High prevalence of hepatitis B and syphilis co-infections among HIV patients initiating antiretroviral therapy in the north-west region of Cameroon

Abstract

Hepatitis B virus (HBV) and syphilis co-infections contribute significantly to HIV-associated morbidity and mortality, but the burden of these diseases is not fully appreciated in sub-Saharan Africa, as prevalence data are scarce. Both infections often remain undiagnosed in resource-limited settings because routine testing is not a part of most of the national guidelines. Epidemiological studies provide important information on prevalence and risk factors for such co-infections and can provide guidance for clinical management and for the development of test strategies. We analysed data on baseline characteristics, CD4 cell counts, HBV and syphilis co-infection rates of 690 patients enrolling for antiretroviral therapy in rural Cameroon. The prevalence of both hepatitis B surface antigen (HBsAg, 12.6%, 95% CI 10.1–15.1) and treponemal antibodies (11.4%, 95% CI 8.9–13.7) was high, with significantly higher prevalences for both infections in men; detection of treponemal antibodies increased with age. Although liver enzyme elevations were common, they were not useful to identify HBsAg-positive patients. In this setting, routine serological screening for HBV and syphilis co-infection should be considered to avoid complications and ongoing transmission.

Keywords

HIV hepatitis B HBV syphilis screening risk factors Africa co-infections

INTRODUCTION

Sub-Saharan Africa is the region with the highest burden of HIV worldwide. Access to antiretroviral therapy (ART) has been improved significantly in the recent years; about 3.9 million patients received ART in sub-Saharan Africa in 2009.¹ The quality of baseline evaluation and biochemical monitoring of ART is limited due to economic constraints, but may be critical for clinical outcome. Side-effects of ART and serious complications occur more often in patients with co-morbidities and co-infections. HIV/hepatitis B virus (HBV) co-infected patients are at an increased risk of liver-related morbidity and mortality,² partially attributable to ART-associated hepatotoxicity, but also to accelerated progression of HBV infection.^3,4 Some drugs used in antiretroviral regimens also have activity against hepatitis B, but the emergence of lamivudine drug resistance may be problematic.⁵

Syphilis is another important co-infection in HIV patients. Besides the fact that syphilis facilitates the acquisition of HIV infection, syphilis may also influence the natural course of HIV infection, although data are inconclusive.^6,7 If the infection remains undiagnosed, complications of late stage syphilis will develop in a proportion of affected individuals.

Few data exist concerning co-infections in HIV-patients enrolling for ART in Africa.^8,9 The Mbingo Baptist hospital has been performing an extended pre-ART work-up including routine hepatitis B and syphilis testing in all patients entering the HIV care program, although this is not mandated by the national strategy. The analysis and dissemination of results are of clinical and epidemiological interest. The identification of patients at particular risk for co-infections can guide test strategies and clinical management.

METHODS

In this report, we retrospectively analysed patient data extracted from medical charts at the Mbingo Baptist hospital, a referral hospital in the north-west region of Cameroon. All subsequent patients presenting to the HIV outpatient clinic between January 2005 and March 2010 who were eligible for ART and above the age of 15 years were included in the analysis. Eligibility for ART was assessed by a committee of treating physicians, social workers and pharmacists based on the national recommendations, current World Health Organization (WHO) guidelines and the availability of antiretroviral drugs. The general work-up for eligible patients included clinical examination and laboratory parameters, including CD4 cell count, full blood count, liver and kidney function tests and serological tests. Patients with clinical or serological evidence for syphilis were treated with antibiotics at the discretion of the attending physician, although this information was not recorded in the medical files. Analysis of CD4 cell count was done with GUAVA Easy CD4 (Millipore Inc, Billerica, MA, USA) according to the instructions of the manufacturer. Liver transaminases and creatinine were determined by using commercial kits (Ortho Clinical Diagnostics, Beerse, Belgium). Normal ranges according to the specifications of the manufacturer are 5–50 U/L for aspartate aminotransferase (AST), 3–61 U/L for alanine aminotransferase (ALT) and 0.7–1.5 mg/dL (62–133 µmol/L) for creatinine. Syphilis and HBV screening were performed from fresh blood samples as part of routine medical care using commercial rapid tests with reported specificities and sensitivities of 99.7% and 99.6%, respectively, for the treponemal antibody test, and 99.2% and 99.8%, respectively, for the hepatitis B surface antigen (HBsAg) test (Acon Biotech, Hangzhou, Zhejiang, China). No data on other HBV and syphilis parameters or other co-infections were available in this retrospective chart review. The performance of this data analysis and publication of the results were approved by the internal review board of the Mbingo Baptist Hospital.

Statistical analysis was performed using the SPSS v15 package and STATA v10. Unpaired t-tests were used for the comparison of age and weight between sexes. Laboratory and serological tests were compared using the Mann–Whitney U-test. Proportions of patients with elevated liver and renal function tests, WHO stages, as well as positive serological tests were compared using chi-squared tests. Patients were categorized into quartiles according to age at diagnosis of HIV infection in order to obtain four equally large groups (15–28 years, n = 196; 29–33 years, n = 165; 34–39, n = 153; >40 years, n = 176). All tests were two-sided. A P value ≤0.05 was considered statistically significant.

RESULTS

A total of 697 patients completed diagnostic work-up prior to starting first-line ART at Mbingo Baptist Hospital during the recruitment period and were included in the analysis. Seven patients were excluded because of incomplete data. Among a total of 690 included patients, mean age at HIV diagnosis was 35.0 years (standard deviation [SD] 9.5), and male patients were significantly older (mean 39.0 years, SD 9.6) than female patients (mean 32.9 years, SD 8.7, P < 0.01).

Median CD4 cell count was 154 cells/µL (n = 687, interquartile range [IQR] 79–245). Four hundred and forty-two patients (64.3%) had a CD4 cell count below 200 cells/µL and 218 patients (31.7%) below 100 cells/µL. No significant difference of CD4 cell counts was noted between men and women. The majority of patients (59.5%) had WHO stage III or IV diseases at the time of diagnosis. However, a considerable number of patients were asymptomatic with a clear female predominance (29.8% WHO stage I disease versus 15.6% in men, P < 0.01, Table 1). Serum levels of ALT (mean 45.6 U/L versus 35.6 U/L, P < 0.01) and AST (50 U/L versus 43 U/L, P = 0.058) were higher in men as compared with women. Notably, 43.6% of the patients had elevated serum AST levels, again with a higher proportion in men compared with women (51.4% versus 39.5%, respectively, P < 0.01, Table 2).

Table 1

Patient characteristics

	Total	Men	Women	P
N	690	243	447
Age at HIV diagnosis, mean years (standard deviation [SD])	35.0 (9.5)	39.0 (9.6)	32.9 (8.7)	<0.01*
WHO Stage I	171 (24.8%)	38 (15.7%)	133 (29.7%)	<0.01**
WHO Stage II	109 (15.7%)	27 (11.2%)	82 (18.1%)
WHO Stage III	315 (45.3%)	129 (53.5%)	186 (41.2%)
WHO Stage IV	98 (14.2%)	48 (19.8%)	50 (11.1%)
Weight, mean kg (SD)	62.0 (11.3)	65.9 (10.9)	60.0 (11.0)	<0.01*

Statistical tests used: *unpaired t-test, **Chi-squared test

Table 2

Results of laboratory investigations and serological tests

	Total	Men	Women	P
Severe anaemia, haemoglobin <80 g/L	31 (4.6%)	12 (5.1%)	19 (4.3%)	0.65
CD4 count (cells/µL), median (IQR)	154 (79–245)	158 (74–227)	154 (82–259)	0.28
Creatinine (µmol/L), median (IQR)	80 (62–88)	88 (71–106)	71 (62–88)	<0.01
Patients with elevated creatinine levels (>133 µmol /L)	43 (6.8%)	26 (11.9%)	17 (4.1%)	<0.01
Aspartate transaminase (AST), median U/L (IQR)	45 (28–70.5)	50 (31–79)	43 (27–67)	0.06
Patients with elevated AST levels (>50 U/L)	265 (43.6%)	107 (51.4%)	158 (39.5%)	0.01
Alanine transaminase (ALT), median U/L (IQR)	30 (17–48)	39 (18–55)	26 (16–43)	<0.01
Patients with elevated ALT levels (>61 U/L)	83 (16.9%)	39 (20.9%)	46 (14.7%)	0.08

IQR, interquartile range

Overall prevalence for HBsAg was 12.6% (95% CI 10.1–15.1). Male patients were more likely to have a positive HBsAg result than female patients (16.5% versus 10.4%, P = 0.03). Women had the lowest HBsAg prevalence in the age group 29–33 years (7.3%), with increasing prevalence in older age groups in contrast with men, where a stable prevalence at a high level was found throughout all age groups (>14.6% in all age groups).

Probability for hepatitis B increased with higher ALT blood levels (P = 0.01), whereas no association with AST was found. However, 70% of the patients with positive HBsAg had normal ALT blood levels and would have been missed by a liver enzyme-guided screening strategy. The positive predictive value of elevated ALT serum levels for HBsAg positivity was only 24%. Positive results for treponemal antibodies (syphilis) were more common in men (15.6% versus 9.1%, P = 0.01), reaching more than 20% in male patients above the age of 40 years. Overall prevalence of syphilis was 11.4% (95% CI 0.9–13.7, Table 3). No association between positive syphilis tests and laboratory tests including AST, ALT or creatinine was found.

Table 3

Hepatitis B surface antigen (HBsAg) and treponemal antibody (syphilis) prevalence according to age group

	HBsAg positive			Treponemal antibody positive
	Men	Women	P*	Men	Women	P*
Age
15–28 years	5/34 (14.7%)	15/156 (9.6%)		5/34 (14.7%)	11/156 (7%)
29–33 years	10/50 (20%)	8/110 (7.3%)		4/50 (8%)	6/110 (5.5%)
34–39 years	9/58 (15.5%)	12/89 (13.5%)		8/58 (13.8%)	10/89 (11.2%)
≥40 years	15/95 (15.8%)	10/76 (13.2%)		20/95 (21.1%)	12/76 (15.8%)
Overall	39/237 (16.5%)	45/431 (10.4%)	0.03	37/237 (15.6%)	39/431 (9.1%)	0.01

*Chi-squared test performed

All patients received lamivudine as part of their initial ART regimen and the combination of stavudine/lamivudine/nevirapine (Triomune) was prescribed in 85.8% of the patients. Protease inhibitor-containing or triple nucleoside therapies were not used as first line therapy.

DISCUSSION

This study documents a high prevalence of HBV and syphilis co-infections in adult HIV patients enrolling for ART in rural Cameroon. Existing data on HIV/HBV co-infection in sub-Saharan Africa indicate a high regional variability, ranging from 9% in Côte d'Ivoire to 17% in South Africa.^10,11 In our study, the overall HBsAg prevalence was 12.6% (95% CI 10.1–15.1) and was significantly higher in men (16.5% versus 10.4% in women, P=0.03). A peak of 20% was reached in male patients aged 29–33 years. The prevalence reported in the rural settings of our study is in line with previously published data from major hospitals in the capital Yaoundé (10.1%).⁹ Although elevations of AST and ALT were common (43.6% and 16.9%, respectively), they were non-specific for HBV co-infection. The positive predictive value of abnormal ALT serum levels to predict HBsAg positivity was only 24%, thus a liver enzyme-guided screening strategy is not suitable in these settings.¹² The vast majority of patients in this study (86%) were started on ART containing nevirapine, which has an associated risk of hepatotoxic reactions, especially in HBV and HCV co-infected patients.³

All patients in our study were started on antiretroviral regimens containing lamivudine, which is active against HBV infection. Use of lamivudine alone in patients with HBV infection is associated with rapid development of resistance and treatment failure. It has been demonstrated that lamivudine resistance occurs in >90% of HIV/HBV co-infected patients exposed to lamivudine for >4 years.⁵ Patients who develop lamivudine resistance are more likely to develop progression of liver disease and the beneficial effect on histological improvement of fibrosis may be blunted in the longer term.^13,14

This has consequences for the infected individual, but also for the community because of the ongoing transmission and the possible transmission of resistant viruses. Recently, tenofovir, a drug with potent anti-HIV and anti-HBV activity, has become available in Cameroon and its use as part of ART regimens in patients with HIV/HBV co-infections is recommended by current WHO guidelines.¹² Among HIV-positive patients in this setting, routine serological HBV screening can help to identify patients at high-risk for liver cirrhosis, liver cancer and liver-associated death who will benefit from ART containing tenofovir, which is associated with a low risk of HBV resistance development and treatment failure even in the longer term.¹⁵ Knowledge on HBV serostatus of HIV patients can further help to prioritize treatment decisions regarding the indicated use of tenofovir when the economic situation does not permit widespread use. The high rates of abnormal creatinine serum levels (11.9% in male patients) in our study is, however, remarkable and warrants further investigations on the safety of a broad introduction of tenofovir in this setting. In addition, vaccination of HBsAg-negative patients could help to reduce HBV infections given the broad availability of HBV vaccines in sub-Saharan Africa.¹⁶

Overall prevalence of syphilis/treponemal antibodies was also high (11.4%, 95% CI 8.9–13.7), and increased with older age, reaching 20.7% in men above the age of 40 years. Published data on the prevalence of syphilis among HIV patients in Cameroon are scarce, but our findings suggest a lower prevalence than reported from HIV-infected pregnant women in the capital Yaoundé (35.9%).¹⁷ Advantages of rapid tests for syphilis have been demonstrated by several studies and routine screening for syphilis including serological tests is currently recommended by WHO in chronic HIV care.¹² Serological testing is particularly important given the paucity of symptoms of early stage syphilis and the clinical consequences of undiagnosed late stage disease. Ideally, non-treponemal tests such as the Venereal Diseases Research Laboratory (VDRL) test should be used to confirm disease activity. However, this is not always feasible because of additional resources required. In addition, once diagnosed, syphilis can effectively be treated with penicillin, which is readily available in resource-poor countries. In antenatal HIV settings routine syphilis screening was shown to be cost-effective, although such analyses are largely missing for other populations at risk.¹⁸ If untreated, syphilis infection may cause long-term complications and facilitate the spread of HIV.⁶

The observed differences between men and women in HBV and syphilis prevalence could be explained by varying susceptibility and transmission, and seroconversion dynamics in the case of HBV. In a previous study among Cameroonian school children, it was shown that males and females were equally exposed to HBV, but females had a lower frequency of HBsAg positivity, possibly indicating a higher seroconversion rate in females.¹⁹ Additionally, in our study men initiating treatment were more often WHO stage III/IV and a preselection of sicker male patients seeking health care could also contribute to the higher observed prevalences of HBV and syphilis.

There were several limitations of this study. Firstly, data were collected in one centre only and validity for other centres is limited. However, the previously reported high regional variations of the frequency of co-infections underline the need for epidemiological studies to guide screening and treatment policies.

Secondly, presence of HBsAg could indicate a chronic carrier state rather than an active disease and could further miss cases of occult HBV disease. In previous studies, 78–100% of patients with positive HBsAg were found to have detectable HBV DNA.⁹ Thus, it is unlikely that the prevalence of HBV co-infection was significantly over-estimated in this study.

Likewise, the presence of treponemal antibodies could represent past or latent syphilis disease and should prompt a VDRL test to assess disease activity, which was not done here. In addition, the possibility of other non-venereal treponemal diseases such as yaws and bejel, which could also result in positive treponemal antibody tests, was not investigated. However, we considered these diseases to be rare in north-western Cameroon.

In conclusion, this study documents the high burden of syphilis and HBV co-infections in HIV infected patients in rural Cameroon and underlines the significance of those co-infections in Central Africa. The prevalence of HBV infection is particularly high in middle-aged male patients and warrants clinical suspicion. HBV screening strategies based on liver enzyme elevations, as often used in clinical practice, are not suitable. Routine serological screening for HBV and syphilis should be considered in HIV-infected high-prevalence populations in order to facilitate adequate clinical management, to avoid complications and to curb the spread of these diseases.

References

UNAIDS Report on the Global AIDS Epidemic 2010. See: http://www.unaids.org/globalreport/documents/20101123_GlobalReport_full_en.pdf (last accessed 29 May 2012)

Thio

, Seaberg

, Skolasky

Jr , HIV-1, hepatitis B virus, and risk of liver-related mortality in the Multicenter Cohort Study (MACS). Lancet 2002;360:1921–6

Livry

, Binquet

, Sgro

, Acute liver enzyme elevations in HIV-1-infected patients. HIV Clin Trials 2003;4:400–10

Puoti

, Torti

, Ripamonti

, Severe hepatotoxicity during combination antiretroviral treatment: incidence, liver histology, and outcome. J Acquir Immune Defic Syndr 2003;32:259–67

Benhamou

, Bochet

, Thibault

, Long-term incidence of hepatitis B virus resistance to lamivudine in human immunodeficiency virus-infected patients. Hepatology 1999;30:1302–6

Zetola

, Klausner

. Syphilis and HIV infection: an update. Clin Infect Dis 2007;44:1222–8

Buchacz

, Patel

, Taylor

, Syphilis increases HIV viral load and decreases CD4 cell counts in HIV-infected patients with new syphilis infections. AIDS 2004;18:2075–9

Lukhwareni

, Burnett

, Selabe

, Mzileni

, Mphahlele

. Increased detection of HBV DNA in HBsAg-positive and HBsAg-negative South African HIV/AIDS patients enrolling for highly active antiretroviral therapy at a Tertiary Hospital. J Med Virol 2009;81:406–12

Laurent

, Bourgeois

, Mpoudi-Ngole

, High rates of active hepatitis B and C co-infections in HIV-1 infected Cameroonian adults initiating antiretroviral therapy. HIV Med 2010;11:85–9

10.

Rouet

, Chaix

, Inwoley

, HBV and HCV prevalence and viraemia in HIV-positive and HIV-negative pregnant women in Abidjan, Cote d'Ivoire: the ANRS 1236 study. J Med Virol 2004;74:34–40

11.

Hoffmann

, Charalambous

, Martin

, Hepatitis B virus infection and response to antiretroviral therapy (ART) in a South African ART program. Clin Infect Dis 2008;47:1479–85

12.

WHO. Antiretroviral therapy for HIV infection in adults and adolescents. Recommendations for a public health approach: 2010 revision. See http://whqlibdoc.who.int/publications/2010/9789241599764_eng.pdf (last accessed 29 May 2012)

13.

Liaw

, Sung

, Chow

, Lamivudine for patients with chronic hepatitis B and advanced liver disease. N Engl J Med 2004;351: 1521–31

14.

Dienstag

, Goldin

, Heathcote

, Histological outcome during long-term lamivudine therapy. Gastroenterology 2003;124:105–17

15.

Snow-Lampart

, Chappell

, Curtis

, No resistance to tenofovir disoproxil fumarate detected after up to 144 weeks of therapy in patients monoinfected with chronic hepatitis B virus. Hepatology 18 November 2010

16.

WHO. Immunization Profile – Cameroon. 2011. See: http://apps.who.int/immunization_monitoring/en/globalsummary/countryprofileresult.cfm (last accessed 25 June 2012)

17.

Mbu

, Kongnyuy

, Mbopi-Keou

, Tonye

, Nana

, Leke

. Gynaecological morbidity among HIV positive pregnant women in Cameroon. Reprod health 2008;5:3

18.

Rydzak

, Goldie

. Cost-effectiveness of rapid point-of-care prenatal syphilis screening in sub-Saharan Africa. Sex Transm Dis 2008;35:775–84

19.

Chiaramonte

, Stroffolini

, Ngatchu

, Hepatitis B virus infection in Cameroon: a seroepidemiological survey in city school children. J Med Virol 1991;33:95–9