The advent of highly active antiretroviral therapy (HAART) in 1996 led to a substantial decrease in the incidence of opportunistic infections among people living with human immunodeficiency virus (HIV) in many regions.1 2 3 Nonetheless, HIV-associated tuberculosis (TB) remains an important global health challenge. The World Health Organization estimated that 10.6 million people were living with TB worldwide in 2021, 6.7% of whom were living with HIV.4 In the same year, 1.6 million people died of TB, including 187 000 people who were living with HIV.4 The burden of HIV-associated TB considerably varies across countries and regions.5 6 The prevalence of HIV-associated TB in individual areas reportedly changes over time.7 8 9 Awareness of changes in the epidemiology and clinical manifestations of patients with HIV-associated TB can help policymakers formulate timely relevant prevention and control measures. It can also help improve treatment outcomes for patients with HIV-associated TB.

In Hong Kong, the TB case notification rate has exhibited an overall decreasing trend over the past few decades.10 In 2021, the provisional number of TB cases reported to the Department of Health was 3741.11 The corresponding TB notification rate was 50.5 per 100 000 inhabitants, a substantial decrease from 84.1 per 100 000 inhabitants in 2006.11 The overall prevalence of HIV infection in the general adult population has remained low (<0.1%).12 The epidemiology and clinical manifestations of HIV-associated TB during the period 1996 to 2006 in Hong Kong have been reported.13 Notably, the report showed that TB had become an increasingly important acquired immunodeficiency syndrome (AIDS)–defining illness in Hong Kong, surpassing Pneumocystis jirovecii pneumonia as the most common primary AIDS-defining illness in 2005; the two illnesses represented 39% and 31% of all such illnesses, respectively, in 2005.13 The presentation of HIV-associated TB is often atypical.13 Considering the declining prevalence of TB in the general population and accompanying decrease in TB transmission in Hong Kong, the implementation of strategies to enhance screening for latent TB infection, the increased use of molecular tests for TB diagnosis, and the expansion of HAART in recent years, we conducted a retrospective review of data regarding patients with HIV-associated TB that were reported to the TB-HIV Registry of the Department of Health during the period 2007 to 2020. We aimed to identify temporal trends in the epidemiology and clinical manifestations of HIV-associated TB during that period.

We retrospectively reviewed data contained within the TB-HIV Registry, which captured information regarding nearly all cases of HIV-associated TB diagnosed in the Tuberculosis and Chest Service and Special Preventive Programme (SPP) under the Department of Health, as well as cases referred from regional hospitals of the Hong Kong Hospital Authority, during the period 2007 to 2020. The details of data contained in the TB-HIV Registry, along with the criteria for TB as a primary AIDS-defining illness, were described in a previous report.13 There have been no changes in the criteria for TB as a primary AIDS-defining illness since the last report. Where necessary, both clinic records and hospital discharge records were reviewed. All data were imported into Epi Info14 and exported to statistical software SPSS (Windows version 26.0; IBM Corp, Armonk [NY], US) for analysis. The Cochran–Armitage trend test in XLSTAT software (Lumivero, Denver [CO], US) was used to identify trends in the proportion of reported AIDS cases with TB as a primary AIDS-defining illness during the study period. The Cochran–Armitage trend test was also used to examine changes in demographic features and clinical manifestations during the same period. Where relevant, we compared the demographic features of patients reported to the TB-HIV Registry during the study period with the features of a historical cohort from the period 1996 to 200613 using the Chi squared test. P values <0.05 were considered statistically significant.

This study was an extension of a previous study designed to evaluate the public health programme for HIV-associated TB in Hong Kong15; it did not constitute research on human participants. Throughout the review process, we implemented all reasonable precautions to protect the confidentiality of personal data and excluded personally identifiable information from the electronic database.

All 390 cases reported to the TB-HIV Registry from 1 January 2007 to 31 December 2020 were included in this retrospective analysis. Information about whether TB constituted a primary AIDS-defining illness was available for 363 of 390 (93.1%) patients, where TB was listed as a primary AIDS-defining illness in 251 of those cases (69.1%). Overall, TB as a primary AIDS-defining illness represented a decreasing trend of 18.3% of 1375 reported AIDS cases during the period 2007 to 202012 (Cochran–Armitage trend test, P<0.001) [Fig], compared with 28.2% (192/680; Chi squared test, P<0.001) among the historical cohort of patients reported during the period 1996 to 2006.13

Trends in demographic features, including age, sex, case category, ethnicity, residence, primary source of care (first presentation), and smoking status, for the 390 HIV-associated TB cases reported to the TB-HIV Registry during the study period are shown in Table 1. The proportion of female patients significantly increased whereas that of ever-smokers significantly decreased (Cochran–Armitage trend test, P=0.035 and P=0.029, respectively). No significant trends were detected in other variables examined. Additionally, the proportions of Chinese individuals and permanent residents were lower in the present cohort than in the historical cohort of 1996 to 200613 (260/390, 66.7% vs 152/190, 80.0%; Chi squared test, P=0.001 and 258/390, 66.2% vs 144/190, 75.8%; Chi squared test, P=0.018, respectively). The proportion of female patients was significantly higher in the present cohort than in the historical cohort13 (82/390, 21.0% vs 21/190, 11.1%; Chi squared test, P=0.003).

Trends in clinical manifestations, including symptoms, presence of pulmonary TB, radiographic features (for cases with abnormalities on chest radiographs), presence of extrapulmonary TB (EPTB), most common EPTB sites, sputum smear positivity status, drug susceptibility patterns, presence of TB risk factors, CD4 cell count at TB diagnosis, presence of other AIDS-defining illnesses at the time of co-infection, and antiretroviral therapy (ART) status (among patients with a diagnosis of HIV infection before TB), among the 390 TB cases are presented in Tables 2 and 3. The proportions of patients presenting with site-specific symptoms (other than chest-related symptoms) and with EPTB both significantly increased during the period 2007 to 2020 (Cochran–Armitage trend test, P=0.029 and P=0.008, respectively) [Table 2]. The most common EPTB sites were lymph nodes (42.8%), pleura (21.5%), and abdomen (13.8%). Among patients who underwent sputum smear tests, the proportion of patients with sputum smear positivity significantly decreased (Cochran–Armitage trend test, P=0.006) [Table 2]. Among patients with lung parenchymal lesions on chest radiographs, a decreasing trend was observed in the proportion of patients in which the lower zone was the predominant lesion site (Cochran–Armitage trend test, P=0.045) [Table 3]. Among patients with a diagnosis of HIV infection before TB, the proportion of patients receiving ART at TB diagnosis significantly increased (Cochran–Armitage trend test, P=0.003) [Table 2].

This study revealed a decreasing trend in the proportion of reported AIDS cases with TB as a primary AIDS-defining illness during the period 2007 to 2020. The overall proportion (18.3%) was also lower than the proportion (28.2%) in the historical cohort of cases reported during the period 1996 to 2006.13 The proportions of Chinese individuals and permanent residents were lower, whereas the proportion of female patients was higher, in our cohort compared with the historical cohort.13 The proportions of female patients and patients with extrapulmonary involvement significantly increased, whereas the proportions of ever-smokers and the proportion with sputum smear positivity among pulmonary TB cases significantly decreased during the period 2007 to 2020. A decreasing trend was observed in the proportion of patients with pulmonary TB in which the lower zone was the predominant site of lung parenchymal lesions. Among patients with a diagnosis of HIV infection before TB, an increasing trend was observed in the proportion of patients receiving ART.

In Hong Kong, TB is considered an AIDS-defining illness when the disease is extrapulmonary. Pulmonary TB and cervical lymph node TB are considered AIDS-defining illnesses only when the CD4 cell count at the time of TB diagnosis is <200/μL, as recommended by the Scientific Committee of the Advisory Council on AIDS in 1994.16 Since then, there have been no changes in the criteria for TB as an AIDS-defining illness among individuals infected with HIV. The decreasing trend in Hong Kong regarding the proportion of reported AIDS cases with TB as a primary AIDS-defining illness might be due to decreased community transmission of TB through better TB control, the expansion of HAART since its introduction in 1997, and (perhaps to a lesser extent) increased acceptance of testing for latent TB infection and preventive treatment for TB among HIV-infected individuals since the early 2000s. Similar decreasing trends related to HAART-induced improvements in immune status among HIV-infected individuals have also been identified during studies conducted in some other countries.7 9 17 In an observational, retrospective study of AIDS cases included in the Barcelona AIDS register between 1994 and 2005, decreases were observed regarding the incidence of TB as an AIDS-defining illness among both native and immigrant populations.7 Another study examining trends in the incidence of AIDS-defining opportunistic illnesses over a 25-year period in Brazil showed a reduction in TB incidence from 1991-1993 to 2009-2012.9 A prospective cohort study of participants in the HIV Outpatient Study at 12 HIV clinics within the US indicated that TB incidence decreased after HAART introduction.17 Conversely, among participants in the HOMER cohort study (HAART Observational Medical Evaluation and Research) conducted during the period 1996 to 2007 in Canada, no statistically significant trends were observed in the proportion of cases with TB as the AIDS-defining illness, probably because the small number of TB cases reported in each time period limited the ability to detect significant changes in the reported cases.18 Another study examining AIDS notification data in Australia during the period 1993 to 2000 revealed that the proportion of AIDS cases with TB as the AIDS-defining illness was higher during 1996 to 2000 (post-HAART era) than during 1993 to 1995 (pre-HAART era); the authors attributed this difference to the increasing proportion of Australian patients with AIDS who had been born in sub-Saharan Africa and Asia during the 1990s, among whom the risk of TB was considerably higher.19 Further studies examining trends in TB as an AIDS-defining illness, as well as location-specific factors that may influence such trends in the post-HAART era, are warranted to facilitate control strategies for HIV-associated TB.

Further attention is needed regarding the observation of higher proportions of non-Chinese individuals (mostly Asians and Africans, who have much higher TB incidence than the rate among Hong Kong Chinese individuals) and non-permanent residents of Hong Kong in the 2007-2020 cohort compared with the historical 1996-2006 cohort. Similar findings of higher incidences of AIDS-associated TB in foreign-born populations from countries with much higher TB incidence compared with the native population have been reported on the basis of some studies conducted in developed countries.7 20 21 These observations highlight the need for TB screening and prophylaxis for people living with HIV who were born in countries with a high background prevalence of TB.

Intriguingly, we observed an increasing trend in the proportion of women during the period 2007 to 2020. The reason for this increase is unclear; it may be related to changes in the societal roles of men and women that influence exposure risk. The role of an increased proportion of EPTB (reportedly associated with female sex and observed throughout our cohort, as discussed below) requires further investigation.

The proportion of ever-smokers significantly decreased during the period 2007 to 2020, consistent with the findings of some studies conducted in the US.22 23 In an analysis of patients from a HIV surveillance system in the US, the prevalence of current smoking declined from 37.6% in 2009 to 33.6% in 2014.22 In another prospective cohort study that examined smoking trends among HIV-positive patients in the US, a decline in the annual prevalence of current smoking from 1984 to 2012 was also reported; however, disparities were noted according to race, ethnicity, and education.23 Nonetheless, because smoking increases HIV-related and non–HIV-related morbidity and mortality among people living with HIV, smoking cessation interventions remain an essential component of routine care for such individuals.

The predominance of the lower zone as the site of lung parenchymal lesions on chest radiographs is a relatively common feature among patients with HIV-associated pulmonary TB, according to our previous report on the 1996 to 2006 cohort13 and some other reports.24 25 The present study showed that the lower zone was less frequently the predominant site of lung parenchymal lesions during the period 2007 to 2020. The proportion of patients in the 2007 to 2020 cohort with the lower zone as the predominant site (16.7%) was also lower compared with the proportion of patients in the historical 1996 to 2006 cohort (32.4%).13 This difference may be related to the higher CD4 cell count at TB diagnosis among patients in the current cohort compared with patients in the historical cohort (median CD4 cell counts at TB diagnosis: 100/μL and 78/μL, respectively). Nonetheless, lower zone involvement was present in approximately one-sixth of pulmonary TB cases reported during 2007 and 2020. A high index of suspicion is required for the accurate and timely diagnosis of pulmonary TB in people living with HIV.

A decreasing trend was observed in the proportion of patients with sputum smear positivity during the period 2007 to 2020. The overall proportion of patients with sputum smear positivity in the 2007 to 2020 cohort (36.6%) was also lower than the proportion in the historical cohort of 1996 to 2006 (42.2%)13; it was similar to the proportion identified during a population database study in South Korea (36.4%).26 These results suggest that TB cases have been diagnosed at increasingly earlier stages due to enhanced active TB screening efforts among people living with HIV, as well as the increased use of molecular testing that enhanced the diagnosis of smear-negative cases in Hong Kong. These results highlight the need for continued TB screening efforts and early detection of TB among people living with HIV.

Our findings indicate that EPTB became more common among HIV-associated TB patients during the period 2007 to 2020. The overall proportion of patients with EPTB was also higher in the present cohort compared to that reported in a local study that examined risk factors for mortality in an earlier cohort (2006 to 2015) and that in the historical 1996 to 2006 cohort (71.0%, 64.9%,15 and 62.6%,13 respectively). These differences may have arisen through enhanced diagnosis of EPTB with the increased use of molecular testing in Hong Kong. An increasing trend of extrapulmonary involvement among TB patients has also been reported in some other studies, although such studies are mostly population-based.27 28 29 Few reports have been published regarding temporal trends in EPTB specifically among HIV-associated TB patients.30 Further studies are needed to examine these trends and associated factors.

Strengths of this study included its use of cohort data from the TB-HIV Registry covering a relatively long period (14 years) to study temporal changes in the epidemiology and clinical manifestations of HIV-associated TB. However, some limitations should be considered when interpreting the results of this study. First, the TB-HIV Registry may not capture all HIV-associated TB cases—some patients encountered in the SPP were not referred to a chest clinic but underwent anti-TB treatment in private clinics or other countries. The total number of HIV-associated TB cases in the HIV Surveillance Report of SPP was approximately 10% higher than the number in the TB-HIV Registry; the difference mostly comprised non-permanent residents temporarily staying in Hong Kong. Nonetheless, data from the HIV Surveillance Report showed a similar decreasing trend in TB as a primary AIDS-defining illness during the study period (data not shown). Second, this study utilised a retrospective design, and data present in the database of the TB-HIV Registry may be incomplete. Information regarding some parameters such as case category, co-morbidities, and CD4 cell count was unavailable for some patients. To overcome this limitation, we traced and reviewed relevant clinical records from chest clinics and hospitals when necessary. Therefore, we expect minimal bias due to missing data. Finally, the sample size may have led to insufficient statistical power for detecting temporal changes in some less common parameters.

This study showed that TB has become less important as a primary AIDS-defining illness in Hong Kong over the 14 years of the study period. Nonetheless, it remains the second most common primary AIDS-defining illness after P jirovecii pneumonia. Important temporal changes were also observed in the patterns of demographic features and clinical manifestations. Continued surveillance regarding the patterns of demographic features and clinical manifestations is needed to inform policymakers during the formulation of TB control strategies to improve patient care and treatment outcomes among people living with HIV. This surveillance is especially important in situations such as the coronavirus disease 2019 era, during which resources from TB programmes may be diverted to management of the global pandemic.

Concept or design: ACK Chan, SS Huang.
Acquisition of data: ACK Chan, SS Huang.
Analysis or interpretation of data: ACK Chan, SS Huang.
Drafting of the manuscript: ACK Chan, SS Huang.
Critical revision of the manuscript for important intellectual content: KH Wong, CC Leung, MP Lee, TY Tsang, WS Law, LB Tai.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

The authors thank all of their colleagues in the Tuberculosis and Chest Service and the Special Preventive Programme of the Department of Health who provided assistance and support to make this paper possible. The authors also thank Ms Ida KY Mak, Research Officer at the Tuberculosis and Chest Service of the Department of Health, for her dedicated efforts in maintaining the Tuberculosis–Human Immunodeficiency Virus Registry and assisting with the analysis.

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

This research was a retrospective analysis of observational data routinely collected in a local Registry as part of the ongoing evaluation of the public health programme for human immunodeficiency virus–associated tuberculosis in Hong Kong. Approval for the evaluation and exemption from obtaining informed patient consent has been granted by the Ethics Committee of the Department of Health of the Hong Kong SAR Government (Ref No.: L/M 416/2017).

1. Low A, Gavriilidis G, Larke N, et al. Incidence of opportunistic infections and the impact of antiretroviral therapy among HIV-infected adults in low- and middleincome countries: a systematic review and meta-analysis. Clin Infect Dis 2016;62:1595-603. Crossref

2. B-Lajoie MR, Drouin O, Bartlett G, et al. Incidence and prevalence of opportunistic and other infections and the impact of antiretroviral therapy among HIV-infected children in low- and middle-income countries: a systematic review and meta-analysis. Clin Infect Dis 2016;62:1586-94. Crossref

3. Rubaihayo J, Tumwesigye NM, Konde-Lule J. Trends in prevalence of selected opportunistic infections associated with HIV/AIDS in Uganda. BMC Infect Dis 2015;15:187. Crossref

4. World Health Organization. Global Tuberculosis Report 2022. Geneva: World Health Organization; 2022. Available from: https://www.who.int/teams/global-tuberculosis-programme/tb-reports/global-tuberculosis-report-2022. Accessed 4 Jul 2024.

5. World Health Organization. WHO releases new global lists of high-burden countries for TB, HIV-associated TB and drug-resistant TB. 2021. Available from: https://www.who.int/news/item/17-06-2021-who-releases-new-global-lists-of-high-burden-countries-for-tb-hiv-associated-tb-and-drug-resistant-tb. Accessed 4 Jul 2024.

6. Benito N, Moreno A, Miro JM, Torres A. Pulmonary infections in HIV-infected patients: an update in the 21st century. Eur Respir J 2012;39:730-45. Crossref

7. Martin V, García de Olalla P, Orcau A, Caylà JA. Factors associated with tuberculosis as an AIDS-defining disease in an immigration setting. J Epidemiol 2011;21:108-13. Crossref

8. González-García A, Fortún J, Elorza Navas E, et al. The changing epidemiology of tuberculosis in a Spanish tertiary hospital (1995-2013). Medicine (Baltimore) 2017;96:e7219. Crossref

9. Coelho L, Cardoso SW, Amancio RT, et al. Trends in AIDS-defining opportunistic illnesses incidence over 25 years in Rio de Janeiro, Brazil. PLoS One 2014;9:e98666. Crossref

10. Centre for Health Protection, Department of Health, Hong Kong SAR Government. Annual Report. Tuberculosis and Chest Service. 2020. Available from: https://www.info.gov.hk/tb_chest/doc/Annual_Report_2020.pdf. Accessed 16 Nov 2022.

11. Centre for Health Protection, Department of Health, Hong Kong SAR Government. Notification & death rate of tuberculosis (all forms), 1947-2022. Available from: https://www.chp.gov.hk/en/statistics/data/10/26/43/88.html. Accessed 17 Nov 2022.

12. Centre for Health Protection, Department of Health, Hong Kong SAR Government. HIV Surveillance Report—2020 Update. 2021. Available from: https://www.chp.gov.hk/ files/pdf/aids20.pdf. Accessed 16 Nov 2022.

13. Chan CK, Alvarez Bognar F, Wong KH, et al. The epidemiology and clinical manifestations of human immunodeficiency virus–associated tuberculosis in Hong Kong. Hong Kong Med J 2010;16:192-8.

14. Centers for Disease Control and Prevention, United States Government. Epi Info. Available from: https://www.cdc.gov/epiinfo/index.html. Accessed 4 Jul 2024.

15. Chan CK, Wong KH, Lee MP, et al. Risk factors associated with 1-year mortality among patients with HIV-associated tuberculosis in areas with intermediate tuberculosis burden and low HIV prevalence. Hong Kong Med J 2018;24:473-83. Crossref

16. Scientific Committee of the Advisory Council on AIDS Hong Kong. Classification system for HIV infection and surveillance case definition for AIDS in adolescents and adults in Hong Kong. 1995. Available from: https://www.aids.gov.hk/pdf/g40.pdf. Accessed 16 Nov 2022.

17. Buchacz K, Baker RK, Palella FJ Jr, et al. AIDS-defining opportunistic illnesses in US patients, 1994-2007: a cohort study. AIDS 2010;24:1549-59. Crossref

18. Jafari S, Chan K, Aboulhosn K, et al. Trends in reported AIDS defining illnesses (ADIs) among participants in a universal antiretroviral therapy program: an observational study. AIDS Res Ther 2011;8:31. Crossref

19. Dore GJ, Li Y, McDonald A, Ree H, Kaldor JM; National HIV Surveillance Committee. Impact of highly active antiretroviral therapy on individual AIDS-defining illness incidence and survival in Australia. J Acquir Immune Defic Syndr 2002;29:388-95. Crossref

20. Camoni L, Regine V, Boros S, Salfa MC, Raimondo M, Suligoi B. AIDS patients with tuberculosis: characteristics and trend of cases reported to the National AIDS Registry in Italy—1993-2010. Eur J Public Health 2012;23:658-63. Crossref

21. Dore GJ, Li Y, McDonald A, Kaldor JM. Spectrum of AIDS-defining illnesses in Australia, 1992 to 1998: influence of country/region of birth. J Acquir Immune Defic Syndr 2001;26:283-90.

22. Frazier EL, Sutton MY, Brooks JT, Shouse RL, Weiser J. Trends in cigarette smoking among adults with HIV compared with the general adult population, United States—2009-2014. Prev Med 2018;111:231-4. Crossref

23. Akhtar-Khaleel WZ, Cook RL, Shoptaw S, et al. Trends and predictors of cigarette smoking among HIV seropositive and seronegative men: the multicenter AIDS cohort study. AIDS Behav 2016;20:622-32. Crossref

24. Affusim C, Abah V, Kesieme EB, et al. The effect of low CD4+ lymphocyte count on the radiographic patterns of HIV patients with pulmonary tuberculosis among Nigerians. Tuberc Res Treat 2013;2013:535769. Crossref

25. Chamie G, Luetkemeyer A, Walusimbi-Nanteza M, et al. Significant variation in presentation of pulmonary tuberculosis across a high resolution of CD4 strata. Int J Tuberc Lung Dis 2010;14:1295-302.

26. Kim JM, Kim NJ, Choi JY, Chin BS. History of acquired immune deficiency syndrome in Korea. Infect Chemother 2020;52:234-44. Crossref

27. Ben Ayed H, Koubaa M, Marrakchi C, et al. Extrapulmonary tuberculosis: update on the epidemiology, risk factors and prevention strategies. Int J Trop Dis 2018;1:006.

28. Peto HM, Pratt RH, Harrington TA, LoBue PA, Armstrong LR. Epidemiology of extrapulmonary tuberculosis in the United States, 1993-2006. Clin Infect Dis 2009;49:1350-7. Crossref

29. Sandgren A, Hollo V, van der Werf MJ. Extrapulmonary tuberculosis in the European Union and European Economic Area, 2002 to 2011. Euro Surveill 2013;18:20431. Crossref

30. Mohammed H, Assefa N, Mengistie B. Prevalence of extrapulmonary tuberculosis among people living with HIV/AIDS in sub-Saharan Africa: a systemic review and meta-analysis. HIV AIDS (Auckl) 2018;10:225-37. Crossref