World Journal of AIDS, 2011, 1, 31-36
doi:10.4236/wja.2011.12006 Published Online June 2011 (
Copyright © 2011 SciRes. WJA
Immunovirological and Biochemical Changes in
Nigerian Patients with Hepatitis B Coinfection on
Antiretroviral Therapy
Jesse Abiodun Otegbayo1, Titilola Stella Akingbola2, Joshua Odunayo Akinyemi3,
Kayode Solomon Adedapo4, Georgina Njideka Odaibo5, Yetunde Adebisi Aken Óva2,
David Olufemi Olaleye5, Isaac Folorunsho Adewole7, Robert Murphy8, Phylis Kanki9
1Departments of Medicine, University of Ibadan, Ibadan, Nigeria; 2Department of Haematology, University of Ibadan, Ibadan, Nige-
ria; 3Department of Epidemiology and Biostatistics, University of Ibadan, Ibadan, Nigeria; 4Department of Chemical Pathology,
University of Ibadan, Ibadan, Nigeria; 5Department of Virology, University of Ibadan, Ibadan, Nigeria; 6Department of Haematology,
University of Ibadan, Ibadan, Nigeria; 7Department of Obstetrics and Gynaecology, University of Ibadan, Ibadan, Nigeria;
8Department of Infectious Diseases, North-western University Feinberg School of Medicine, Chicago, USA; 9Department of cancer
Biology, Harvard School of Public Health, Harvard University, Boston, USA.
Received April 3rd, 2011; Revised May 3rd, 2011; Accepted May 24th, 2011.
Hepatitis B virus (HBV) co-infection with HIV is high among Nigerians. Some studies have suggested impaired CD4
recovery among coinfected patients compared to the HIV mono infected. This retrospective study of treatment-naïve
HIV infected patients was aimed at determining the trend of changes in CD4+ counts, HIV-RNA, renal and liver func-
tion tests in response to combined antiretroviral therapy (CART). A questionnaire was utilised to extract clinical and
laboratory da ta of HBV co infected HIV /AIDS patients before treatment and at six, twelve and eighteen months of ther-
apy with CART. Findings were compared to those of HIV mono infected. Relevant statistical instruments were used to
analyse for comparisons of means of Log10 HIV viral load and CD4 count using SPSS package 15.0. All levels of sig-
nificance were at 5%. Two thousand five hundred and sixty two patients were analysed. Of these, 354 (13.8%) were
HBsAg positive. Majority (63.1%) were females. Most of the recruited patients were on combivir and nevirapine. The
median CD4 count fo r the HBsAg negative was 104 cells/(mm3 IQR 34-171) and it was significantly higher than those
of the positive (91 cells/mm3) (p < 0.05). ALT and AST were higher among HBsAg positives, while urea and creatinine
levels were similar. The median change in CD4 count from baseline and during the course of therapy were similar in
the two groups. S imilarly, virological responses were not different in the two g roups at the various time points. In con-
clusion no significant difference in the rate of CD4 recovery and HIV-RNA decline in among coinfected and monoin-
fected HIV patients at different stages of therapy.
Keywords: Hepatitis B, Coinfection, CD4, Viral Load, Antiretroviral, Nigeria
1. Introduction
Hepatitis B virus (HBV) is a common major endemic
infection in Nigeria [1,2] and its association with com-
mon major liver diseases is well established in our envi-
ronment [3,4] as in other parts of the world. HIV/AIDS
continue to receive significant attention and a sizeable
proportion of the health care budget because of the yet to
be conquered pandemic. Previous studies in Nigeria have
shown a significant hepatitis B and C virus co-infection
with HIV infection [5,6]. These studies however showed
that hepatitis C, either as mono-infection or as co-
infection with HIV infection is relatively low compared
with HBV. Many of the antiretroviral drugs in current
use are associated with hepatotoxicity among other
side-effects. Hepatotoxicity of antiretroviral drugs espe-
cially co-infected patients with hepatitis viruses is a ma-
jor concern, with increasing incidence of liver lidease [7].
Some studies have addressed the immunological (CD4+)
and viral (HIV-RNA) responses in HIV/AIDS patients on
combined antiretroviral therapy (CART). Micheloud et al.
in their study among HCV-coinfected paediatric patients
32 Immunovirological and Biochemical Changes in Nigerian Patients with Hepatitis B Coinfection on Antiretroviral Therapy
found no difference in clinical, immunological and viral
response to CART over a six year period, save for in-
creased transaminases among HCV coinfected [8]. A
meta-analysis had shown that patients with HIV-HCV
coinfection have less immune reconstitution, as deter-
mined by CD4 cell count after 48 weeks of CART, than
patients with HIV monoinfection [9] suggesting some
influence of hepatitis C on immunologic recovery in HIV
infection. Clinical progression of HIV-1 disease after
starting potent antiretroviral therapy is accelerated by
concomitant infection with HCV. De Luca et al. con-
cluded in their study that compared with patients without
HCV coinfection, coinfected patients showed impaired
CD4+ cell recovery, despite similar virologic response to
HIV-1 therapy [10]. Since HBV is the most important
co-infection among Nigerians, we embarked on a retro-
spective study of CART-naive HIV infected patients.
The objective of the study was to determine the trend of
changes in CD4+ counts, HIV-RNA load as well as bio-
chemical, renal and liver function tests in response to
2. Materials and Methods
The database of the PEPFAR clinic at the University
College Hospital, Ibadan, Nigeria was searched between
January 2006 and December 2007 to retrieve the pre-
sented data. The total number of patients on CART on
the PEPFAR database at the time of data extraction was
4354 as at December 2007. A questionnaire was de-
signed to extract the clinical data, liver function tests,
CD4+ counts and HIV viral load as well as the antiretro-
viral combination of patients who were HBV-coinfected
and compare with those who were mono infected with
HIV. Hepatitis B surface antigen was detected from sera
using commercially available 3rd generation ELISA
technique (ABBOT Murex, Germany), while HIV anti-
body was initially detected by ELISA test followed by
confirmation test using the Western blot hybridization.
HIV-RNA quantification was by reverse transcriptase
polymerase chain reaction (RT-PCR), with a lower de-
tection rate of 400 copies/ml. All patients with HIV were
tested for HBsAg at first visit, but HBeAg, anti-HBeAg
and HBV-DNA were not determined. The results were
noted at entry (baseline) before commencement of CART.
HIV-RNA, CD4+ counts, ALT, AST, urea and creatinine
were then determined at six, twelve and eighteen months
respectively. Only patients with HIV-1 and HIV-2 with
detectable HIV-RNA and CD4+ count of 250 cells/ul
were included in the study. Patients who were anti-HCV
positive were excluded. Also excluded were those that
had missing CD4, ALT and AST or were non-compliant
with medication and those who were not able to do CD4
count at the time of entry due to equipment breakdown.
2.1. Statistical Methods
Quantitative variables were assessed for normality using
visual inspection of box-plots and normal probability
plots. The Chi square test was used to assess differences
in frequencies for categorical variables. Levene’s statistic
was used to assess equality of variance between groups
before independent two-sample tests for comparisons of
means of Log10 HIV viral load at baseline. Given the sig-
nificant non-normality of CD4 count and liver function
test results, a non-parametric two-sample (Mann-
Whitney) test was performed to determine differences in
their distribution between HBsAg status. Six hundred
and sixteen cases with missen values (for follow up CD4
count/Viral load, AST/ALT) were excluded list wise
automatically by the statistical software. The generalized
linear model repeated measures analysis was performed
to assess the effect of HBsAg status on treatment over
time. Sex, education, baseline AST and ALT were ad-
justed for, and all statistical tests were carried out at the
5% significance level using SPSS 15.0. Ethical approval
was obtained from the Joint University College Hospital,
Ibadan/University of Ibadan Institutional Review Board.
3. Results
3.1. Socio-Demographic and Clinical
Two thousand five hundred and sixty two patients who
had sufficient retrievable data were included in the
analysis. Of the 2562 patients, 354 (13.8%) were HBsAg
positive. Majority (63.1%) were females. In terms of
educational attainment, secondary (37.9%) and primary
(29.7%) were the most common. A little above half
(58.1%) were married. Table 1 shows the socio- demo-
graphic and clinical characteristics of the patients ac-
cording to the HBsAg status. The mean ages were (36 +
8.9) years and (37.1 + 9.9) years for the HBsAg positive
and negative patients respectively (p > 0.05). There is a
higher proportion of males (48.3%) in the HBsAg posi-
tive group than in the HBsAg negative (35.1%). (p <
0.05). HBsAg positivity was also found to be higher
among those with primary (32.9%) and secondary
(41.7%) level education (p < 0.05).
The median CD4 count for the HBsAg negative was
104 cells/mm3 (IQR 34-171) and it was significantly
higher than those of the positive (91 cells/mm3) (p <
0.05). Similar pattern was observed for ALT and AST
with the HBsAg negative recording significantly lower
values. Urea and creatinine levels did not show any sig-
nificant difference between the two groups of patients.
3.2. Immunological and Virological Response
The median change in CD4 count from baseline time was
Copyright © 2011 SciRes. WJA
Immunovirological and Biochemical Changes in Nigerian Patients with Hepatitis B Coinfection on Antiretroviral Therapy
Copyright © 2011 SciRes. WJA
similar in the two groups (Figure 1). At 6 months, the
median change was 143 cells/mm3 and 148 cells/mm3 for
the negative and positive respectively. Although the me-
to adjust for some baseline characteristics that differs be-
tween the groups. Results from the model shows that nei-
ther of the covariates nor HBsAg status affects CD4 count
recovery. HBV status, education, ALT and AST all did
not affect CD4 count recovery (Table 2).
dian change at 18 months was higher for the positives
(218 cells/mm3) than negatives (206 cells/mm3), the dif-
ference did not attain statistical significance (p > 0.05). A The virological response was similar in the two groups
at the various time points (Figure 2). At 6 months, the generalized linear model was fitted for CD4 count change
Table 1. Clinical and demographic characteristics of subjects.
Characteristic HBV+ (n = 354) HBV– (n = 2208) p-value
Age in Yrs Mean(SD) 36.1 (8.9) 37.1 (9.9) 0.076
Female 51.7 64.9
Primary 32.9 29.2
Secondary 41.7 37.3
Tertiary 20.1 23.8
Marital status
Married 59.8 57.9
Separated 12.3 10.9
Single 13.7 13.5
Widowed 9.7 13.4
CD4 count, Median (IQR)
91.5 (32 - 152)
104 (34 - 171) 0.02
10 Viral Load, Mean (SD) 5.029 (0.85) 5.044 (0.91) 0.787
AST, Median (IQR) 49.0 (33.8 - 69.2) 40.0 (29.0 - 63.0) 0.005
ALT, Median (IQR) 31.0 (20 - 40) 27.0 (17 - 41) 0.004
Urea, Median (IQR) 19.0 (15.0 - 28.0) 19.0 (14.0 - 28.0) 0.731
Creatinine, Median (IQR) 0.8 (0.6 - 1.0) 0.8 (0.6 - 1.0) 0.281
6 month12 month18 month
Time period
Percentage with undetectable viral load
Figure 1. Proportion of patients by HBV infection status who achieved viral suppression.
Immunovirological and Biochemical Changes in Nigerian Patients with Hepatitis B Coinfection on Antiretroviral Therapy
percentage of patients who had achieved viral suppression
(HIV RNA level < 500 copies/mL) was 46.7% in the
negative group and 43.3% in the positive group (p > 0.05).
The percentages at 18 months were 71.0% and 68.8% for
the negatives and positives respectively. Repeated meas-
ures analysis showed that HBsAg status did not have any
effect on viral suppression over time (Table 2).
3.3. AST, ALT, Urea and Creatinine
At baseline, ALT and AST levels were higher for the
HBsAg positive patients (p < 0.05). The levels of these
parameters at 6, 12 and 18 months are shown in Table 3.
at the time points, these parameters were similar for the
two groups of patients. Majority of the monoinfected
Table 2. Summary results for GLM repeated measures model for CD4 count and Log10 RNA Viral Load over time.
CD4 count Log10 RNA Viral Load
Variable Model coefficient (β) 95% Confidence Interval
for β Model coefficient (β) 95% Confidence Interval for β
Baseline ALT 0.16 –0.60 - 0.92 0.003 –0.001 - 0.007
Baseline AST –0.06 –0.42 - 0.30 –0.001 –0.003 - 0.001
HBV 12.79 –33.08 - 58.65 0.063 –0.192 - 0.319
Gender 55.78 22.52 - 89.04* –0.002 –0.187 - 0.183
Education 10.35 –6.69 - 27.39 –0.002 –0.137 - 0.054
Marital Status –8.37 –25.52 - 4.7 –0.017 –0.09 - 0.056
Baseline CD4 0.87 0.69 - 1.00* –0.001 –0.002 - 003
Baseline Log10 RNA Viral
Load 25.85 3.94 - 47.75* –0.014 –0.135 - 0.106
HAART regimen 40.49 –17.42 - 98.39 –0.019 –0.353 - 0.314
* p < 0.05; Model R2 = 0.216
Change in CD4 Count from baseline (cells/cubic mm)
18 mths
12 mths
6 mths
Figure 2. Change in CD4 counts from baseline.
Table 3. Laboratory parameters of subjects over time by HBV status.
Characteristic 6 month 12 month 18 month
HBV+ HBV p-value HBV+ HBV p-valueHBV+ HBV p-value
ALT 30.0 (19 - 42) 27.0 (20 - 41) 0.689 26.0 (16.0 - 36.8)23.0 (15.8 - 39.3)0.692 16.0 (15.0 - 31.8) 23.5 (16.0 - 30.8)0.860
AST 41.0 (30 - 68) 35.0 (26 - 49) 0.131 NA NA NA NA
Urea 21.0 (14 - 26) 18.0 (14 - 24) 0.338 18.5 (13.5 -26.5)20 (13.0 - 24.5)0.335 20.0 (18.0 - 31.0) 21.5 (13.3 - 30.0)0.865
Creatinine 1.0 (0.8 - 1.1) 0.9 (0.7 - 1.0) 0.426 0.85 (0.7 - 1.0)0.9 (0.7 - 1.0)0.075 0.6 (0.5 - 1.0) 0.7 (0.5 - 1.1) 0.224
Median (IQR).
Copyright © 2011 SciRes. WJA
Immunovirological and Biochemical Changes in Nigerian Patients with Hepatitis B Coinfection on Antiretroviral Therapy 35
patients were on combivir and nevirapine, while the coin-
fected patients were mainly on truvada and nevirapine.
4. Discussion
This study has further shown a high HBV coinfection
rate among our patient cohorts as previously noted by us
and other authors in Africa [6,11,12] but higher than
values obtained in the Western world with lower inci-
dence of HBV [13,14]. The preponderance of the female
gender had been suggested to be due to the fact that
women may be more sensitive to changes in their health
and probably socioculturally conditioned to seek health
care more than men [6]. The significantly higher coinfec-
tion among men compared with women in spite of a
predominant female gender affected with HIV is likely a
reflection of HBV prevalence among the general popula-
tion which has been noted to be higher amongst men [15].
The lower prevalence of coinfection among the better
educated individuals generally could be due to the level
of awareness that may be associated with education and
therefore behaviours and practices that may reduce HBV
transmission [16]. It is however surprising that no dif-
ference in HBV prevalence was found across the differ-
ent marital status. It would have been expected, arguably
though, that divorcees and the unmarried would be more
exposed to transmission of HBV sexually. This phe-
nomenon might be confirming the well known fact that
HBV transmission is mainly horizontal in childhood
among Africans. Ola et al. however found out that HIV
monoinfection is commoner among married people
(70.4%), suggesting heterosexual transmission of HIV in
our environment as previously documented [16]. It
would appear that the immunological status of HBV
coinfection and monoinfection is similar, as evidenced
by the similar CD4 counts in HIV monoinfected and
HIV-coinfected at the start of CART. Similar observa-
tions were made Idoko et al. in the Northern part of Ni-
geria [17]. It is evident from our study that CART has no
influence on CD4 recovery regardless of HBV status
with time and duration of therapy. Idoko et al. however
found an association with Hepatitis B e antigen (HBeAg)
and CD4 T-cell recovery and HIV-RNA clearance [17].
Apart from the higher baseline values of transaminases
among HBV coinfected patients, no temporal difference
was observed in the normalization of ALT, urea and
creatinine among coinfected and monoinfected patients,
suggesting that HBV status did not significantly influ-
ence the immunovirological response to CART, renal
function and transaminase levels. The retrospective na-
ture of this study and our inability to determine the date
of acquisition of both HIV and HBV (acute or chronic),
as well as unavailability of HBV DNA are some of the
limitations of our study. The exclusion of subjects with
missen data may also affect the result as it is unclear
whether such cases were missing for reasons connected
with the study outcome.
[1] J. A. Otegbayo, F. A. Fasola and A. Abja, “Prevalence of
Hepatitis B Surface and E Antigens, Risk Factors for Vi-
ral Acquisition and Serum Transaminase among Blood
Donors in Ibadan, Nigeria,” Trop Gastroenterol, Vol. 24,
No. 4, 2003, pp. 196-197.
[2] N. D. Sirisena, M. O. Njoku, J. A. Idoko, et al., “Carriage
Rate of Hepatitis B Surface Antigen in an Urban Com-
munity in Jos Plateau State, Nigeria,” The Nigerian Post-
graduate Medical Journal, Vol. 9, No. 1, 2002, pp. 7-10.
[3] S. O. Ola, J. A. Otegbayo, G. N. Odaibo, O. D. Olaleye
and I. O. Olubuyide, “Serum Anti-HCV and Hepatitis B
Surface Antigenaemia in Nigerian Patients with Acute
Icteric Hepatitis,” West African Journal of Medicine, Vol.
21, Vol. 3, No. 2, 2002, pp. 215-217.
[4] I. O. Olubuyide, B. Aliyu, O. A. Olalelye, S. O. Ola, F.
Olawuyi, U. H. Malabu, S. O. Odemuyiwa, G. N. Odaibo
and G. C. Cook, “Hepatitis B and C Virus and Hepato-
cellular Carcinoma,” Transactions of the Royal Society of
Tropical Medicine, Vol. 91, No. 1, 1997, pp. 38-41.
[5] C. J. Uneke, O. Ogbu, P. U. Inyama, G. I. Anyanwu, M.
O. Njoku and J. H. Idoko, “Prevalence of Hepatitis-B
Surface Antigen among Blood Donors and Human Im-
munodeficiency Virus-Infected Patients in Jos, Nigeria,”
Memórias do Instituto Oswaldo Cruz, Vol. 100, No. 1,
2005, pp. 13-16.
[6] J. A. Otegbayo, B. O. Taiwo, T. S. Akingbola, G. N.
Odaibo, K. S. Adedapo, S. Penugonda, I. F. Adewole, D.
O. Olaleye, R. Murphy and P. Kanki, “Prevalence of
Hepatitis B and C Seropositivity in a Nigerian Cohort of
HIV-Infected Patients,” Annals of Hepatol, Vol. 7, No. 2,
2008, pp. 152-156.
[7] D. L. Thomas, “Growing Importance of Liver Disease in
HIV-Infected Persons,” Hepatology, Vol. 43, No. 2, 2006,
pp. S221-S229. doi:10.1002/hep.21033
[8] D. Micheloud, J. Jensen, J. M. Bellón, R. Gonzalez, M. J.
Mellado, M. L. Navarro, M. A. Muñoz-Fernández and S.
Resino, “Spanish Group of Pediatric HIV Infection.
Long-Term Response to Highly Active Antiretroviral
Therapy in Human Immunodeficiency Virus and Hepati-
tis C Virus Coinfected Children: 6 Years of Follow-up,”
The Pediatric Infectious Disease Journal, Vol. 26, No. 11,
2007, pp. 1061-1064.
[9] M. F. Miller, C. Haley, M. J. Koziel and C. F. Rowley,
“Impact of Hepatitis C Virus on Immune Restoration in
HIV-Infected Patients Who Start Highly Active Antiret-
roviral Therapy: A Meta-Analysis,” Clinical Infectious
Diseases, Vol. 41, No. 5, 2005, pp. 713-720.
[10] Luca, R. Bugarini, A. C. Lepri, M. Puoti, E. Girardi,
Copyright © 2011 SciRes. WJA
36 Immunovirological and Biochemical Changes in Nigerian Patients with Hepatitis B Coinfection on Antiretroviral Therapy
Copyright © 2011 SciRes. WJA
A. Antinori, A. Poggio, G. Pagano, G. Tositti, G. Cadeo,
A. Macor, M. Toti and A. D’Arminio Monforte, “Italian
Cohort Naive Antiretrovirals Study Group. Coinfection
with Hepatitis Viruses and Outcome of Initial Antiretro-
viral Regimens in Previously Naive HIV-Infected Sub-
jects,” Archives of Internal Medicine, Vol. 162, No. 18,
2002, pp. 2125-2132.
[11] J. C. Forbi, S. Gabadi, R. Alabi, H. O. Iperepolu, C. R.
Pam, P. E. Entonu and S. M. Agwale, “The Role of Triple
Infection with Hepatitis B Virus, Hepatitis C Virus, and
Human Immunodeficiency Virus (HIV) Type-1 on CD4+
Lymphocyte Levels in the Highly HIV Infected Popula-
tion of North-Central Nigeria,” Memórias do Instituto
Oswaldo Cruz, Vol. 102, No. 4, 2007, pp. 535-537.
[12] A. M. di Bisceglie, M. Maskew, D. Schulze, A. Reyneke,
L. McNamara and C. Firnhaber, “HIV-HBV Coinfection
among South African Patients Receiving Antiretroviral
Therapy,” Antiviral Therapy, Vol. 15, No. 3, 2010, pp.
499-503. doi:10.3851/IMP1494
[13] P. R. Spradling, J. T. Richardson, K. Buchacz, A. C.
Moorman and J. T. Brooks, “The HIV Outpatient Study
(HOPS) Investigators*. Prevalence of Chronic Hepatitis
B Virus Infection among Patients in the HIV Outpatient
Study, 1996-2007,” Journal of Viral Hepatitis, Vol. 17,
No. 12, 2010, pp. 879-886.
[14] C. L. Thio and B. Hepatitis, “Human Immunodeficiency
Virus Coinfection,” Hepatology, Vol. 49, No. 5, 2009, pp.
S138-145. doi:10.1002/hep.22883
[15] M. P. Adoga, S. D. Gyar, S. Pechulano, O. D. Bashayi, S.
E. Emiasegen, T. Zungwe, O. H. Iperepolu, C. Agupugo
and S. M. Agwale, “Hepatitis B Virus Infections in Ap-
parently Healthy Urban Nigerians: Data from Pre- Vac-
cination Tests,” The Journal of Infection in Developing
Countries, Vol. 4, No. 6, 2010, pp. 397-400.
[16] S. O. Ola, M. M. Ladipo, J. A. Otegbayo, G. N. Odaibo,
E. A. Bamgboye, O. G. Nwaorgu, S. W. Shokunbi and O.
D. Olaleye, “Demographic Factors in HIV Infected Pa-
tients Seen at UCH, Ibadan, Nigeria,” African Journal of
Medicine & Medical Sciences, Vol. 34, No. 3, 2005, pp.
[17] J. Idoko, S. Meloni, M. Muazu, L. Nimzing, B. Badung,
C. Hawkins, J. L. Sankalé, E. Ekong, R. Murphy, P.
Kanki and C. L. Thio, “Impact of Hepatitis B virus Infec-
tion on Human Immunodeficiency Virus Response to
Antiretroviral Therapy in Nigeria,” Clinical Infectious
Diseases, Vol. 49, No. 8, 2009, pp. 1268-1273.