Abstract
Objectives:
Worldwide, adult men experience an excess burden of tuberculosis (TB) disease compared to women, but few studies have examined sex differences in TB among people with HIV. In this study, we aimed to investigate sex differences in TB infection and disease among people with HIV in Rio de Janeiro, Brazil.
Design:
Analysis of data from a randomized controlled trial and retrospective cohort study.
Methods:
We analyzed data from two studies conducted between 2005 and 2017. The THRio Study (2005–2012) evaluated increasing tuberculin skin testing (TST) and TB preventive therapy (TPT) and UnivART (2010–2017) was a virtual cohort study of people with HIV and TB with data from four national electronic registries.
Results:
Among 4,606 people with HIV in THRio, 2,992 (65.0%) had a TST placed and read, of whom 312/1,865 (17%) males and 203/1,127 (18%) females (p=0.37) had prevalent TB infection. TB disease incidence was higher among males compared to females overall (IRR 1.33, 95% CI 1.04–1.69), among males compared to females who did not receive TPT (IRR 1.30, 95% CI 1.01–1.67), and among males compared to females on ART (IRR 1.64, 95% CI 1.17–2.29). Among 54,957 people with HIV in UnivART, TB disease incidence rates were higher among males than females overall (IRR 1.28, 95% CI 1.18–1.39), among males compared to females on ART (IRR 1.58, 95% CI 1.40–1.77), and among males compared to females not on ART (IRR 1.11, 95% CI 0.99–1.25).
Conclusions:
In this medium TB and HIV burden setting, TB disease incidence was higher among males than females with HIV, despite similar prevalence of TB infection.
Keywords: tuberculosis, HIV, sex differences, gender differences, Brazil
BACKGROUND
Globally, rates of tuberculosis (TB) disease are markedly higher among adult men compared to women, with men accounting for 65–75% of all TB cases[1]. In addition, men have more severe TB disease at diagnosis and experience substantially poorer treatment responses and a higher risk of death[2].Few studies have evaluated sex differences in TB among people with HIV (PWH), for whom TB remains the leading cause of death[1]. Prevalence surveys show a nearly 2-fold higher TB disease prevalence and higher prevalence-to-notification ratios among males compared to females in TB/HIV endemic areas,[3–5] suggesting that males are more likely to both develop TB disease and go undiagnosed than females; however, sex-disaggregated data on TB prevalence, incidence, and diagnosis specifically among PWH are limited. Furthermore, whether males and females with HIV have differential susceptibility to TB infection—or only progression to TB disease—remains unclear, with few published data on prevalence of TB infection among males and females with HIV.
Brazil has a medium burden of TB and HIV, with an HIV-associated TB incidence rate of 9 per 100,000 population[6]. Overall, TB disease incidence is more than 2-fold higher among adolescent and adult males compared to females i[6] and differences in TB notifications by sex have not been reported. In this study, we explored sex differences in TB infection and disease among PWH in Rio de Janeiro, Brazil, using data from two large studies conducted over a twelve-year period.
METHODS
Study design and data sources
We used data from two studies conducted in Rio de Janeiro between 2005 and 2017.
THRio:
The TB/HIV in Rio (THRio) study was a cluster-randomized trial conducted in 29 HIV clinics evaluating the impact of an intervention to increase use of tuberculin skin testing (TST) and isoniazid preventive therapy (IPT) on TB incidence and mortality[7–10]. Briefly, clinicians were trained to perform TST for eligible patients according to Brazilian national guidelines and to prescribe IPT for patients with a positive TST if TB disease was ruled-out. Demographic, clinical, treatment and outcome data on were collected from medical records from September 1, 2005-August 31, 2010. Additional TB disease diagnoses and deaths were obtained by linkage with the Rio de Janeiro TB and mortality registries through October 31, 2012.
UnivART:
The Universal ART in Rio (UnivART) study was a virtual cohort study of PWH registered in Rio de Janeiro from 2008–2016, with follow-up through 2017. We used a validated, probabilistic strategy to link data from four surveillance databases for PWH registered in Rio de Janeiro City from 2008–2016[11]. The databases—managed by the Brazilian Ministry of Health and Rio de Janeiro Municipal Health Secretariat—capture information for all individuals diagnosed with HIV: The Sistema de Informação de Agravos de Notificação (SINAN) includes data for all individuals diagnosed with HIV/AIDS or TB disease reported to the Rio de Janeiro Municipal Health Secretariat; the Sistema de Controle de Exames Laboratoriais (SISCEL) captures data on CD4 counts and viral loads; the Sistema de Controle Logístico de Medicamentos (SICLOM) captures data on ART; and the Sistema de Informação sobre Mortalidade (SIM) captures dates and causes of death. HIV (since 2014), AIDS (since 1986), and TB disease are notifiable in Brazil; CD4 and viral load measurements are reported for all patients receiving care in the public sector; and all ART is dispensed through the public sector. Data on TST and TB preventive therapy were not included in the surveillance registries.
Brazilian TB/HIV guidelines
Brazilian guidelines guided care for patients in both studies. During the study periods, national guidelines recommended annual screening for TB infection via TST for all PWH with no history of TB, TB preventive therapy, or positive TST. Patients with a positive TST (≥5mm induration) within 2–4 days were eligible for IPT (6-months daily isoniazid 300mg with pyridoxine 25mg) if TB disease was ruled-out based on clinical history and chest radiography. TB disease was confirmed based on ≥1 positive culture for Mycobacterium tuberculosis, acid-fast bacilli smear, and/or clinical and radiographic presentation consistent with TB and response to anti-TB treatment[12]; additionally, Xpert MTB/RIF (Cepheid) became available in 2011[13]. ART was recommended for PWH with CD4 <200 cells/μL through 2007, <350 cells/μL from 2008–2009, <500 cells/μL from 2010–2013, and for all PWH starting in 2014.
Statistical analysis
THRio:
We included adults (≥18 years) newly registered at THRio clinics from 2005–2009, excluding those already registered at study start and those with prevalent TB disease (diagnosed within 30 days of clinic registration). We used Chi-square and Wilcoxon rank-sum tests to compare patient characteristics and prevalence of TB infection (defined as a positive TST) by sex. We followed patients for up to 7 years, from THRio enrollment to incident TB disease, death, or administrative censoring on October 31, 2012. We calculated TB disease incidence rates per 100 person-years with 95% confidence intervals, and incidence rate ratios (IRR) comparing incidence rates between males and females. In addition, we calculated the cumulative incidence of TB disease using Kaplan-Meier approach, and compared cumulative incidence between males and females using log-rank tests. Time in the study was cross classified by TST status (unknown, negative, positive), IPT (not initiated vs. initiated), and ART (not initiated vs. initiated), and analyses were adjusted for baseline CD4 count.
UnivART:
We included all PWH in Rio de Janeiro City from 2010–2016, with follow-up through 2017. We compared patient characteristics by sex using Chi-square and Wilcoxon rank-sum tests. We followed patients from January 1, 2010 (for those with prevalent HIV, defined as being present in SINAN-HIV, SICLOM, or SISCEL from 2008–2009 or having a documented HIV diagnosis date prior to 2010) or entry into care (for those with incident HIV, defined as being present in SINAN-HIV, SICLOM, or SISCEL on or after January 1, 2010, without being in the databases from 2008–2009 or having a documented HIV diagnosis date prior to 2010) until TB diagnosis, death, or administrative censoring on December 31, 2017. We defined entry into care as the earliest date a patient appeared in SINAN-HIV, SICLOM, or SISCEL. We calculated TB incidence rates per 100 person-years and IRRs comparing TB incidence rates of males to females. Patients diagnosed with HIV within 180 days of TB diagnosis (corresponding to a 6-month TB treatment regimen), were given 1 day of follow-up time. We performed analyses for all prevalent and incident HIV patients from 2010–2016 and people with incident HIV diagnosed from 2010–2016. Time in the study was cross classified by ART status (not initiated vs. initiated), and analyses for people with incident HIV were adjusted for baseline CD4 count.
Ethics approval
Both studies were approved by the institutional review boards of Johns Hopkins University School of Medicine and the Municipal Health Secretariat of Rio de Janeiro, with the requirement for informed consent waived.
RESULTS
Sex differences in TB infection, THRio, 2005–2012
Of 4,606 PWH entering care from 2005–2009 without prevalent TB disease, 2,867 (62.2%) were male, median age was 34 years (IQR 28–42), and 2,138 (46.4%) had a baseline CD4 count ≤350 cells/μL (Table 1). Males and females were of similar age (median 34 vs. 34 years, p=0.74), while males had lower baseline CD4 counts compared to females (median 326 vs. 359 cells/μL, p<0.001).
Table 1.
Characteristics of people with HIV in THRio, 2005–2012
| Total N=4,606 |
Male N=2,867 |
Female N=1,739 |
p-value | |
|---|---|---|---|---|
| Median age, years (IQR) | 34 (28–42) | 34 (28–42) | 34 (27–43) | 0.74 |
| CD4 count | ||||
| ≤350 cells/μL | 2,138 (46.4%) | 1,380 (48.1%) | 758 (43.6%) | 0.02 |
| 351–500 cells/μL | 806 (17.5%) | 481 (16.8%) | 325 (18.7%) | |
| >500 cells/μL | 1,170 (25.4%) | 700 (24.4%) | 470 (27.0%) | |
| Unknown | 492 (10.7%) | 306 (10.7%) | 186 (10.7%) | |
| Median CD4 count, cells/μL (IQR) | 338 (170–535) | 326 (158–522) | 359 (190–561) | <0.001 |
| TST placed and read | 2,992 (65.0%) | 1,865 (65.1%) | 1,127 (64.8%) | 0.87 |
| TST results | ||||
| Negative | 2,322 (77.6%) | 1,450 (77.8%) | 872 (77.4%) | 0.81 |
| Positive 1 | 670 (22.4%) | 415 (22.3%) | 255 (22.7%) | |
| Median days to first TST (IQR) | 45 (10–177) | 42 (9–172) | 50 (11–192) | 0.04 |
| Median days to positive TST (IQR) | 103 (16–478) | 120 (21–540) | 79 (14–394) | 0.04 |
| IPT and ART initiation | ||||
| IPT | 627 (13.6%) | 368 (12.8%) | 259 (14.9%) | 0.05 |
| ART 2 | 2,976 (64.6%) | 1,895 (66.1%) | 1,081 (62.2%) | 0.01 |
| IPT and ART | 390 (8.5%) | 242 (8.4%) | 148 (8.5%) | 0.93 |
| Diagnosed with TB | 300 (6.5%) | 204 (7.1%) | 96 (5.5%) | 0.03 |
| Died 3 | 472 (10.3%) | 300 (10.5%) | 172 (9.9%) | 0.53 |
Abbreviations: IQR, interquartile range; TST, tuberculin skin test; IPT, isoniazid preventive therapy; ART, antiretroviral therapy; TB, tuberculosis
Includes 155 patients who converted from TST-negative to -positive over follow-up (103 males, 52 females)
434 patients were on ART at baseline (283 [9.9%] male, 151 [8.7%] female, p=0.18)
89 patients died after a TB diagnosis (58 [65.2%] male, 31 [34.8%] female, p=0.73)
A total of 2,992 (65.0%) patients had a TST placed and read, including 1,865 (65.1%) males and 1,127 (64.8%) females (p=0.87, Table 1). Median time to first TST was 45 days (IQR 10–177), and was slightly longer for females than males (median 50 vs. 42 days, p=0.04). Of those tested, the prevalence of TB infection at first TST was 17.2% overall, with 312/1,865 (16.7%) TST-positive males and 203/1,127 (18.0%) TST-positive females (p=0.37). Among the 2,477 with a negative first TST, 155 (6.3%) later converted to positive on a follow-up TST (103/1,553 [6.6%] males vs. 52/924 [5.6%] females, p=0.32); there was no difference in time to TST conversion between males and females (median 581 vs. 548 days, p=0.29). The proportion of patients with a positive TST increased with increasing baseline CD4 count but did not differ by sex regardless of CD4 count (Table S1). Among those with a positive TST, 336/415 (81.0%) males and 208/255 (81.6%) females received IPT (p=0.85); and 266/415 (64.1%) males and 134/255 (52.6%) females received ART (p=0.003).
Sex differences in TB disease, THRio, 2005–2012
There were 300 (6.5%) patients diagnosed with TB disease over the follow-up period, for an incidence rate of 1.37 per 100 person-years (95%CI 1.22–1.53, Table 2). Among males, 204 (7.1%) were diagnosed with TB disease, for an incidence rate of 1.51 per 100 person-years (95%CI 1.32–1.74). Among females, 96 (5.5%) were diagnosed with TB disease, for an incidence rate of 1.14 per 100 person-years (95%CI 0.93–1.39). The adjusted TB disease incidence rate ratio (IRR) for males compared to females was 1.29 (95%CI 1.02–1.65). The 7-year cumulative incidence of TB disease was 8.9% for males and 6.2% for females (log-rank p=0.03, Figure 1a).
Table 2.
Incidence of TB disease among people with HIV in THRio, 2005–2012
| Incidence rate per 100 person-years (95%CI) |
IRR (95%CI) |
Adjusted IRR1 (95%CI) |
|||
|---|---|---|---|---|---|
| Total N=4,606 |
Male N=2,867 |
Female N=1,739 |
|||
| Total | 1.37 (1.22–1.53) | 1.51 (1.32–1.74) | 1.14 (0.93–1.39) | 1.33 (1.04–1.69) | 1.29 (1.02–1.65) |
| TST status | |||||
| TST-negative | 0.87 (0.71–1.07) | 1.00 (0.79–1.27) | 0.66 (0.45–0.96) | 1.53 (0.97–2.41) | 1.50 (0.96–2.34) |
| TST-positive | 2.46 (1.95–3.12) | 2.50 (1.86–3.38) | 2.40 (1.63–3.52) | 1.04 (0.64–1.70) | 1.00 (0.62–1.63) |
| TST-unknown | 1.64 (1.39–1.94) | 1.85 (1.51–2.27) | 1.32 (0.98–1.78) | 1.40 (0.98–2.00) | 1.34 (0.93–1.92) |
| Baseline CD4 count | |||||
| ≤350 cells/μL | 1.69 (1.46–1.97) | 1.82 (1.52–2.18) | 1.47 (1.12–1.92) | 1.24 (0.90–1.72) | — |
| 351–500 cells/μL | 1.14 (0.85–1.52) | 1.33 (0.94–1.89) | 0.86 (0.51–1.45) | 1.55 (0.82–2.90) | — |
| >500 cells/μL | 0.81 (0.61–1.08) | 0.93 (0.66–1.32) | 0.63 (0.38–1.04) | 1.48 (0.80–2.74) | — |
| Unknown | 1.81 (1.31–2.48) | 1.85 (1.24–2.76) | 1.73 (1.02–2.92) | 1.07 (0.55–2.07) | — |
| IPT status | |||||
| No IPT | 1.44 (1.28–1.62) | 1.58 (1.37–1.82) | 1.22 (0.99–1.50) | 1.30 (1.01–1.67) | 1.27 (0.99–1.64) |
| Initiated IPT | 0.82 (0.54–1.25) | 0.98 (0.59–1.62) | 0.62 (0.29–1.30) | 1.58 (0.64–3.87) | 1.57 (0.68–3.62) |
| ART status | |||||
| No ART | 1.27 (1.06–1.52) | 1.28 (1.02–1.61) | 1.26 (0.95–1.66) | 1.01 (0.71–1.45) | 1.00 (0.70–1.42) |
| Initiated ART | 1.45 (1.25–1.68) | 1.69 (1.42–2.00) | 1.03 (0.77–1.38) | 1.64 (1.17–2.29) | 1.61 (1.15–2.25) |
Abbreviations: CI, confidence interval; IRR, incidence rate ratio; TST, tuberculin skin test; IPT, isoniazid preventive therapy; ART, antiretroviral therapy
Legend: Individuals may have contributed person-time to some or all rows, according to changing TST, IPT, and ART status.
Adjusted for baseline CD4 count
Figure 1.
Cumulative incidence of TB disease among men and women with HIV in a) THRio*; b) all people with HIV in UnivART†; c) all people with incident HIV in UnivART‡
* Log-rank p=0.03
† Log-rank p<0.001
‡ Log-rank p<0.001
Comparing males to females, the adjusted TB disease IRR was 1.50 (95%CI 0.96–2.34) for TST-negatives, 1.00 (95%CI 0.62–1.63) for TST-positives, and 1.34 (95%CI 0.93–1.92) for those that never had a TST placed and read (“TST-unknown,” Table 2, Figure S1).
Comparing males to females who initiated IPT, the adjusted TB disease IRR was 1.53 (95%CI 0.61–3.85) for TST-positives and 0.75 (95%CI 0.05–11.99) for TST-negatives (Table 3). No patients with an unknown TST who initiated IPT were diagnosed with TB disease. Comparing males to females who did not initiate IPT, the adjusted TB disease IRR was 0.78 (95%CI 0.44–1.38) for TST-positives, 1.51 (95%CI 0.96–2.37) for TST-negatives, and 1.32 (95%CI 0.92–1.90) for those with an unknown TST.
Table 3.
Incidence of TB disease among people with HIV in THRio, stratified by IPT, ART, and TST status
| Incidence rate per 100 person-years (95%CI) |
IRR (95%CI) |
Adjusted IRR1 (95%CI) |
|||
|---|---|---|---|---|---|
| Total N=4,606 |
Male N=2,867 |
Female N=1,739 |
|||
| No IPT | |||||
| TST-negative | 0.88 (0.71–1.07) | 1.00 (0.79–1.28) | 0.66 (0.45–0.97) | 1.52 (0.96–2.39) | 1.51 (0.96–2.37) |
| TST-positive | 10.34 (7.82–13.68) | 9.37 (6.51–13.48) | 12.17 (7.85–18.87) | 0.77 (0.44–1.36) | 0.78 (0.44–1.38) |
| TST-unknown | 1.66 (1.40–1.96) | 1.86 (1.52–2.28) | 1.37 (1.02–1.84) | 1.39 (0.97–1.99) | 1.32 (0.92–1.90) |
| Initiated IPT | |||||
| TST-negative | 0.75 (0.19–2.99) | 0.94 (0.13–6.64) | 0.62 (0.09–4.41) | 1.50 (0.09–24.05) | 0.75 (0.05–11.99) |
| TST-positive | 0.86 (0.55–1.33) | 0.99 (0.59–1.68) | 0.65 (0.29–1.45) | 1.52 (0.59–3.97) | 1.53 (0.61–3.85) |
| TST-unknown | 0 | 0 | 0 | — | — |
| No ART | |||||
| TST-negative | 0.83 (0.56–1.22) | 0.90 (0.57–1.43) | 0.70 (0.35–1.39) | 1.30 (0.57–2.99) | 1.30 (0.57–3.00) |
| TST-positive | 3.18 (2.29–4.41) | 2.89 (1.82–4.59) | 3.53 (2.22–5.60) | 0.82 (0.43–1.58) | 0.63 (0.33–1.19) |
| TST-unknown | 1.13 (0.88–1.45) | 1.19 (0.86–1.65) | 1.04 (0.70–1.55) | 1.15 (0.69–1.92) | 1.11 (0.67–1.84) |
| Initiated ART | |||||
| TST-negative | 0.99 (0.78–1.26) | 1.15 (0.87–1.52) | 0.73 (0.47–1.15) | 1.57 (0.92–2.67) | 1.51 (0.89–2.56) |
| TST-positive | 2.16 (1.53–3.03) | 2.46 (1.66–3.64) | 1.55 (0.78–3.11) | 1.58 (0.71–3.51) | 1.61 (0.74–3.52) |
| TST-unknown | 1.98 (1.58–2.48) | 2.29 (1.77–2.97) | 1.40 (0.90–2.20) | 1.63 (0.97–2.75) | 1.65 (0.98–2.78) |
Abbreviations: CI, confidence interval; IRR, incidence rate ratio; TST, tuberculin skin test; IPT, isoniazid preventive therapy; ART, antiretroviral therapy
Legend: Individuals may have contributed person-time to some or all rows, according to changing TST, IPT, and ART status.
Adjusted for baseline CD4 count
Comparing males to females who initiated ART, the adjusted TB disease IRR was 1.51 (95%CI 0.89–2.56) for TST-negatives, 1.61 (0.74–3.52) for TST-positives, and 1.65 (95%CI 0.98–2.78) for those with an unknown TST (Table 3). For those not on antiretroviral therapy (ART), the adjusted TB disease IRR was 1.30 (95% 0.57–3.00) for TST-negatives, 0.63 (95%CI 0.33–1.19) for TST-positives, and 1.11 (0.67–1.84) for those with an unknown TST.
Sex differences in TB disease, UnivART, 2010–2017
Among 54,957 PWH registered in Rio de Janeiro from 2010–2016, 35,880 (65.3%) were male and 19,077 (34.7%) were female. Median age at entry into HIV care was 35 years (IQR 28–43); 14,715 (26.8%) were identified as brown, black, or Indigenous; and 4,810 (8.8%) reported having less than a high school education. Males and females were of similar age (median 35 vs. 35 years, p=0.85), while a higher proportion of females were brown, black, or Indigenous (30.2% vs. 25.0%, p<0.001) and had less than a high school education (11.4% vs. 7.4%).
There were 2,665 (4.8%) patients diagnosed with TB disease over the follow-up period, for an incidence rate of 0.91 per 100 person-years (95%CI 0.87–0.94). Among males, 1,835 (5.1%) were diagnosed with TB disease, for an incidence rate of 0.99 per 100 person-years (95%CI 0.94–1.03). Among females, 830 (4.4%) were diagnosed with TB disease, for an incidence rate of 0.77 per 100 person-years (95%CI 0.72–0.82). The TB disease IRR for males compared to females was 1.28 (95%CI 1.18–1.39). The 8-year cumulative incidence of TB disease was 6.1% for males and 5.2% for females (log-rank p<0.001, Figure 1b). Comparing males to females, the TB disease IRR was 1.58 (95%CI 1.40–1.77) for those not on ART and 1.11 (95%CI 0.99–1.25) for those who initiated ART.
Among 30,485 people with incident HIV during follow-up, 20,958 (68.8%) were male and 9,527 (31.3%) were female. There were 1,667 (5.5%) patients with incident HIV diagnosed with TB disease over follow-up, for an incidence rate of 1.45 per 100 person-years (95%CI 1.38–1.52, Table 4). Among males with incident HIV, 1,235 (5.9%) were diagnosed with TB disease, for an incidence rate of 1.60 per 100 person-years (95%CI 1.52–1.70). Among females with incident HIV, 432 (4.5%) were diagnosed with TB disease, for an incidence rate of 1.14 per 100 person-years (95%CI 1.04–1.25). The TB disease IRR for males compared to females was 1.41 (95%CI 1.26–1.57). The 8-year cumulative incidence of TB disease was 7.3% for males and 5.6% for females (log-rank p<0.001, Figure 1c). Comparing males to females, the TB disease IRR was 1.52 (95%CI 1.33–1.74) for those not on ART and 1.38 (95%CI 1.14–1.67) for those who initiated ART.
Table 4.
Incidence of TB disease among people with incident HIV in UnivART
| Incidence rate per 100 person-years (95%CI) |
IRR (95%CI) |
Adjusted IRR (95%CI)1 |
|||
|---|---|---|---|---|---|
| Total N=30,485 |
Male N=20,958 |
Female N=9,527 |
|||
| Total | 1.45 (1.38–1.52) | 1.60 (1.52–1.70) | 1.14 (1.04–1.25) | 1.41 (1.26–1.57) | 1.43 (1.28–1.60) |
| ART status | |||||
| No ART | 2.81 (2.65–2.98) | 3.19 (2.98–3.42) | 2.10 (1.87–2.36) | 1.52 (1.33–1.74) | 1.45 (1.27–1.66) |
| Initiated ART | 0.75 (0.69–0.82) | 0.82 (0.75–0.91) | 0.60 (0.51–0.70) | 1.38 (1.14–1.67) | 1.43 (1.19–1.73) |
| Baseline CD4 count | |||||
| ≤350 cells/μL | 3.20 (2.95–3.48) | 3.73 (3.39–4.10) | 2.28 (1.94–2.68) | 1.64 (1.36–1.97) | — |
| 351–500 cells/μL | 1.05 (0.86–1.30) | 1.19 (0.93–1.53) | 0.83 (0.57–1.21) | 1.44 (0.92–2.26) | — |
| >500 cells/μL | 0.74 (0.62–0.88) | 0.82 (0.66–1.02) | 0.62 (0.46–0.85) | 1.31 (0.90–1.91) | — |
| Unknown | 1.23 (1.15–1.31) | 1.32 (1.23–1.43) | 1.00 (0.88–1.14) | 1.32 (1.13–1.54) | — |
Abbreviations: CI, confidence interval; IRR, incidence rate ratio; ART, antiretroviral therapy
Legend: Individuals may have contributed person-time to some or all rows, according to changing ART status.
Adjusted for baseline CD4 count
DISCUSSION
In this analysis from two large cohorts with over ten years of follow-up, we found that rates of TB disease were higher in males compared to females with HIV, despite similar prevalence of TB infection as assessed by TST. M. tuberculosis acquisition is known to be influenced by multiple socio-behavioral factors; progression to TB disease is more influenced by intrinsic factors, such as immunodeficiency, co-morbidities, and body habitus, though behaviors such as ART adherence and tobacco and alcohol use are also important. The elevated risk of TB disease observed among men in our study (despite a similar prevalence of TB infection as women) suggests that biological factors may play an important role in driving sex differences in TB disease for PWH, and should prompt additional research to elucidate both the mechanisms underlying this sex difference, as well as strategies to reduce sex- and gender-based disparities in TB incidence and outcomes.
While some studies show similar prevalence of TB infection by sex[14–16], others document a higher prevalence among either males[17–20] or females[21]. Here, we confirmed that prevalence of TB infection was similar among males and females with HIV in THRio: Overall, approximately 14% of males and females were TST-positive; among those who completed TST, approximately 17% were TST-positive on their first test, and 22% were TST-positive by the end of follow-up, with no difference in testing uptake or positivity between males and females. Furthermore, prevalence of TB infection was similar for males and females regardless of CD4 count, and TST positivity was higher with increasing CD4 count regardless of sex. Despite a similar prevalence of TB infection, males in THRio had a 29% higher TB incidence rate compared to females. While we did not have data on TB infection in the UnivART cohort, we observed similar trends in TB disease incidence: Among all PWH during the UnivART study period, the TB incidence rate was 24% higher among males compared to females; when restricting to those with incident HIV over the study period, the TB incidence rate was 40% higher for males regardless of ART status.
Interestingly, we found higher rates of TB disease among TST-positive females who did not initiate IPT (male:female IRR 0.78, 95%CI 0.44–1.38) and ART (male:female IRR 0.63, 95%CI 0.33–1.19) in THRio, though the confidence intervals around these estimates are wide. Importantly, TST-positive females were less likely to receive ART compared to males (53% vs. 64%), which may have contributed to this finding. Another potential explanation is that women with positive TSTs were more recently infected than men and therefore at higher risk of progression to disease; however, we were unable to directly assess timing of TB infection in this study. Qualitative analyses in Brazil[22] show that health care providers often order testing for TB infection for patients with TB symptoms or contact with a TB case, suggesting that testing for TB infection is often prompted by suspicion of TB disease, and it is possible that females in THRio were more likely than males to undergo TST when TB disease was suspected. Supporting this hypothesis, we found that median time from a positive TST to TB diagnosis was shorter for females compared to males (median 84 vs. 247 days), though the difference was not statistically significant (p=0.21).
Historically, differences in TB between men and women have been attributed to gendered behavioral norms and sociocultural roles, which may contribute to higher risk of TB infection, progression to TB disease, and poor outcomes for men[23–36]. In many settings, men have a higher risk of exposure to M. tuberculosis, with a greater number of contacts and more occupational exposures[28]. In contrast, our data suggest that risk of TB infection was similar for men and women with HIV entering care in Rio, with 22% of men and women who completed TST testing positive. Although behavioral data were not available for our study populations, behavioral risk factors—such as smoking and alcohol use—are typically more common among men and are often cited as contributing to disparities in TB incidence[29–32,34,37]. Poorer ART and TB preventive therapy adherence may similarly help drive elevated TB risk for men, and suboptimal TB treatment adherence may lead to worse clinical outcomes for men[33–36]. Importantly, differential access to healthcare[23–27,38] and poorer quality of TB evaluation[39–41] may contribute to under-diagnosis of TB among women in some settings. However, data from prevalence surveys (which use systematic screening to reduce risk of bias in case detection) confirm the higher burden of TB among males in TB/HIV endemic settings, and additionally suggest that males are often more likely to go undiagnosed than females, with substantially higher prevalence-to-notification ratios for males[1,3–5]. In Brazil, TB and HIV testing and treatment are freely available via the public sector, which may help reduce barriers to care and TB diagnosis; however, challenges in accessing care likely remain for some individuals, with possible differential effects for men and women. In THRio, we found that women had slightly higher baseline CD4 counts, suggesting earlier HIV diagnosis compared to men and therefore lower underlying TB risk. However, our analyses of the UnivART cohort found similar CD4 counts for men and women at entry into HIV care, and a higher prevalence of sociodemographic TB risk factors among women. Differences in CD4 counts between males and females entering care between the THRio and UnivART cohorts may have been due to several factors, including a strong emphasis on HIV screening among pregnant women during THRio (leading to earlier diagnosis and higher CD4 counts for women), and scale-up of HIV “test and treat” initiatives in later years (leading to similar CD4 counts at entry into HIV care for males and females in UnivART).
Emerging data suggest that biological differences between males and females likely help drive elevated TB risk for males, in addition socio-behavioral differences between men and women[42]. Animal models have demonstrated greater TB susceptibility[43,44] and disease severity[45,46] in male mice compared to female mice; similarly, female mice have been shown to have lower rates of disease progression for some non-tuberculous mycobacterial infections[47–49]. Sex hormones—which influence mycobacterial growth and immune responses to infection[47,48,50–52]—are a likely source of male-female differences in TB risk. In females of reproductive age, those with TB have been shown to have lower concentrations of estradiol and progesterone compared to those without TB[53]. Conversely, estrogen concentrations have been shown to be higher in post-menopausal females with TB compared to post-menopausal females without TB[54]. However, understanding of the role of sex hormones on immune responses to TB is poor, and additional human and animal studies are needed. In addition, limited data have suggested potential sex differences in TST performance may exist, with larger indurations among females with confirmed TB than males, and lower reactivity among females who were exposed to, but did not develop, TB compared to males[55]. Additional studies evaluating sex differences in performance of tests for TB infection are needed.
This analysis has several limitations. First, we were unable to assess the impact of multiple important socio-behavioral and biologic factors that likely affect TB risk, as our data were collected from medical records and surveillance databases. However, the limited demographic data that were available suggested that women had greater socioeconomic risks compared to men. Second, despite mandatory reporting of all TB and HIV diagnoses through Brazilian surveillance systems, it is possible that some cases were either not detected or not reported. While underreporting of individuals diagnosed with TB is likely to have been non-differential by sex in our studies (leading to underestimation of sex differences in TB risk), gaps in case detection have been shown to be more substantial for men in many settings[3–5]. In contrast, some prior studies have found that quality of TB evaluation and testing is worse for women in several settings, leading to a greater risk of under-diagnosis compared to men[39–41]. Third, we did not have data on TB infection for those in the UnivART cohort, as TST/IGRA results were not collected via national surveillance systems during the study period. Finally, we may have had insufficient power to detect sex differences in the THRio study cohort, particularly among TST-positives. THRio promoted screening for TB infection and isoniazid preventive therapy for those who were TST-positive, and the majority of TST-positives received IPT and/or ART,[9,56] substantially lowering their risk of TB regardless of sex; future studies with a larger population of individuals with confirmed TB infection would provide more precise estimates of sex differences in this important population at high TB risk. Despite these limitations, this study provides valuable data demonstrating similar burdens of TB infection among males and females with HIV in Rio de Janeiro, but elevated risk of TB disease among males. These data should inform future studies designed to assess socio-behavioral and biologic factors, and their influences on sex differences in TB risk among PWH.
In conclusion, we found that prevalence of TB infection was similar between males and females with HIV in a TB/HIV endemic setting, but that risk of TB disease was higher for males. These results suggest that biologic factors may play an important role in observed sex differences in TB prevalence and incidence worldwide, in addition to gendered socio-behavioral factors that are known to influence TB risk. While strategies such as enhanced TB screening and scale-up of TB preventive therapy for males with HIV may help reduce sex differences in TB, rigorous studies elucidating the precise mechanisms driving the enhanced TB risk for males are needed.
Supplementary Material
Figure S1. Cumulative incidence of TB disease among men and women with HIV in THRio, stratified by TST status: a) TST-positive*, b) TST-negative†, c) TST-unknown‡
* Log-rank p=0.09
† Log-rank p=0.72
‡ Log-rank p0.07
ACKNOWLEDGEMENTS
We thank the patients, clinicians, and staff who participated in the THRio and UnivART studies.
Conflicts of Interest and Source of Funding:
These studies were supported by the Bill & Melinda Gates Foundation (19790.01 to the Consortium to Respond Effectively to the AIDS-Tuberculosis Epidemic); the Johns Hopkins Center for AIDS Research (P30 AI094189 to REC); the Johns Hopkins Tuberculosis Research Advancement Center (P30 AI168436 to REC) and the Johns Hopkins HIV Epidemiology and Prevention Sciences Training Program (T32 AI102623 to LHC). The funding organizations had no role in the design, collection, analysis, or interpretation of data. Earlier versions of this analysis were presented at the Conference on Retroviruses and Opportunistic Infections (CROI) in 2022 and the Union World Conference on Lung Health in 2023. The authors declare no conflicts of interest.
References
- 1.World Health Organization. Global Tuberculosis Report 2023. Geneva: ; 2023. [Google Scholar]
- 2.Chidambaram V, Tun NL, Majella MG, Ruelas Castillo J, Ayeh SK, Kumar A, et al. Male Sex Is Associated With Worse Microbiological and Clinical Outcomes Following Tuberculosis Treatment: A Retrospective Cohort Study, a Systematic Review of the Literature, and Meta-analysis. Clin Infect Dis 2021; 73:1580–1588. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Borgdorff MW, Nagelkerke NJ, Dye C, Nunn P. Gender and tuberculosis: a comparison of prevalence surveys with notification data to explore sex differences in case detection. Int J Tuberc Lung Dis 2000; 4:123–132. [PubMed] [Google Scholar]
- 4.Horton KC, MacPherson P, Houben RMGJ, White RG, Corbett EL. Sex Differences in Tuberculosis Burden and Notifications in Low- and Middle-Income Countries: A Systematic Review and Meta-analysis. PLoS Med 2016; 13:e1002119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.World Health Organization. National tuberculosis prevalence surveys, 2007–2016. Geneva: ; 2021. [Google Scholar]
- 6.World Health Organization. Tuberculosis profile: Brazil. ; 2023. https://worldhealthorg.shinyapps.io/tb_profiles/?_inputs_&entity_type=%22country%22&iso2=%22BR%22&lan=%22EN%22 (accessed 6 Feb2024).
- 7.Moulton LH, Golub JE, Durovni B, Cavalcante SC, Pacheco AG, Saraceni V, et al. Statistical design of THRio: a phased implementation clinic-randomized study of a tuberculosis preventive therapy intervention. Clin Trials 2007; 4:190–199. [DOI] [PubMed] [Google Scholar]
- 8.Durovni B, Cavalcante SC, Saraceni V, Vellozo V, Israel G, King BS, 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:S49–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Durovni B, Saraceni V, Moulton LH, Pacheco AG, Cavalcante SC, King BS, et al. Effect of improved tuberculosis screening and isoniazid preventive therapy on incidence of tuberculosis and death in patients with HIV in clinics in Rio de Janeiro, Brazil: a stepped wedge, cluster-randomised trial. Lancet Infect Dis 2013; 13:852–858. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Golub JE, Cohn S, Saraceni V, Cavalcante SC, Pacheco AG, Moulton LH, et al. Long-term protection from isoniazid preventive therapy for tuberculosis in HIV-infected patients in a medium-burden tuberculosis setting: the TB/HIV in Rio (THRio) study. Clin Infect Dis 2015; 60:639–645. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Camargo K, Coeli C. OpenRecLink. 2016.http://reclink.sourceforge.net/
- 12.Conde MB, Melo FAF de, Marques AMC, Cardoso NC, Pinheiro VGF, Dalcin P de TR, et al. III Brazilian Thoracic Association Guidelines on tuberculosis. J Bras Pneumol 2009; 35:1018–1048. [DOI] [PubMed] [Google Scholar]
- 13.Durovni B, Saraceni V, van den Hof S, Trajman A, Cordeiro-Santos M, Cavalcante S, et al. Impact of replacing smear microscopy with Xpert MTB/RIF for diagnosing tuberculosis in Brazil: a stepped-wedge cluster-randomized trial. PLoS Med 2014; 11:e1001766. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ting W-Y, Huang S-F, Lee M-C, Lin Y-Y, Lee Y-C, Feng J-Y, et al. Gender disparities in latent tuberculosis infection in high-risk individuals: a cross-sectional study. PLoS One 2014; 9:e110104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ncayiyana JR, Bassett J, West N, Westreich D, Musenge E, Emch M, et al. Prevalence of latent tuberculosis infection and predictive factors in an urban informal settlement in Johannesburg, South Africa: a cross-sectional study. BMC Infect Dis 2016; 16:661. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Teklu T, Legesse M, Medhin G, Zewude A, Chanyalew M, Zewdie M, et al. Latent tuberculosis infection and associated risk indicators in pastoral communities in southern Ethiopia: a community based cross-sectional study. BMC Public Health 2018; 18:266. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.de Souza FM, do Prado TN, Pinheiro J dos S, Peres RL, Lacerda TC, Loureiro RB, et al. Comparison of interferon-γ release assay to two cut-off points of tuberculin skin test to detect latent Mycobacterium tuberculosis infection in primary health care workers. PLoS One 2014; 9:e102773. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.He G, Li Y, Zhao F, Wang L, Cheng S, Guo H, et al. The Prevalence and Incidence of Latent Tuberculosis Infection and Its Associated Factors among Village Doctors in China. PLoS One 2015; 10:e0124097. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Sabri A, Quistrebert J, Naji Amrani H, Abid A, Zegmout A, Abderrhamani Ghorfi I, et al. Prevalence and risk factors for latent tuberculosis infection among healthcare workers in Morocco. PLoS One 2019; 14:e0221081. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Reichler MR, Khan A, Yuan Y, Chen B, McAuley J, Mangura B, et al. Duration of Exposure Among Close Contacts of Patients With Infectious Tuberculosis and Risk of Latent Tuberculosis Infection. Clin Infect Dis 2020; 71:1627–1634. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Wada PY, Costa AG, Araújo-Pereira M, Barreto-Duarte B, Souza AB, Rocha MS, et al. Possible sex difference in latent tuberculosis infection risk among close tuberculosis contacts. Int J Infect Dis 2022; 122:685–692. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Salles I, Travassos P, Spener-Gomes R, Loch AP, Saraceni V, Lauria L, et al. Contextualizing and optimizing novel strategies to improve the latent TB continuum of care: Insights from people living with HIV and health care providers in Brazil. PLOS Glob Public Health 2023; 3:e0001251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Holmes CB, Hausler H, Nunn P. A review of sex differences in the epidemiology of tuberculosis. Int J Tuberc Lung Dis 1998; 2:96–104. [PubMed] [Google Scholar]
- 24.Long NH, Johansson E, Lönnroth K, Eriksson B, Winkvist A, Diwan VK. Longer delays in tuberculosis diagnosis among women in Vietnam. Int J Tuberc Lung Dis 1999; 3:388–393. [PubMed] [Google Scholar]
- 25.Thorson A, Hoa NP, Long NH, Allebeck P, Diwan VK. 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 Clin Epidemiol 2004; 57:398–402. [DOI] [PubMed] [Google Scholar]
- 26.Storla DG, Yimer S, Bjune GA. A systematic review of delay in the diagnosis and treatment of tuberculosis. BMC Public Health 2008; 8:15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Cai J, Wang X, Ma A, Wang Q, Han X, Li Y. Factors associated with patient and provider delays for tuberculosis diagnosis and treatment in Asia: a systematic review and meta-analysis. PLoS One 2015; 10:e0120088. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Hertz D, Schneider B. Sex differences in tuberculosis. Semin Immunopathol 2019; 41:225–237. [DOI] [PubMed] [Google Scholar]
- 29.Maurya V, Vijayan VK, Shah A. Smoking and tuberculosis: an association overlooked. Int J Tuberc Lung Dis 2002; 6:942–951. [PubMed] [Google Scholar]
- 30.Bates M, O’Grady J, Mwaba P, Chilukutu L, Mzyece J, Cheelo B, et al. Evaluation of the burden of unsuspected pulmonary tuberculosis and co-morbidity with non-communicable diseases in sputum producing adult inpatients. PloS one 2012; 7:e40774. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Wilsnack RW, Wilsnack SC, Kristjanson AF, Vogeltanz-Holm ND, Gmel G. Gender and alcohol consumption: patterns from the multinational GENACIS project. Addiction 2009; 104:1487–1500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Amere GA, Nayak P, Salindri AD, Narayan KMV, Magee MJ. Contribution of Smoking to Tuberculosis Incidence and Mortality in High-Tuberculosis-Burden Countries. Am J Epidemiol 2018; 187:1846–1855. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Dooley KE, Lahlou O, Ghali I, Knudsen J, Elmessaoudi MD, Cherkaoui I, et al. Risk factors for tuberculosis treatment failure, default, or relapse and outcomes of retreatment in Morocco. BMC Public Health 2011; 11:140. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Danarastri S, Perry KE, Hastomo YE, Priyonugroho K. Gender differences in health-seeking behaviour, diagnosis and treatment for TB. Int J Tuberc Lung Dis 2022; 26:568–570. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Gilmour B, Xu Z, Bai L, Alene KA, Clements ACA. Risk factors associated with unsuccessful tuberculosis treatment outcomes in Hunan Province, China. Trop Med Int Health 2022; 27:290–299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Barathi A, Krishnamoorthy Y, Sinha P, Horsburgh C, Hochberg N, Johnson E, et al. Effect of treatment adherence on the association between sex and unfavourable treatment outcomes among tuberculosis patients in Puducherry, India: a mediation analysis. J Public Health (Oxf) 2023; 45:304–311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Watkins RE, Plant AJ. Does smoking explain sex differences in the global tuberculosis epidemic? Epidemiol Infect 2006; 134:333–339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Ward J, Siskind V, Konstantinos A. Patient and health care system delays in Queensland tuberculosis patients, 1985–1998. Int J Tuberc Lung Dis 2001; 5:1021–1027. [PubMed] [Google Scholar]
- 39.Begum V, de Colombani P, Das Gupta S, Salim AH, Hussain H, Pietroni M, et al. Tuberculosis and patient gender in Bangladesh: sex differences in diagnosis and treatment outcome. Int J Tuberc Lung Dis 2001; 5:604–610. [PubMed] [Google Scholar]
- 40.Thongraung W, Chongsuvivatwong V, Pungrassamee P. Multilevel factors affecting tuberculosis diagnosis and initial treatment. J Eval Clin Pract 2008; 14:378–384. [DOI] [PubMed] [Google Scholar]
- 41.Miller CR, Davis JL, Katamba A, Sserwanga A, Kakeeto S, Kizito F, et al. Sex disparities in tuberculosis suspect evaluation: a cross-sectional analysis in rural Uganda. Int J Tuberc Lung Dis 2013; 17:480–485. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Gupta M, Srikrishna G, Klein SL, Bishai WR. Genetic and hormonal mechanisms underlying sex-specific immune responses in tuberculosis. Trends Immunol 2022; 43:640–656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Ulrichs T, Kosmiadi GA, Trusov V, Jörg S, Pradl L, Titukhina M, et al. Human tuberculous granulomas induce peripheral lymphoid follicle-like structures to orchestrate local host defence in the lung. J Pathol 2004; 204:217–228. [DOI] [PubMed] [Google Scholar]
- 44.Hertz D, Dibbern J, Eggers L, von Borstel L, Schneider BE. Increased male susceptibility to Mycobacterium tuberculosis infection is associated with smaller B cell follicles in the lungs. Sci Rep 2020; 10:5142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Bini EI, Mata Espinosa D, Marquina Castillo B, Barrios Payán J, Colucci D, Cruz AF, et al. The influence of sex steroid hormones in the immunopathology of experimental pulmonary tuberculosis. PLoS One 2014; 9:e93831. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Dibbern J, Eggers L, Schneider BE. Sex differences in the C57BL/6 model of Mycobacterium tuberculosis infection. Sci Rep 2017; 7:10957. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Yamamoto Y, Tomioka H, Sato K, Saito H, Yamada Y, Setogawa T. Sex differences in the susceptibility of mice to infection induced by Mycobacterium intracellulare. Am Rev Respir Dis 1990; 142:430–433. [DOI] [PubMed] [Google Scholar]
- 48.Yamamoto Y, Saito H, Setogawa T, Tomioka H. Sex differences in host resistance to Mycobacterium marinum infection in mice. Infect Immun 1991; 59:4089–4096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Tsuyuguchi K, Suzuki K, Matsumoto H, Tanaka E, Amitani R, Kuze F. Effect of oestrogen on Mycobacterium avium complex pulmonary infection in mice. Clin Exp Immunol 2001; 123:428–434. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Liva SM, Voskuhl RR. Testosterone acts directly on CD4+ T lymphocytes to increase IL-10 production. J Immunol 2001; 167:2060–2067. [DOI] [PubMed] [Google Scholar]
- 51.Zhao Y, Ying H, Demei J, Xie J. Tuberculosis and sexual inequality: the role of sex hormones in immunity. Crit Rev Eukaryot Gene Expr 2012; 22:233–241. [DOI] [PubMed] [Google Scholar]
- 52.Klein SL, Flanagan KL. Sex differences in immune responses. Nat Rev Immunol 2016; 16:626–638. [DOI] [PubMed] [Google Scholar]
- 53.A Aswathy T, Purohit Gopal, Vyas Sunil. Evaluation of hormones in the menstrual cycle of women with pulmonary tuberculosis. Eur Respir J 2018; 52:PA2703. [Google Scholar]
- 54.Erbay G, Senol G, Anar C, Meral AR, Tuzel O. RELATIONSHIP BETWEEN TUBERCULOSIS AND FEMALE HORMONE LEVELS IN POST-MENOPAUSAL WOMEN. Southeast Asian J Trop Med Public Health 2016; 47:78–83. [PubMed] [Google Scholar]
- 55.Bothamley GH. Male Sex Bias in Immune Biomarkers for Tuberculosis. Front Immunol 2021; 12:640903. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Chaisson LH, Saraceni V, Cohn S, Seabrook D, Cavalcante SC, Chaisson RE, et al. CD4+ cell count stratification to guide tuberculosis preventive therapy for people living with HIV. AIDS 2020; 34:139–147. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1. Cumulative incidence of TB disease among men and women with HIV in THRio, stratified by TST status: a) TST-positive*, b) TST-negative†, c) TST-unknown‡
* Log-rank p=0.09
† Log-rank p=0.72
‡ Log-rank p0.07