Study selection and baseline characteristics
Based on Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and strategy, a total of 2,321 studies were retrieved from the searched databases; 587 from PubMed, 544 from Scopus, 15 from African Journal Online (AJOL), 693 from Google Scholar, 67 from Cochrane, 103 from Willy library and 312 from EMBASE. After removal of duplicates and screening all records, 60 studies meeting the previously described eligibility criteria were used in the meta-analysis. The article number and reason for exclusion in each screening steps are shown in Fig 1.
Overall, a sample of three thousand two hundred eighty-two (3,283) participants was pooled from 60 eligible studies for inclusion in our data synthesis. By region, 10 studies (27)(28)(29)(30)(31)(32)(33)(34)(35)(36) were from east Africa with a pooled sample size of 531, 7 studies (37)(38)(39)(40)(41)(42)(43) from north Africa with a pooled sample size of 621, 12 studies (44)(45)(46)(47)(48)(49)(50)(51)(52)(53)(54)(55) from southern Africa with a pooled sample size of 483 and 31 studies (56)(57)(58)(59)(60) (61)(62)(63)(64)(65)(66)(67)(68)(69)(70)(71)(72)(73)(74)(75)(76)(77)(78)(79)(80)(81)(82)(83)(84)(85) (86) from west Africa with a pooled sample size of 1,648. By country, 23 countries out of the 54 African Union member countries (42.6%) had eligible studies for inclusion in the meta-analysis. One study was from each of the following countries; Cape Verde(84), Angola(47), Eretria(35), Guinea Bissau(63), Libya(38), Malawi(46), Mozambique(51), Rwanda(30) and Uganda(27) with sample sizes 96, 30, 122, 26, 60, 20, 9, 45 and 93 respectively.
Table 3: Characteristics of eligible studies.
First author, Year
|
Study design
|
Country
|
Region
|
Genotypes (n)
|
Sample
|
Patient type
|
Genotyping Method
|
de Pina-Araujo et al., 2018
|
Cross-sectional
|
Cape Verde
|
West Africa
|
D (1), A (75), E (19)
|
95
|
Community based
|
Sequencing
|
Fujiwara et al., 2005
|
Cross-sectional
|
Benin
|
West Africa
|
E (20), A (1)
|
21
|
Blood donors
|
RFLP
|
De Paschale et al., 2014
|
Prospective cohort
|
Benin
|
West Africa
|
E (19)
|
19
|
ANC women
|
INNO-LiPA
|
Diarra et al., 2018
|
Cross-sectional
|
Burkina Faso
|
West Africa
|
A (4), E (17)
|
21
|
Occult
HBV
|
Sequencing
|
Compaore et al., 2016
|
Cross-sectional
|
Burkina Faso
|
West Africa
|
E (120)
|
120
|
HIV+ Cohort
|
Multiplex PCR
|
Candotti et al et al., 2016
|
Prospective cohort
|
Burkina Faso
|
West Africa
|
E (71), A (28)
|
99
|
Blood Donors
|
Sequencing
|
Lago et al., 2014
|
Cross-sectional
|
Angola
|
South Africa
|
E (30)
|
30
|
Staff and visitors of a private hospital
|
Sequencing
|
Anderson et al., 2018
|
Cross-sectional
|
Botswana
|
South Africa
|
A (12), D (12), E (1)
|
25
|
HIV+ Cohort
|
Sequencing
|
Choga et al., 2018
|
Cross-sectional
|
Botswana
|
South Africa
|
A1 (13), D3 (21), D2 (1) NT (1)
|
36
|
Blood donors
|
Sequencing
|
Matthews et al., 2015
|
Cross-sectional
|
Botswana
|
South Africa
|
A (14), D (2)
|
16
|
HIV+ Cohort
|
Sequencing
|
Angounda et al., 2016
|
Cross-sectional
|
DRC
|
East Africa
|
A (24), E (58)
|
82
|
CHBV
|
Sequencing
|
Shindano et al., 2018
|
Cross-sectional
|
DRC
|
East Africa
|
A (40), E (1)
|
41
|
Hospital attendees
|
Sequencing
|
Elmaghloub et al., 2017
|
Cross-sectional
|
Egypt
|
North Africa
|
D (3), D/E (4), E (7)
|
14
|
HCWs
|
Sequencing
|
Iman et al., 2010
|
Cross-sectional
|
Egypt
|
North Africa
|
D (87), D/F (17)
|
100
|
HBV patients
|
INNO-LiPA
|
Zekri et al., 2007
|
Cross-sectional
|
Egypt
|
North Africa
|
A (7), B (18), C (6), D (26), A/D (5), C/D (2), B/D (2), B/C (2) ND (2)
|
70
|
Pediatric HCC cohort
|
PCR and RFLP
|
Archampong et al., 2017
|
Cross-sectional
|
Ghana
|
West Africa
|
E (58), A4 (3), A1 (1), D8 (1)
|
63
|
HIV-HBV co-infected
|
Sequencing
|
Ampah et al et al., 2016
|
Prospective cohort
|
Ghana
|
West Africa
|
E (52)
|
52
|
Randomized volunteers
|
Sequencing
|
Boyce et al et al., 2017
|
Case report
|
Ghana
|
West Africa
|
D/E (3)
|
3
|
HIV-HBV co-infected
|
Sequencing
|
Candotti et al., 2006
|
Cross-sectional
|
Ghana
|
West Africa
|
A (10), D (3), E (87), A/E (1)
|
101
|
Blood donors
|
INNO-LiPA
|
Candotti et al et al., 2007
|
Cross-sectional
|
Ghana
|
West Africa
|
E (69), E (1)
|
70
|
Mothers and their neonates
|
Sequencing
|
Geretti et al et al., 2010
|
Cross-sectional
|
Ghana
|
West Africa
|
E (82), A (4)
|
86
|
HIV+
|
Sequencing
|
Huy et al et al., 2006
|
Cross-sectional
|
Ghana
|
West Africa
|
E (12)
|
12
|
Blood donors
|
Sequencing
|
Dongdem et al., 2016
|
Cross-sectional
|
Ghana
|
West Africa
|
A (8), D (3), E (47)
|
58
|
CHBV
|
RFLP
|
Honge et al., 2014
|
Cross-sectional
|
Guinea-Bissau
|
West Africa
|
E (25), D (1)
|
26
|
HIV+
|
|
Boyd et al et al., 2016
|
Prospective cohort
|
Ivory Coast
|
West Africa
|
E (98), A (2)
|
100
|
HBV-HIV co-infected
|
Not specified
|
Anders et al., 2016
|
Prospective cohort
|
Ivory Coast
|
West Africa
|
A (1), E (92)
|
93
|
HBV-HIV co-infected
|
PCR based methods
|
Suzuki et al., 2003
|
Cross-sectional
|
Ivory Coast
|
West Africa
|
A (3), D (3), E (42)
|
48
|
HBV carriers
|
Sequencing
|
Lawson-Ananissoh et al., 2017
|
Cross-section
|
Ivory Coast
|
West Africa
|
A (6), E (27)
|
33
|
CHBV patients
|
INNO-LiPA
|
Day et al., 2013
|
Cross-sectional
|
Kenya
|
East Africa
|
A (10)
|
10
|
HIV+ women on ART
|
Sequencing
|
Mabeya et al., 2017
|
Cross-sectional
|
Kenya
|
East Africa
|
A (11)
|
11
|
HIV+ patients
|
Sequencing
|
Mwangi et al., 2009
|
Cross-sectional
|
Kenya
|
East Africa
|
A (46), E (4), D (2)
|
52
|
Blood donors
|
Sequencing
|
Ochwoto et al., 2013
|
Cross-sectional
|
Kenya
|
East Africa
|
A (38), D (3), D/E (1)
|
42
|
Liver disease patients
|
Sequencing
|
Webale et al., 2015
|
Cross-sectional
|
Kenya
|
East Africa
|
A (33)
|
33
|
HIV1+ and un-HIV infected
|
Sequencing
|
Salem MA et al., 2012
|
Cross-sectional
|
Libya
|
North Africa
|
A (1), D (54), E (1), D/E (4)
|
60
|
HBV infected
|
INNO-LiPA
|
Traore et al et al., 2015
|
Cohort
|
Mali
|
West Africa
|
E (82), D/E (5), D4 (1), A3 (2)
|
90
|
HBV infected and liver disease patients
|
Sequencing
|
Cella et al et al., 2017
|
Cross-sectional
|
Mali
|
West Africa
|
E (16)
|
16
|
HBV infected
|
Sequencing
|
Sugauchi et al., 2003
|
Cross-sectional
|
Malawi
|
South Africa
|
A (20)
|
20
|
HBV chronic carriers
|
Sequencing
|
Mabunda et al., 2020
|
Cross-sectional
|
Mozambique
|
South Africa
|
A1 (8), E (1)
|
9
|
Occult HBV among blood donors
|
Sequencing
|
Chekaraou et al et al., 2010
|
Cross-sectional
|
Niger
|
West Africa
|
E (20), D/E (4)
|
24
|
Blood donors
|
Sequencing
|
Brah et al., 2016
|
Cross-sectional
|
Niger
|
West Africa
|
E (21), D/E (1), A3/E (1)
|
23
|
HBV infected
|
Sequencing
|
Ahmad et al., 2019
|
Cross-sectional
|
Nigeria
|
West Africa
|
A (2), B (1), E/B (82), A/B/C/E (22), E (21), E/B/A (7), E/B/C (2), D/B/A (1)
|
138
|
HBV carriers
|
Nested PCR
|
Ayodele et al., 2019
|
Cross-sectional
|
Nigeria
|
West Africa
|
B (5), E (10)
|
15
|
HBV-HIV co-infected
|
Sequencing
|
Faleye et al., 2015
|
Cross-sectional
|
Nigeria
|
West Africa
|
E (19), NT (3)
|
22
|
ANC mothers
|
Sequencing
|
Forbi et al., 2010
|
Cross-sectional
|
Nigeria
|
West Africa
|
E (53), A (2)
|
55
|
ANC mothers and HIV+
|
Sequencing
|
Opaleye et al et al., 2016
|
Cross-sectional
|
Nigeria
|
West Africa
|
E (17)
|
17
|
CHBV
|
Sequencing
|
Oyinloye et al., 2021
|
Cross-sectional
|
Nigeria
|
West Africa
|
B/E (9)
|
9
|
CHBV
|
Nested PCR
|
Hubschen et al., 2009
|
Cross-sectional
|
Rwanda
|
East Africa
|
A (30), B (1), C (4), D (10)
|
45
|
HIV+ women
|
Sequencing
|
Maylin et al et al., 2015
|
Cohort
|
Senegal
|
West Africa
|
A (22), E (65)
|
87
|
CHBV
|
Sequencing
|
Vray et al., 2006
|
Cross-sectional
|
Senegal
|
West Africa
|
A (9), E (23)
|
32
|
Blood donors
|
Affymetrix system
|
Bowyer et al., 1997
|
Cross-sectional
|
South Africa
|
South Africa
|
A (24), D (3), B (1), C (1)
|
29
|
CHBV and Acute HBV
|
Sequencing
|
Kew et al., 2005
|
Cross-sectional
|
South Africa
|
South Africa
|
A (172), D (35), E (3) ND (12), A1 (146)
|
222
|
HCC patients
|
PCR and RFLP
|
Makondo et al., 2012
|
Cross-sectional
|
South Africa
|
South Africa
|
A (28), D (1)
|
29
|
|
Sequencing
|
Maponga et al. 2020
|
Cross-sectional
|
South Africa
|
South Africa
|
A (34), D (7), E (1)
|
42
|
HCC patients
|
Real time PCR
|
Selabe et al., 2009
|
Cross-sectional
|
South Africa
|
South Africa
|
A (10), B (4), C (2), D (1)
|
17
|
CHBV
|
Sequencing
|
Gededzha et al., 2016
|
Cross-sectional
|
South Africa
|
South Africa
|
A (5), D (3)
|
8
|
HBV-HIV co-infected and uninfected
|
Sequencing
|
Ayed et al., 2006
|
Cross-sectional
|
Tunisia
|
North Africa
|
A (1), B (1), C (3), D (139), Mixed (20)
|
164
|
CHBV
|
INNO-LiPA
|
Bahri et al., 2006
|
Cross-sectional
|
Tunisia
|
North Africa
|
A (7), D (66), E (6), NT (4)
|
83
|
CHBV
|
RFLP
|
Hannachi et al., 2010
|
Cross-sectional
|
Tunisia
|
North Africa
|
D (125), A (5)
|
130
|
CHBV
|
TSP-PCR
|
Hamida et al., 2021
|
Cross-section
|
Eretria
|
East Africa
|
A (3), B (1), C (21), D (26), E (19), A/D (6), A/E (1), B/E (1), C/D (16), C/E (13), D/E (5), A/D/C (1), C/D/E (9)
|
122
|
Patients with liver disease
|
Multiplex-nested PCR
|
Zirabamuzale et al., 2016
|
Cross-sectional
|
Uganda
|
East Africa
|
A (16), D (47), E (1), A/D (6), A/D/E (3), A/D/E/G (1), A/E (9), B/C/D (1), D/E/G (1), D/E (2), D/G (3), Intermediate (3)
|
93
|
Delinked stored samples
|
INNO-LiPA
|
Two studies were from each of the following countries; Benin(86)(69), DRC(34)(32), Mali(72)(77), Niger(66)(82) and Senegal(56)(71) with pooled sample sizes of 30, 123, 106, 47 and 119 respectively. In addition, three studies each were reported from Botswana(48)(50)(49), Burkina Faso(78)(59)(74), Egypt(40)(39)(43) and Tunisia(42)(41)(37) with pooled sample sizes of 77, 240, 184 and 377 respectively. Finally, four studies with a pooled sample size of 274 were from Ivory Coast(80)(58)(61)(61), five studies with a sample size of 148 were reported in Kenya(31)(28)(33)(29)(36), six studies were conducted from each of Nigeria (75)(68)(81)(70)(79)(60)and South Africa (53)(87)(54)(44)(52)(55)with sample sizes of 256 and 347 respectively and, eight studies from Ghana(65)(67)(64)(57)(85) (76)(83)(73)with a sample size of 445. By year of publication, 27(45%) of the studies that were eligible for inclusion in our systematic review and meta-analysis were published between 2016 to 2021, 13(21.67%) between 2011 to 2015, 15(25%) between 2006 to 2010 and 5(8.3%) between 1997 to 2005. Regarding the genotyping method used, 40(66.67%) of the studies used sequencing, 8(13.3%) INNO LiPA, 7(11.67%) RFLP, 2(3.33%) real time PCR while 1(1.67%) study for each of nested PCR, Affymetrix system and Multiplex PCR (Table 3, Figure 2).
Relative prevalence of the hepatitis B genotypes in Africa
Our systematic review and meta-analysis have reported genotypes A-E as the circulating genotypes on the African continent. Overall, genotype E had the highest pooled prevalence of 71.2% from 45 eligible significantly higher than all the other genotypes (p<0.001, 95% CI= [56.512 to 83.943]). This was followed by genotype A from 42 eligible studies at a pooled prevalence of 40.581%; 95% CI= [28.407 to 53.373] and genotype D with a pooled prevalence of 24.942%; 95% CI= [13.547 to 38.459] from 25 studies. In contrast, genotypes B and C had the lowest prevalence of 19.739%; 95%CI= [1.923 to 49.624] from 8 studies and 10.674%; 95% CI= [4.716 to 18.665] from 6 studies respectively. For all the analyses, the heterogeneity remained high (I2>79%, p<0.05) and the random effect model (REM) was used to pool the genotype prevalence (Table 4, Figures 3 and 4).
Regional prevalence of the hepatitis B virus genotypes
When we disaggregated our data by regions, all genotypes, except genotype C in west Africa, were represented in all the regions with sufficient number of eligible studies for meta-analysis. However, there was a disproportionate distribution of the genotypes in these regions. In eastern Africa genotype A posted the highest pooled prevalence of 72.791%; 95%CI= [41.7 to 94.86] from 10 studies significantly higher than the prevalence of all other genotypes (p<0.0001). The heterogeneity of the included studies for the analysis was high (I2= 98.17%, 95%CI= [97.55 to 98.63], p<0.0001) and the random effect model was used for the analysis.
Table 4: Meta-analysis of the prevalence of the hepatitis B virus genotypes in Africa
Sub-groups
|
|
|
Analysis of HBV prevalence
|
|
Analysis of heterogeneity
|
|
|
Genotypes
|
No
|
Sample size
|
Prevalence %(95%CI)
|
P value
|
I2% (95%CI)
|
P het
|
Model
|
Genotype E
|
45
|
2800
|
71.2 (56.512 to 83.943) Ref
|
|
98.54 (98.35 to 98.71)
|
< 0.0001**
|
Random
|
Genotype A
|
42
|
2729
|
40.581 (28.407 to 53.373)
|
< 0.0001
|
97.84 (97.51 to 98.13)
|
< 0.0001**
|
Random
|
Genotype B
|
8
|
604
|
19.739 (1.923 to 49.624)
|
< 0.0001
|
98.34 (97.72 to 98.79)
|
< 0.0001**
|
Random
|
Genotype C
|
6
|
556
|
10.674 (4.716 to 18.665)
|
< 0.0001
|
85.10 (69.43 to 92.74)
|
< 0.0001**
|
Random
|
Genotype D
|
25
|
1946
|
24.942 (13.547 to 38.459)
|
< 0.0001
|
97.64 (97.18 to 98.02)
|
< 0.0001**
|
Random
|
Similarly, in southern Africa, our meta-analysis reported genotype A as the most highly circulating genotype with a pooled prevalence of 79.441%; 95% CI= [68.8 to 88.34] from 9 eligible studies significantly higher than the prevalence of any other genotype (p<0.0001). Statistical heterogeneity was observed among the included studies for the analysis (I2=77.94% 95% CI= [58.26 to 88.35], p<0.0001). In contrast, our meta-analysis reported genotype E as the most prevalent in west Africa with a pooled prevalence of 87.44%; 95% CI= [78.16 to 94.43] from 9 eligible studies significantly higher than all other genotypes (p<0.0001).
Table 5: Meta-analysis of the regional prevalence of the hepatitis B virus genotypes
Sub-groups
|
|
|
Analysis of HBV prevalence
|
|
Analysis of heterogeneity
|
|
|
Genotype
|
No
|
Sample size
|
Prevalence %(95%CI)
|
P value
|
I2% (95%CI)
|
P het
|
Model
|
East Africa
|
|
|
|
|
|
|
|
Genotype A
|
10
|
531
|
72.791 (41.7 to 94.86) Ref
|
|
98.17 (97.55 to 98.63)
|
< 0.0001**
|
Random
|
Genotype B
|
2
|
167
|
1.632(0.307 to 4.889)
|
< 0.0001
|
0.00(0.00 to 0.00)
|
0.4183
|
Fixed
|
Genotype C
|
2
|
167
|
15.192(10.141 to 21.507)
|
< 0.0001
|
42.20 (0.00 to 0.00)
|
0.1884
|
Fixed
|
Genotype D
|
5
|
354
|
19.638(6.564 to 37.508)
|
< 0.0001
|
92.92(86.43 to 96.30)
|
< 0.0001
|
Random
|
Genotype E
|
5
|
390
|
15.440 (0.819 to 43.183)
|
< 0.0001
|
97.43 (95.85 to 98.41)
|
< 0.0001**
|
Random
|
South Africa
|
|
|
|
|
|
|
|
Genotype A
|
9
|
408
|
79.441 (68.8 to 88.34) Ref
|
|
77.94 (58.26 to 88.35)
|
< 0.0001**
|
Random
|
Genotype B
|
2
|
46
|
12.638(0.481 to 37.349)
|
< 0.0001
|
75.41(0.00 to 94.43)
|
0.0437
|
Random
|
Genotype C
|
2
|
46
|
7.720 (2.003 to 19.176)
|
< 0.0001
|
10.76 (0.00 to 0.00)
|
0.2898
|
Fixed
|
Genotype D
|
8
|
388
|
17.248 (10.042 to 25.917)
|
< 0.0001
|
67.36 (31.21 to 84.51)
|
0.0032
|
Random
|
Genotype E
|
5
|
328
|
26.662(1.717 to 66.518)
|
< 0.0001
|
97.32(95.66 to 98.35)
|
< 0.0001
|
Random
|
West Africa
|
|
|
|
|
|
|
|
Genotype E
|
30
|
1669
|
87.44 (78.16 to 94.43) Ref
|
|
96.19 (95.33 to 96.89)
|
< 0.0001**
|
Random
|
Genotype A
|
17
|
1238
|
13.283 (5.739 to 23.338)
|
< 0.0001
|
95.41 (93.86 to 96.56)
|
< 0.0001**
|
Random
|
Genotype B
|
2
|
153
|
11.832(2.578 to 57.786)
|
< 0.0001
|
93.64(79.48 to 98.03)
|
0.0001
|
Random
|
Genotype D
|
5
|
328
|
3.746 (1.979 to 6.387)
|
< 0.0001
|
0.00 (0.00 to 79.74)
|
0.4247
|
Fixed
|
North Africa
|
|
|
|
|
|
|
|
Genotype D
|
7
|
621
|
75.074 (58.01 to 88.85) Ref
|
|
95.07 (92.03 to 96.95)
|
< 0.0001**
|
Random
|
Genotype E
|
3
|
157
|
13.758(1.353 to 35.991)
|
< 0.0001
|
89.54(71.8 to 96.13)
|
0.0001**
|
Random
|
Genotype C
|
2
|
234
|
4.774(0.391 to 13.618)
|
< 0.0001
|
80.66(17.29 to 95.48)
|
0.0230
|
Random
|
Genotype B
|
2
|
234
|
9.459 (1.435 to 45.997)
|
< 0.0001
|
97.42 (93.53 to 98.97)
|
< 0.0001**
|
Random
|
Genotype A
|
5
|
507
|
4.510 (1.530 to 8.958)
|
< 0.0001
|
75.52 (39.89 to 90.03)
|
0.0026**
|
Random
|
For all the included studies, there was statistical heterogeneity (I2=96.19%; 95% CI= [95.33 to 96.89], p<0.0001) and the random effect model was used to pool the prevalence. In north Africa, genotype D reported the highest pooled prevalence of 75.074%; 95% CI= [58.01 to 88.85] from 7 eligible studies. The statistical heterogeneity was found among the included studies (I2=95.07% CI= [92.03 to 96.95], p<0.0001) and the random effect model was used to pool the prevalence (Table 5, Figure 5).
Furthermore, we compared the relative prevalence of the two emerging genotypes B and C on the African continent by region. Although the published data on these emerging genotypes is still limited, our results have shown that the prevalence of genotype B was significantly higher in southern Africa than eastern Africa (X2=11.299, p=0.0008, 95%CI= [3.472% to 23.5976]). In contrast, the prevalence of genotype C was significantly higher in eastern Africa than in northern Africa (X2=13.062, p=0.0003, 95%CI= [4.6197% to 16.968]) (Table 6).
Table 6: Meta-analysis of the relative prevalence of the emerging genotypes B and C on the African continent by region
Genotype
|
Region
|
No
|
Sample size
|
Prevalence %(95%CI)
|
X2
|
P value (95%CI)
|
B
|
South Africa
|
2
|
46
|
12.638(0.48 to 37.349) ref
|
|
|
|
East Africa
|
2
|
167
|
1.632(0.307 to 4.889)
|
11.299
|
0.0008(3.472% to 23.5976)
|
|
West Africa
|
2
|
153
|
11.832(2.578 to 57.786)
|
0.0216
|
0.8831(-8.3008% to 14.04)
|
|
North Africa
|
2
|
243
|
9.459 (1.435 to 45.997)
|
0.434
|
0.5100(-4.872% to 16.093)
|
C
|
East Africa
|
2
|
167
|
15.192(10.14 to 21.51) ref
|
|
|
|
South Africa
|
2
|
46
|
7.720 (2.003 to 19.176)
|
1.708
|
0.1913(-4.799% to 15.352)
|
|
North Africa
|
2
|
243
|
4.774(0.391 to 13.618)
|
13.062
|
0.0003(4.6197% to 16.968)
|
Prevalence of the sub-genotype by country.
The prevalence of sub-genotypes by country was evaluated by using Fisher’s exact test because of the scant published data on the prevalence of the sub-genotypes in Africa. Hence, we had insufficient data to pool the prevalence of the sub-genotypes because only one eligible study was obtained for each sub-genotype except for genotype A1 with four eligible studies. Besides, data on sub-genotypes was only available for published studies from four countries which included Botswana, Ghana, Mali and Mozambique. Overall, sub-genotype A1 was the most prevalent in proportions higher than any other sub-genotype (168/222, 75.7%) (p<0.05). (Table 7, Figure 3).
Table 7: Prevalence of the sub-genotypes by country
Sub-genotype
|
Country
|
Number of studies
|
Prevalence
|
Sample size (% prevalence)
|
Fisher’s test p value
|
A1
|
Ghana, Botswana, Mozambique, South Africa
|
4
|
168
|
222(75.7) Ref
|
|
D3
|
Botswana
|
1
|
21
|
36(58.3)
|
<0.0001*
|
A3
|
Mali
|
1
|
2
|
90(2.2)
|
<0.0001*
|
A4
|
Ghana
|
1
|
3
|
63(4.8)
|
<0.0001*
|
D2
|
Botswana
|
1
|
1
|
36(2.8)
|
<0.0001*
|
D4
|
Mali
|
1
|
1
|
90(1.1)
|
< 0.0001*
|
D8
|
Ghana
|
1
|
1
|
63(1.6)
|
<0.0001*
|
*P value of sub genotype prevalence with respect to the reference sub genotype < 0.05 statistically significant
Prevalence of genotype mixtures
We evaluated the prevalence of the genotype mixtures on the African continent by using the Fishers exact test. Data was available from only nine countries. The prevalence of the genotype mixture B/E reported from Nigeria and Eretria (93/269, 34.2%) was significantly higher than the prevalence of the other genotype mixture (p<0.05). Interestingly, recombinant genotype D/E was the most diverse reported by 9 studies in eight countries followed by A/E and A/D from 4 and 3 studies respectively (Table 7).
Table 7: Recombinant/mixed genotype prevalence by country
Genotype mixtures
|
Country
|
No
|
Prevalence
|
Sample size (% prevalence)
|
Fisher’s test p value
|
B/E
|
Eretria, Nigeria
|
3
|
93
|
269(34.2)
|
|
A/B/C/E
|
Nigeria
|
1
|
22
|
138(15.9)
|
0.003*
|
A/D/C
|
Eretria
|
1
|
1
|
122(0.82)
|
< 0.0001*
|
A/D/E
|
Uganda
|
1
|
3
|
93(3.2)
|
< 0.0001*
|
B/C/D
|
Uganda
|
1
|
1
|
93(1.1)
|
< 0.0001*
|
C/D
|
Egypt
|
1
|
2
|
70(2.9)
|
< 0.0001*
|
C/D/E
|
Eretria
|
1
|
9
|
122(7.4)
|
< 0.0001*
|
C/E
|
Eretria
|
1
|
13
|
122(10.7)
|
< 0.0001*
|
D/B/A
|
Nigeria
|
1
|
1
|
138(0.7)
|
< 0.0001*
|
D/E/G
|
Uganda
|
1
|
1
|
93(1.1)
|
< 0.0001*
|
D/F
|
Egypt
|
1
|
17
|
100(17.0)
|
0.012*
|
D/G
|
Uganda
|
1
|
3
|
93(3.2)
|
< 0.0001*
|
D/E
|
Egypt, Ghana, Kenya, Libya, Mali, Niger, Eretria, Uganda
|
9
|
29
|
471(6.2)
|
< 0.0001*
|
A/E
|
Uganda, Eretria, Ghana, Niger
|
4
|
12
|
339(3.5)
|
< 0.0001*
|
A/D
|
Egypt, Eretria, Uganda
|
3
|
14
|
285(4.9)
|
< 0.0001*
|
E/B/A
|
Nigeria
|
1
|
7
|
138(5.1)
|
< 0.0001*
|
E/B/C
|
Nigeria
|
1
|
2
|
138(1.5)
|
< 0.0001*
|
*P value of recombinant/mixed genotype prevalence with respect to the reference recombinant genotype < 0.05 statistically significant
Variation in the HBV dominant genotypes, other genotypes, sub-genotypes and recombinant genotypes.
We performed a meta-regression analysis on the variations in the dominant genotype over the past 25 years on the continent. Our results reported a general decline in the genotypes known to be dominating in the respective regions. However, the decline was only statistically significant in western Africa (r=0.36, p=0.049). Similarly, by meta-regression analysis of the variation in the prevalence of other genotypes other than the dominant genotype over the years, there was increase in the prevalence of these genotypes (Figure 6).
Non the less, the increase was not statistically significant (p>0.05) although a positive correlation was observed. Finally, meta-regression analysis of the variation in the prevalence of sub-genotypes/recombinant genotypes over the years only significantly increased in north Africa (r=0.9, p=0.036) (Figures 7 and 8).
Table 8: Role of intra and inter-continental migrations in the apparent distribution of HBV genotypes in Africa
Genotype(s)
|
Migrations
|
A and E in eastern and southern Africa
|
Ancient pastoral communities from eastern Africa to central-southern Africa(88)
Ancient farmers migrations from western Africa to southern-eastern Africa(89)
Migrations of labourers from West Africa to Congo to construct the Congo railway during the colonial times(90)
Bantu-speaking farmers migrations from western-central Africa around Niger and Benue rivers across central Africa to the southern-eastern Africa due to desertification pressure around 4,000 years ago(91)
Migrations from west Africa to south Africa for employment and education (92)
|
D in North Africa
|
Immigrations of French and Italians into north Africa during the transoceanic immigrations(93)
|
B and C in Africa
|
Trans-continental migrations from south east Asia and the far East into Eastern and southern Africa(92)
Colonialist immigrants from Europe to eastern and southern Africa during African colonization(94)(95).
Migrations of Asians to east Africa during the construction of the east Africa railway(96).
Business and education trips to and from Asia(96)
|