who did not develop incident TB (0.9 cases per 100 PY vs. 0.4 cases per 100 PY; adjusted IRR, 2.2; 95% CI, 0.6–3.8) (figure 1).

Infant characteristics and outcomes. Infants born to women who subsequently developed TB did not differ from infants born to women who did not develop incident TB with regard to median birth weight (2500 g vs. 2600 g; P = .66), median gestational age (38 weeks vs. 38 weeks), and median duration of breastfeeding (105.5 days vs. 100 days) (table 2). Infants with mothers with incident TB, however, were more likely to have HIV infection (determined by infant HIV DNA PCR) by 1 year of age than were infants with mothers who did not have incident TB (37.5% vs. 9.1%; P < .001). Thirteen infants received TB prophylaxis after their mothers developed TB with either a combination of isoniazid and rifampin (administered to 12 infants) or isoniazid alone (administered to 1 infant). Three infants did not complete a full course of prophylaxis. The remaining infants did not receive prophylaxis because parents refused treatment (1 infant), the mother had extrapulmonary TB (4 infants) or was smear-negative for TB (2 infants), the infant was lost to follow-up (1 infant), or the infant died before prophylaxis was initiated (3 infants). Two infants were suspected to have developed active TB (a third suspected case occurred in an infant whose father, but not mother, received a diagnosis of TB; data not shown). Four (16.7%) of 24 infants of mothers with TB died, compared with 28 (4.1%) of 691 infants of mothers who did not have TB (P < .001, by χ2 test). The unadjusted infant mortality rates were 25.8 cases per 100 PY for infants with mothers with TB, compared with 5.9 cases per 100 PY for infants with mothers without TB (IRR, 4.71; 95% CI, 1.19–13.57). Adjusting for infant gestational age, low birth weight, infant mode of feeding, and infant HIV status, as well as for maternal factors, such as CD4 cell count, hemoglobin concentration, and HIV load, the probability of infant mortality was 8.5 cases per 100 PY among infants with mothers who developed incident TB, compared with 2.5 cases per 100 PY among infants whose mothers did not develop TB (adjusted IRR, 3.4; 95% CI, 1.22–10.59) (figure 1).

Table 2
View larger version:
Table 2

Characteristics and outcomes of infants born to HIV-infected women, by maternal incident tuberculosis (TB) status, in a cohort in Pune, India, 2002–2005.

Previous SectionNext Section

Discussion

In a cohort of postpartum HIV-infected women in India, we identified a high maternal TB incidence of 5 cases per 100 PY in the absence of IPT and found that many cases developed early in the postpartum period. Our incidence rate is consistent with the 3%–10% annual incidence estimated by other studies, including 1 study in India [20,21–22]. We also confirmed several findings involving postpartum HIV-infected women that have been described in other cohorts involving nonpregnant adult HIV-infected women. A CD4 cell count <200 cells/mm3 was the strongest predictor of incident TB, associated with an 8-fold increased incidence of TB, compared with a CD4 cell count ⩾200 cells/mm3 [23]. However, nearly two-thirds of TB cases occurred in women with a CD4 cell count <200 cells/mm3, which is above the threshold recommended for HAART initiation in India. Therefore, many postpartum Indian women who are at high risk for TB would not benefit from the reduction in TB incidence that has been demonstrated with a combination of HAART and IPT or HAART alone [24, 25]. We also determined that a baseline HIV load <50,000 copies/mL was independently associated with a 3-fold increased risk of incident TB [26, 27]. Finally, an important finding was the impact that maternal postpartum TB had on overall infant mortality within the first year of life. Infants with mothers who developed incident TB had a 3-fold increased mortality rate, compared with infants whose mothers did not have TB, controlling for infant HIV status and several other important potential confounders.

There are limited published data about postpartum TB in the HIV era. A study of 120 South African women with positive TST results (screening performed after delivery) found that 11% of the women had active TB [3]. In our study, 7 cases were confirmed within the first 2 weeks after delivery; less than one-half of the women in whom these cases occurred had a positive TST result. This finding is concerning, because it suggests that these women may have had subclinical active TB at the time of delivery. Some researchers have also suggested that the early postpartum period may be a time of higher risk because of the nutritional stress of lactation, lack of sleep, rapid hormonal changes, and depression of cell-mediated immunity [28, 29]. Our finding that the median onset of maternal TB was 3 months after delivery lends some credence to the early postpartum period being a time of higher risk for developing TB, but larger studies are needed to definitively address this controversy [29, 30]. Small studies have suggested that TB and HIV coinfection—associated deaths have accounted for ∼15% of maternal mortality within the first year after delivery in teaching hospitals in southern Africa [7, 8]. We found that 20% of HIV-associated maternal deaths were due to complications of TB.

Maternal illness has an important influence on infant health, but few studies have examined the impact of maternal TB on infant outcomes in the HIV era [31]. Although our sample was small, our study highlights that infants with mothers with incident TB have higher all-cause mortality. Although some of the infant deaths are likely to be attributable to TB, it is biologically plausible that infants with mothers with poor nutritional status or illness due to active TB would have an increased risk of death [11].

Recent data reveal that providing IPT to HIV-infected children is beneficial for preventing mortality in areas where the prevalence of TB is high [32]. Prophylaxis is generally safe, and isoniazid for 6 months or isoniazid plus rifampin for 3 months are typically administered to TB-exposed infants in India. In our study, 72% of HIV-exposed infants who met criteria for prophylaxis received it. The rest of the infants did not have therapy initiated, because they were lost to follow-up, their parents refused treatment, or the infants died before therapy could be initiated. Furthermore, 23% of the infants did not complete therapy, suggesting that this strategy of preventing infant TB and mortality is suboptimal. Thus, consideration of the role of IPT for mothers as a strategy for reducing TB incidence in mothers, as well as morbidity and mortality in infants, may be useful. Currently, many guidelines recommend screening for TB during pregnancy using TSTs.

In India, the TST is not used as a routine screening test, and its use is limited to the diagnosis of TB in children and to epidemiological surveys. A 15-year follow-up of 280,000 HIV-uninfected individuals revealed that TST response is associated with development of active TB [33]. We also found that a positive TST result was an independent predictor of incident postpartum maternal TB, but the test had poor sensitivity (37.5%) and positive predictive value (6.2%), although 92.9% of women had a CD4 cell count <200 cells/mm3. Our findings are consistent with previous studies demonstrating poor predictive value of TST in regions where TB is highly endemic [34]. However, it is possible that many of these women were recently infected with HIV. Because our follow-up of these women was limited to 12 months after delivery, our study did not assess the longer-term predictive value of TST screening among this population. Despite its overall low sensitivity, TST may still have a role in identifying who is most likely to benefit from IPT.

The optimal screening strategy and timing of IPT initiation in HIV-infected pregnant women is not known but likely varies in different regions, depending on access to prenatal care, TB risk and incidence, and relative willingness of the patient to receive IPT. Current US guidelines recommend delaying IPT until after delivery, largely because of concern that pregnancy may be a risk factor for isoniazid-induced hepatitis [1]. This concern is largely based on a retrospective study of Hispanic women, in which a 2.5-fold increased risk of isoniazid-induced hepatitis and a 4-fold increased mortality rate occurred, but both of these effects were not statistically significant [35]. In the absence of controlled trials to determine the optimal time to initiate IPT, a Markov decision analysis of HIV-uninfected women found that IPT was most effective when initiated during the antepartum period (at 20 weeks gestation) and that the benefits outweighed the risks if the case-fatality rate was <0.45% [36]. The benefit would be expected to be higher for HIV-infected women. Because many cases in our study were diagnosed around the time of delivery and during the early postpartum period, a strategy of antepartum IPT initiation would likely be most effective. Assuming published efficacies for IPT and our results, irrespective of CD4 cell count or TST results, 56 women would have to receive IPT during the antepartum period to avert 1 case of TB (table 3) [24, 25, 37]. Using a more targeted approach that involves administering IPT to women with positive TST results and CD4 cell counts ⩾200 cells/mm3 and administering HAART plus IPT to women with CD4 cell counts <200 cells/mm3, 5–23 women would need to be given IPT to avert 1 case of TB. Thus, 30% of antepartum women, who account for 63% of patients with incident TB, would be targeted. With any strategy, stringent efforts to rule out active TB would be necessary (e.g., using sensitive symptom screening, including cough, fever, and weight loss; chest radiography; and, ideally, collection of sputum samples for acid-fast bacilli testing).

Table 3
View larger version:
Table 3

Expected reduction in postpartum incident tuberculosis (TB) cases with isoniazid preventive therapy (IPT) initiated during the antepartum period, using different criteria for targeted intervention in an Indian antenatal clinic population.

Even in the HAART era, TB will remain an important cause of morbidity and mortality [24, 38]. In the absence of preventive TB vaccines, the United Nations AIDS Program and World Health Organization, as well as the Centers for Disease Control and Prevention guidelines, currently recommend IPT for HIV-infected adults with latent TB [17, 39]. Targeted IPT has been shown to reduce the risk of active TB by up to 60% both in regions where TB is endemic and in regions where it is not endemic [37, 39, 40]. Despite India's large HIV and TB burden, IPT is not currently recommended. Indian HIV-infected women have a high risk of TB during the postpartum period, regardless of CD4 cell count, which is associated with the need for infant TB prophylaxis and high rates of postpartum maternal and infant death. Therefore, although widespread use of IPT in India may be problematic, targeted administration of IPT to antepartum HIV-infected women is likely to have great benefit for preventing postpartum maternal TB and to reduce the associated high incidence of morbidity and mortality caused by mother-to-child transmission of infection.

Previous SectionNext Section

Acknowledgments

We thank Shuchi Anand, Joline Choi, and all of the Maternal Infant Transmission Study participants and clinical staff.

Financial support. The US National Institutes of Health (NIH; R01 AI45462) and the Fogarty International Center/USNIH (2 D 43 TW000010-19-AITRP). This study was undertaken in collaboration with Byramjee Jeejeebhoy Medical College in Pune, India.

Potential conflicts of interest. All authors: no conflicts.

Previous SectionNext Section

Footnotes

  • The views expressed in this manuscript do not necessarily represent the views of the National Institutes of Health, Fogarty International Centre, the Johns Hopkins University, or Byramjee Jeejeebhoy Medical College.

  • Received January 14, 2007.
  • Accepted March 13, 2007.
Previous Section
 

References

  1. ↵
    1. World Health Organization
    . HIV/TB clinical manual. 2nd ed. Geneva, Switzerland: World Health Organization; 2004.
  2. ↵
    1. Kali PB,
    2. Gray GE,
    3. Violari A,
    4. Chaisson RE,
    5. McIntyre JA,
    6. Martinson NA
    . Combining PMTCT with active case finding for tuberculosis. J Acquir Immune Defic Syndr 2006;42:379-81.
    CrossRefMedlineWeb of Science
  3. ↵
    1. Nachega J,
    2. Coetzee J,
    3. Adendorff T,
    4. et al
    . Tuberculosis active case-finding in a mother-to-child HIV transmission prevention programme in Soweto, South Africa. AIDS 2003;17:1398-400.
    CrossRefMedlineWeb of Science
  4. ↵
    1. Mofenson LM,
    2. Rodriguez EM,
    3. Hershow R,
    4. et al
    . Mycobacterium tuberculosis infection in pregnant nonpregnant women infected with HIV in the Women Infants Transmission Study. Arch Intern Med 1995;155:1066-72.
    Abstract/FREE Full Text
  5. ↵
    1. Pillay T,
    2. Khan M,
    3. Moodley J,
    4. Adhikari M,
    5. Coovadia H
    . Perinatal tuberculosis HIV-1: considerations for resource-limited settings. Lancet Infect Dis 2004;4:155-65.
    CrossRefMedlineWeb of Science
  6. ↵
    1. Juneja Y,
    2. Goel U,
    3. Sood M
    . Changing trends in maternal mortality over a decade. Int J Gynaecol Obstet 1994;46:265-9.
    CrossRefMedline
  7. ↵
    1. Ahmed Y,
    2. Mwaba P,
    3. Chintu C,
    4. Grange JM,
    5. Ustianowski A,
    6. Zumla A
    . A study of maternal mortality at the University Teaching Hospital, Lusaka, Zambia: the emergence of tuberculosis as a major non-obstetric cause of maternal death. Int J Tuberc Lung Dis 1999;3:675-80.
    MedlineWeb of Science
  8. ↵
    1. Khan M,
    2. Pillay T,
    3. Moodley J,
    4. Connolly C
    . Maternal mortality associated with tuberculosis-HIV coinfection in Durban, South Africa. Ann N Y Acad Sci 2000;918:367-9.
    CrossRefMedlineWeb of Science
  9. ↵
    1. Pillay T,
    2. Khan M,
    3. Moodley J,
    4. et al
    . The increasing burden of tuberculosis in pregnant women, newborns infants under 6 months of age in Durban, KwaZulu-Natal. S Afr Med J 2001;91:983-7.
    MedlineWeb of Science
  10. ↵
    1. Marais BJ,
    2. Gie RP,
    3. Schaaf HS,
    4. et al
    . The natural history of childhood intra-thoracic tuberculosis: a critical review of literature from the pre-chemotherapy era. Int J Tuberc Lung Dis 2004;8:392-402.
    MedlineWeb of Science
  11. ↵
    1. Marais BJ,
    2. Gie RP,
    3. Schaaf HS,
    4. et al
    . The clinical epidemiology of childhood pulmonary tuberculosis: a critical review of literature from the pre-chemotherapy era. Int J Tuberc Lung Dis 2004;8:278-85.
    MedlineWeb of Science
  12. ↵
    1. TB Control of India
    . TB India 2006 RNTCP status report—DOTS for all: all for DOTS. Available at: http://www.tbcindia.org/documents.asp#. Accessed 28 June 2006.
  13. ↵
    1. World Health Organization (WHO)
    . Global tuberculosis control—surveillance, planning, financing: WHO report 2006. Geneva, Switzerland: WHO; 2006.
  14. ↵
    1. United Nations AIDS Program
    . 2006 Report on the global AIDS epidemic. 2006. Available at: http://www.unaids.org/en/HIV_data/2006GlobalReport/default.asp. Accessed 12 July 2006.
  15. ↵
    1. Padmapriyadarsini C,
    2. Swaminathan S
    . Preventive therapy for tuberculosis in HIV-infected individuals. Indian J Med Res 2005;121:415-23.
    MedlineWeb of Science
  16. ↵
    1. Huebner RE,
    2. Schein MF,
    3. Bass JB Jr
    . The tuberculin skin test. Clin Infect Dis 1993;17:968-75.
    FREE Full Text
  17. ↵
    Targeted tuberculin testing, treatment of latent tuberculosis infection American Thoracic Society. MMWR Recomm Rep 2000;49(RR-6):1-51.
    Medline
  18. ↵
    1. Enarson PM,
    2. Enarson DA,
    3. Gie R
    . Management of tuberculosis in children in low-income countries. Int J Tuberc Lung Dis 2005;9:1299-304.
    MedlineWeb of Science
  19. ↵
    1. Burnham K,
    2. Anderson D
    . Model selection multimodel inference: a practical information-theoretic approach. New York: Springer-Verlag; 2002.
  20. ↵
    1. Selwyn PA,
    2. Hartel D,
    3. Lewis VA,
    4. et al
    . A prospective study of the risk of tuberculosis among intravenous drug users with human immunodeficiency virus infection. N Engl J Med 1989;320:545-50.
    MedlineWeb of Science
  21. ↵
    1. Whalen CC,
    2. Nsubuga P,
    3. Okwera A,
    4. et al
    . Impact of pulmonary tuberculosis on survival of HIV-infected adults: a prospective epidemiologic study in Uganda. AIDS 2000;14:1219-28.
    CrossRefMedlineWeb of Science
  22. ↵
    1. Swaminathan S,
    2. Ramachandran R,
    3. Baskaran G,
    4. et al
    . Risk of development of tuberculosis in HIV-infected patients. Int J Tuberc Lung Dis 2000;4:839-44.
    MedlineWeb of Science
  23. ↵
    1. Antonucci G,
    2. Girardi E,
    3. Raviglione MC,
    4. Ippolito G
    . Risk factors for tuberculosis in HIV-infected persons: a prospective cohort study. The Gruppo Italiano di Studio Tubercolosi e AIDS (GISTA). JAMA 1995;274:143-8.
    Abstract/FREE Full Text
  24. ↵
    1. Lawn SD,
    2. Badri M,
    3. Wood R
    . Tuberculosis among HIV-infected patients receiving HAART: long term incidence risk factors in a South African cohort. AIDS 2005;19:2109-16.
    MedlineWeb of Science
  25. ↵
    1. Golub JE,
    2. Saraceni V,
    3. Cavalcante S,
    4. et al
    . The impact of antiretroviral therapy isoniazid preventative therapy on tuberculosis incidence in HIV-infected patients in Rio de Janeiro, Brazil. AIDS. (in pres).
  26. ↵
    1. Day JH,
    2. Grant AD,
    3. Fielding KL,
    4. et al
    . Does tuberculosis increase HIV load? J Infect Dis 2004;190:1677-84.
    Abstract/FREE Full Text
  27. ↵
    1. Wolday D,
    2. Hailu B,
    3. Girma M,
    4. Hailu E,
    5. Sanders E,
    6. Fontan L,
    7. et al
    . Low CD4+ T-cell count high HIV viral load precede the development of tuberculosis disease in a cohort of HIV-positive ethiopians. Int J Tuberc Lung Dis 2003;7:110-6.
    MedlineWeb of Science
  28. ↵
    1. Yancey MK
    . Host defenses bacterial resistance. Obstet Gynecol Clin North Am 1992;19:413-34.
    MedlineWeb of Science
  29. ↵
    1. Krishna U
    1. Saraiya UB,
    2. Bhalerao SA
    . Pregnancy tuberculosis. In: Krishna U, editor. Pregnancy at risk: current concepts. 4th ed. New Delhi, India: The Federation of Obstetric Gynecological Societies of India, Jaypee Brothers Medical Publishers; 2001. p. 132.
  30. ↵
    1. Robinson CA,
    2. Rose NC
    . Tuberculosis: current implications management in obstetrics. Obstet Gynecol Surv 1996;51:115-24.
    CrossRefMedline
  31. ↵
    1. Newell ML,
    2. Coovadia H,
    3. Cortina-Borja M,
    4. Rollins N,
    5. Gaillard P,
    6. Dabis F
    . Mortality of infected uninfected infants born to HIV-infected mothers in Africa: a pooled analysis. Lancet 2004;364:1236-43.
    CrossRefMedlineWeb of Science
  32. ↵
    1. Zar HJ,
    2. Cotton MF,
    3. Strauss S,
    4. et al
    . Effect of isoniazid prophylaxis on mortality incidence of tuberculosis in children with HIV: randomised controlled trial. BMJ 2007;334:136.
    Abstract/FREE Full Text
  33. ↵
    1. Radhakrishna S,
    2. Frieden TR,
    3. Subramani R
    . Association of initial tuberculin sensitivity, age sex with the incidence of tuberculosis in south India: a 15-year follow-up. Int J Tuberc Lung Dis 2003;7:1083-91.
    MedlineWeb of Science
  34. ↵
    1. Pai M,
    2. Gokhale K,
    3. Joshi R,
    4. et al
    . Mycobacterium tuberculosis infection in health care workers in rural India: comparison of a whole-blood interferon gamma assay with tuberculin skin testing. JAMA 2005;293:2746-55.
    Abstract/FREE Full Text
  35. ↵
    1. Franks AL,
    2. Binkin NJ,
    3. Snider DE Jr,
    4. Rokaw WM,
    5. Becker S
    . Isoniazid hepatitis among pregnant postpartum Hispanic patients. Public Health Rep 1989;104:151-5.
    MedlineWeb of Science
  36. ↵
    1. Boggess KA,
    2. Myers ER,
    3. Hamilton CD
    . Antepartum or postpartum isoniazid treatment of latent tuberculosis infection. Obstet Gynecol 2000;96:757-62.
    CrossRefMedlineWeb of Science
  37. ↵
    1. Woldehanna S,
    2. Volmink J
    . Cochrane Database Syst Rev. 2004. Treatment of latent tuberculosis infection in HIV infected persons; p. CD000171.
  38. ↵
    1. Lawn SD,
    2. Myer L,
    3. Bekker LG,
    4. Wood R
    . Burden of tuberculosis in an antiretroviral treatment programme in sub-Saharan Africa: impact on treatment outcomes implications for tuberculosis control. AIDS 2006;20:1605-12.
    MedlineWeb of Science
  39. ↵
    1. World Health Organization Global Tuberculosis Programme
    . Policy statement on preventive therapy against tuberculosis in people living with HIV. Geneva, Switzerland: World Health Organization; 1998.
  40. ↵
    1. Grant AD,
    2. Charalambous S,
    3. Fielding KL,
    4. et al
    . Effect of routine isoniazid preventive therapy on tuberculosis incidence among HIV-infected men in South Africa: a novel randomized incremental recruitment study. JAMA 2005;293:2719-25.
    Abstract/FREE Full Text
| Table of Contents

This Article

  1. Clin Infect Dis. (2007) 45 (2): 241-249. doi: 10.1086/518974
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most