Prevalence of HIV infection among tuberculosis patients in a teaching ...

HIV & AIDS Review 15 (2016) 136–140

Contents lists available at ScienceDirect

HIV & AIDS Review journal homepage: www.elsevier.com/locate/hivar

Original Research Article

Prevalence of HIV infection among tuberculosis patients in a teaching hospital in south-west Nigeria: A four-year retrospective study Kolade O. Ranti a, Atilola O. Glory a,*, Babalola T. Victoria b, Komolafe O. Isaac a a b

Department of Biological Sciences/Microbiology Unit College of Natural Sciences, Redeemer’s University, PMB 210 Ede, Osun State, Nigeria Department of Community Medicine and Primary Care, Olabisi Onabanjo University Teaching Hospital, Sagamu, Ogun State, Nigeria

A R T I C L E I N F O

A B S T R A C T

Article history: Received 22 March 2016 Accepted 5 November 2016 Available online 23 November 2016

Introduction: According to the World Health Organization, Nigeria remains one of the worst affected countries with HIV and Mycobacterium tuberculosis infections in the world. Of a critical concern is the emerging threat of co-infection at an unprecedented proportion in the country. In south-west Nigeria, few studies have assessed the prevalence of HIV infection among TB patients. Findings aimed at estimating the prevalence of HIV co-infection among TB patients enrolled in a teaching hospital in Ogun State, south-west Nigeria between 2008 and 2011 are presented in this study. Materials and methods: This is a retrospective study of 386 TB patients admitted between January 2008 and 2011 to Olabisi Onabanjo University Teaching Hospital, Ogun State, Nigeria of which 206 (53%) were males and 180 (47%) were females. Descriptive statistics, chi-square test and logistic regression were utilized in the analysis of the data. Results: An overall HIV prevalence of 29.27% (95% CI: 24.78–34.09) was found. Female TB patients had higher odds of HIV infection (OR 1.91, 95% CI: 1.03–3.55) while patients with the highest education had the least odds of HIV infection [OR 0.08, 95% CI: 0.01–0.56]. Conclusion: Evidence of a rising tide of HIV infection in TB patients was found, in particular among single middle-aged women with low education. Further research also needs to be conducted to provide more insight into the epidemiology of co-infection in order to better address the dual burden of HIV and TB among tuberculosis patients in Nigeria. ß 2016 Polish AIDS Research Society. Published by Elsevier Sp. z o.o. All rights reserved.

Keywords: HIV infection TB patients Nigeria

1. Introduction Human immunodeficiency virus (HIV) and Mycobacterium tuberculosis infections remain the leading causes of morbidity and mortality from infectious disease in both children and adults in sub-Saharan Africa [1]. Most recent global estimates of incident cases of HIV and active TB infections stood at 2.5 million and 9 million respectively with 13% of TB cases co-infected with HIV [2,3]. The mechanisms of susceptibility of tuberculosis patients to co-infection with HIV remain largely unknown and promoted by a complex set of socio-behavioral and environmental factors [4]. The overall aim of this study was to determine the prevalence of HIV among TB patients at the Olabisi Onabanjo University Teaching Hospital, Ogun State, Nigeria between the period of 2008 and 2011. Nigeria has recorded an overall steady decline in the proportion of people infected with HIV from a peak of 5.8% in 2001 to 3.2% in

* Corresponding author. E-mail address: [email protected] (A.O. Glory).

2013. The country remains the second worst HIV affected country in the world with a total estimated burden of 3.4 million people living with HIV/AIDS [3]. There has been a rapidly growing trend in the epidemiologic synergy forged between the two diseases particularly in sub-Saharan Africa as one potentiates the other and complicate other unrelated infectious diseases among vulnerable groups. While the literature is sparse with regard to attempt to define the prevalence magnitude of the rising co-epidemic and its socio-economic dimension in south-west Nigeria, a few studies have signaled the potential of the two diseases to erode past public health gains as well as slowing down effectiveness of current control efforts in Nigeria [4,5]. Furthermore, TB is known to be preventable and curable; however, it remains the most common cause of AIDS-related deaths in Nigeria and sub-Saharan Africa. Conversely, among individuals with latent TB within a population, HIV remains the most significant cause of reactivation of tuberculosis, and rapid progression to death if not diagnosed early or effectively treated [6]. However, being a disease that thrives in poor living conditions, TB ensures that effective control can be attained by proper

http://dx.doi.org/10.1016/j.hivar.2016.11.001 1730-1270/ß 2016 Polish AIDS Research Society. Published by Elsevier Sp. z o.o. All rights reserved.

K.O. Ranti et al. / HIV & AIDS Review 15 (2016) 136–140

implementation of sanitary and hygienic practices which include avoidance of overcrowding (especially in households, schools, correctional and health facilities) and the minimization of air pollution [7,8]. Evidence of high burden TB reactivation has also been found among immigrants in countries with low incidence of tuberculosis as reported by Lillebaek and are colleagues [9]. In Nigeria, few studies have assessed presentation and treatment outcome of HIV and TB co-infection both in children and adult TB patient populations. A five-year retrospective assessment conducted by Daniel and his Colleagues in Ogun State reported a 10.5% period prevalence of HIV among 76 children with TB [10]. In a separate study conducted by Umeh and others among patients who presented with respiratory symptoms in a referral chest clinic, the risk of co-infection was about three times higher among HIV positives than TB positives [11]. The study also reported a 12.8% prevalence of HIV/TB co-infection [11]. Certain socio-cultural factors have been shown to exacerbate co-infection in Nigeria which include gender inequality and sexual violence entrenched in diverse Afro-cultural norms particularly among rural and urban slum dwellers in sub-Saharan Africa [12,13]. However, while ample studies exist with respect to the prevalence of HIV in TB patients across different parts of Nigeria, a dearth of studies exist with regard to the sero-prevalence of HIV co-infection among TB patients on DOTS therapy in south-west Nigeria. In the present study findings from a four-year retrospective assessment of HIV prevalence among TB patients in a teaching hospital in south-west Nigeria are presented. 2. Materials and methods 2.1. Study design, setting and target population This study was a retrospective descriptive and analytic assessment of TB patient records. A retrospective review of all TB patient records enrolled at the Olabisi Onabanjo University Teaching Hospital (OOUTH) in Ogun State from 2008 to 2011 was conducted. The primary study was conducted at the department of Virology and Directly Observed Treatment Short Course (DOTS) Clinic, at the OOUTH Sagamu, Ogun State in south-west, Nigeria. The town (Sagamu) is an urban area located about 50 km from the metropolitan city of Lagos with an estimated population of 253 421 as of 2006 census. 2.2. TB test and HIV diagnosis in enrolled patients TB case diagnosis was conducted using acid-fast bacilli (AFB) test, Mantoux test and chest X-ray radiography among patients attending OOUTH, Sagamu, between 2008 and 2011. This period was selected due to completeness of the record. This was followed by clinical diagnosis of HIV infection and subsequent confirmation using Determine1 and Starpac1 Diagnostic kits. 2.3. Data collection, management and statistical analysis Two independent study personnel conducted a record review on 386 TB patient registers. Information on demographic parameters of patients including age, gender, marital status, occupation and level of education were retrieved from each patient’s record. Extracted data was captured in excel spreadsheets. As part of the data management procedure, cleaning, coding and recoding of extracted data were carried out. Descriptive analysis was carried out to assess the distribution of TB and HIV co-infection with respect to socio-demographic variables using frequencies and proportions. Significant difference between patients who had only TB and patients with TB-HIV coinfection with respect to socio-demographic variables (at 5% alpha

137

level) was also reported. Bivariate analysis was then conducted to determine independent associations of socio-demographic variables with the study outcome (TB-HIV co-infection). Bivariate associations significant at 10% alpha level were considered for inclusion using the stepwise-variable selection procedure in the multivariable binary logistic regression model. All statistical significance was reported at 5% alpha level along with the 95% confidence intervals (95% CI) while analyses were implemented in STATA general statistical software package version 12.1. 2.4. Ethical approval Approval for this study and access to patient records were given by the Ethical Review Committee of Olabisi Onabanjo University Teaching Hospital. In order to forestall a breach in patient confidentiality, record review and data extraction were conducted in a secluded compartment of the health facility. 3. Results 3.1. Descriptive Of the 386 TB patients that were screened for HIV infection within the study period, an estimated 113 (29.3%) had HIV infection. The highest number of TB cases screened was 139 in 2008 of which 20% had HIV infection while the lowest was 71 in 2011 with 42% sero-positive for HIV infection. A total of 47 (22.8%) male TB patients tested positive to HIV compared to 66 (37%) of females (p < 0.003). HIV positivity rate was found to be highest (34% of all HIV infections) in TB patients in the modal middle age group (25–49 years). Results also showed that while Table 1 Demographic distribution of study population in relation to HIV status. Total TB patients enrolled (N = 386)

HIV positive N (%) 113 (29.27)

HIV negative N (%) 273 (70.73)

Gender Male Female

206 180

47 (22.82) 66 (36.67)

159 (77.18) 114 (63.33)

Age 0–24 25–49 50

121 226 39

26 (21.49) 77 (34.07) 10 (25.64)

95 (78.51) 149 (65.93) 29 (74.36)

Marital status Single Married Widowed/divorced

138 97 13

34 (24.64) 23 (23.71) 6 (46.15)

104 (75.36) 74 (76.29) 7 (53.85)

Occupation Unemployed Student Employed Business Artisan

5 93 13 121 16

2 (40) 24 (25.81) 2 (15.38) 34 (28.10) 1 (6.25)

3 (60.00) 69 (74.19) 11 (84.62) 87 (71.90) 15 (93.75)

Education None Primary Secondary Higher

11 78 132 27

5 (45.45) 27 (34.62) 29 (21.97) 2 (7.41)

6 (54.55) 51 (65.38) 103 (78.03) 25 (92.59)

Year of diagnosis 2008 2009 2010 2011

139 93 83 71

28 33 22 30

111 (79.86) 60 (64.52) 61 (73.49) 41 (57.75)

Variable

P-value ( chi2 = 0.9801; area under ROC curve: 0.7840. SE: standard error.


out an independent assessment of possible associations between TB-HIV co-infection and socio-demographic factors in female patients. Results of this model are presented in Table 3. The results showed remarkable peculiarities with regard to how sociodemographic factors predict differentials and likelihood of coinfection across the various variable categories compared to the general model. Overall, age groups and marital status also emerged alongside other previously identified variables to be significant predictors of co-infection in TB patients (Table 3). For instance, female patients aged 25–49 years were 16 times more likely to have TB-HIV co-infection relative to female patients aged 0–24 years (Table 3). A similar pattern with regard to the level of education was also observed in female patients with higher education lowering the odds of TB-HIV co-infection. With regard to marital status, married (OR: 0.05; 95% CI: 0.01–0.31) and widowed/divorced female patients had lower odds of co-infection when compared to single female TB patients holding other factors constant in the model. The model also revealed that female patients diagnosed in 2009 and beyond had significantly higher odds of co-infection relative to female patient diagnosed in 2008, holding other factors constant in the model. 4. Discussion In our study, we set out to assess the prevalence of HIV infection among patients with active Mycobacterium bacterium infection at the Olabisi Onabanjo University Teaching Hospital in Ogun State, Nigeria. An overall HIV prevalence of 29.3% was observed among the 386 TB patients enrolled during the study period with the highest rate observed in 2011 (42%) and lowest (20%) in 2008 (Table 1). This finding was consistent with other health facilitybased studies conducted in different parts of the country [11,14– 17]. However, co-infection estimate was lower when compared to similar studies conducted by Pennap and others (44.2%) [12]; and Ofoegbu and Odume (42.7%) [18], and relatively higher, compared to other studies with a range from 0% [19], 5.9% [20], 11.9% [21], 18.2% [22], to 18.8% [23]. Prevalence finding was also significantly high compared to study by Okonko and others [19], who found no evidence of TB/HIV co-infection among patients on directly observed treatment of short course in Ogun state. A comparison of this prevalence finding to estimates reported in a recent systematic review and meta-analysis showed approximately a 6% point higher than the pooled prevalence estimate (29.3% versus 23.5%) and a 2% point lower than the average estimate in African countries (29.3% versus 31.3%) [21]. Furthermore, we found a significant difference in the prevalence of HIV infection between male and female TB patients with higher likelihood of co-infection observed in female patients (36.7% as against 22.8% in male). This finding was consistent with the results of studies carried out by others in different parts of Nigeria [12,14,17,21,23]. This pattern of higher prevalence of HIV coinfection among female TB patients was found to be reported in all the studies carried out so far in Nigeria particularly in the productive age group (25–49 years). While multiple factors may account for this trend, the higher probability of females for HIV acquisition may account for the most significant part of the multiple causal model of the observed outcome [25]. Furthermore, a positive linear dose–response association was observed between the HIV co-infection and education. The strong link between the risk of HIV acquisition and level of education has been shown by several studies which clearly demonstrate the benefit of secondary and higher education in reducing the risk of HIV infection [24]. Conversely, lower level or no education among women in this population may also detract from their regular access to relevant information and health-seeking behavior. This vulnerable group of the women population has a high level of

139

dependency on their male partners, thus creating perfect breeding ground for sexual exploitation and domestic violence [25]. In addition, a sub-group analysis of the female TB patients showed that married female TB patients are 95% less likely to be infected with HIV infection compared to single female patients (Table 3). Our study found no statistically significant association between the risk of co-infection and patient’s age group. However, with regard to age differential, highest HIV prevalence was observed in TB patients in the middle age group (25–49 years) a reflection of the predilection of this highly sexually active age group. The high potential for unrestrained sexual tendencies is known to increase the risk of STIs (including HIV) in untold millions of persons in the middle age of life across the globe [25]. Commercial sex work, transactional and concurrent sexual partnerships are more often than not, associated with female individuals within this age group which further strengthens the evidence for higher observed rate of coinfection in the female gender [25]. It is worth mentioning that our study found the highest HIV positivity in 2011 which is significantly higher when compared to the figure for 2008. The only possible explanation that might explain this observed trend was the increase in intervention coverage, availability and access to HIV testing services coupled with heightened awareness of infection among patients with active TB infections, thus leading to higher rate of detection and uncovering existing cases of HIV infection previously unknown. In conclusion, a profile analysis revealed that in the study population, middle-aged female TB patients with low or no education are the most at risk of acquiring HIV infection. Therefore, HIV prevention effort and integrated response targeted at this subpopulation of TB patients in high-risk communities across the country will go a long way in significantly stemming the rising tide of HIV/TB co-infection among individuals with underlying TB infection. 5. Limitations of the study The authors are aware of the limitations posed by the retrospective nature of the study particularly with regard to the number of explanatory variables considered and in turn, residual confounding that may not be adequately accounted for in the fitted regression models to better explain the observed associations. Conflict of interest None declared. Financial disclosure None declared. Acknowledgement The authors would like to acknowledge the Management board of OOU Teaching Hospital for granting access to review the data authors employed in this retrospective study. References [1] E. Corbett, B. Marston, G. Churchyard, K. DeCock, Tuberculosis in sub-Saharan Africa: opportunities, challenges, and change in the era of antiretroviral treatment, Lancet 367 (2006) 926–937. [2] WHO Global Tuberculosis Report, 2014 Available at: http://apps.who.int/iris/ bitstream/10665/137094/1/9789241564809_eng.pdf. [3] UNAIDS, UNAIDS Report on the Global AIDS Epidemic, 2013 Available at: http:// www.unaids.org/sites/default/files/en/media/unaids/contentassets/documents/ epidemiology/2013/gr2013/UNAIDS_Global_Report_2013_en.pdf.

140


[4] A. Otu, A review of the national tuberculosis and leprosy control programme (NTBLCP) of Nigeria: challenges and prospects, Ann. Trop. Med. Public Health 6 (5) (2013) 491–500. [5] S. AbdoolKarim, G. Churchyard, Q. AbdoolKarim, S. Lawn, HIV infection and tuberculosis in South Africa: an urgent need to escalate the Public Health response, Lancet 374 (9693) (2010) 921–933. [6] C. Kwan, J. Ernst, HIV and tuberculosis: a deadly human syndemic, Clin. Microbiol. Rev. 24 (2) (2011) 351–376. [7] J. Hargreaves, D. Boccia, C. Evans, M. Adato, M. Petticrew, J. Porter, The social determinants of tuberculosis: from evidence to action, Am. J. Public Health 101 (4) (2011) 654–662. [8] C. Ogbudebe, J. Chukwu, C. Nwafor, A. Meka, N. Ekeke, N. Madichie, et al., Reaching the underserved: active tuberculosis case finding in urban slums in Southeastern Nigeria, Int. J. Mycobacteriol. 4 (1) (2015) 18–24. [9] T. Lillebaek, A. Andersen, A. Dirksen, E. Smith, L. Skovgaard, A. Kok-Jensen, Persistent high incidence of tuberculosis in immigrants in a low-incidence country, Emerg. Infect. Dis. 8 (7) (2002) 679–684. [10] O.J. Daniel, O.B. Ogunfowora, O.T. Oladapo, HIV sero-prevalence among children diagnosed with TB in Nigeria, Trop. Doct. 37 (4) (2007) 268–269. [11] E. Umeh, D. Ishaleku, C. Iheukwumere, HIV/tuberculosis coinfection among patients attending a referral chest clinic in Nasarawa state, Nigeria, J. Appl. Sci. 7 (6) (2007) 933–935. [12] G. Pennap, S. Makpa, S. Ogbu, Sero-prevalence of HIV infection among tuberculosis patients in a rural tuberculosis referral clinic in northern Nigeria, Pan Afr. Med. J. 5 (2010) 22. [13] A. Daftary, N. Padayatchi, Social constraints to TB/HIV healthcare: accounts from coinfected patients in South Africa, AIDS Care 24 (12) (2012) 1480–1486. [14] S.O. Kalu, P.U. Ele, R.O. Okonkwo, C.N. Ogbuagu, B.O. Oluboyo, Seroprevalence of human immunodeficiency virus (HIV) infection among tuberculosis patients attending TB/DOTS centre in Nnewi, Sex. Transm. Infect. 89 (2013) A215.

[15] A.H. Kwaru, A.B. Muhammad, E.E. Nwokedi, A. Umar, M.A. Magashi, Prevalence of human immunodeficiency virus among tuberculosis patients in Kano, Nigeria, Highl. Med. Res. J. 6 (1–2) (2008). [16] O. Erhabor, Z.A. Jeremiah, T.C. Adias, C.E. Okere, The prevalence of human immunodeficiency virus infection among TB patients in Port Harcourt Nigeria, HIV/AIDS Res. Palliat. Care 2 (2010) 1–5. [17] O. Victor, O. Stanley, Prevalence of HIV infection among tuberculosis (TB) patients in a TB/HAART-HAART referral centre in Nigeria, Int. J. Sci. Nat. 3 (1) (2012) 88–92. [18] O.S. Ofoegbu, B.B. Odume, Treatment outcome of tuberculosis patients at National Hospital Abuja Nigeria: a five year retrospective study, S. Afr. Fam. Pract. 57 (1) (2015) 50–56. [19] I.O. Okonko, F.A. Soleye, F.O. Adeniji, P.O. Okerentugba, HIV and TB co-infection among patients on directly observed treatment of short course in Abeokuta, Ogun state, Nigeria, Nat. Sci. 10 (6) (2012) 10–14. [20] C. Nwabuko, O. Ejele, A. Chukw, M. Nnoli, I. Chukwuonye, Prevalence of tuberculosis-HIV coinfection and relationship between tuberculosis and cd4/ESR in HIV patients in Niger Delta region of Nigeria, IOSR J. Dent. Med. Sci. 2 (4) (2012) 1–4. [21] O. Chuks, A. Anyabolu, I. Martin, K. Stephen, O. Maria, C. Chiamaka, Prevalence of HIV infection in pulmonary tuberculosis suspects; assessing the Nnamdi Azikiwe University Teaching Hospital, Nnewi, Nigeria, Adv. Life Sci. Technol. 14 (2013) 87–90. [22] V.U. Nwadike, G.C. Mabata, I.E. Kalu, K.C. Ojide, I.G. Nweke, C. Nwokeji, et al., Seroprevalence of HIV antibodies among patients presenting in a tuberculosis clinic South Eastern Nigeria, S. Pac. J. Technol. Sci. 1 (2) (2014) 201–206. [23] M. Yusuf, O. Azeez-Akande, M. Yusha’u, HIV sero-prevalence among adult with newly diagnosed pulmonary tuberculosis in Kano, Nigeria, J. Med. Trop. 15 (2) (2013) 140–143. [24] E. Amuta, T. Mary, A. Ishaku, A retrospective study on the epidemiological trend of human immunodeficiency virus (HIV) and pulmonary tuberculosis (PTB) coinfection in Nasarawa State, Nigeria, J. Nat. Sci. Res. 3 (7) (2013) 45–51. [25] R. Stephenson, A. Winter, M. Elfstrom, Community environments shaping transactional sex among sexually active men in Malawi, Nigeria and Tanzania, AIDS Care 25 (6) (2013) 784–792.