Hepatitis B virus (HBV) represents an important health problem in the Maghreb countries, Algeria, Libya, Mauritania, Morocco and Tunisia, but no detailed synthesis of its epidemiology is available. In this review, we systematically searched for data about HBV in the Maghreb in peer-reviewed databases and included in our analysis works written in English and French, as well as institutional reports and regional conference meeting abstracts. We estimated national and regional prevalence of chronic HBV infection. In addition, we discuss molecular features of the viral strains circulating in the region.
Data analysis suggests that in the Maghreb region HBs antigen carriage concerns 1.8–4.9% of the population for an estimated number of 2.7 million persons. Genotype D, subtype D7, is predominant and mutations in the precore region of HBV genome are highly prevalent.
This epidemiological situation requires obviously widespread active interventions for prevention and control. In addition, anti-hepatitis B vaccination programme should be applied with the utmost discipline in the five countries considered in this present review. This systematic review will, hopefully, increase knowledge at disposal of Public Health authorities, enabling better resource allocation and healthcare delivery. The present synthesis intends to stimulate policies aiming at preventing the spread of HBV, keeping in mind that eradication of the virus from Maghrebi populations should be the ultimate objective of Public Health authorities.
Hepatitis B is an affection of global distribution that still represents worldwide major causes of severe liver disease, including cirrhosis and hepatocellular carcinoma (HCC). The people with chronic hepatitis B are estimated to 360 million resulting in 0.5–1 million deaths annually . According the World Health Organization, chronic hepatitis B prevalence is intermediate (HBsAg prevalence range from 2 to 7%) in the Maghreb region. The recurrent observation in various European countries that immigrants from the Maghreb region are more frequently infected with HBV than the general population illustrates this preoccupying situation [2-4]. The epidemiology of hepatitis B virus is not well defined in the Maghreb. Epidemiological data are, however, a mandatory starting point for the development of appropriate diagnostic and preventive measures. We aimed to systematically review publications related to HBV epidemiology, including their major modes of transmission as well as the molecular features of viral strains circulating in the Maghreb populations. Such systematic review will, hopefully, increase knowledge at disposal of Public Health authorities, enable better resource allocation or healthcare delivery and stimulate policies to prevent the spread of HBV.
Study design and search strategy
We undertook our review in line with PRISMA guidelines relevant to a descriptive review of this nature . Our procedures consisted of multiple stages of searches of the peer-reviewed published work until 20 June 2012. Beside documents available from PubMed, we included in our analysis works written in French, as well as data from institutional reports, regional meeting abstracts updated in April 2012. Our purpose was to produce a synthesis from the available corpus of data emanating from a region where investigators do not publish readily in English and to provide, therefore, genuinely new information to the reader. In addition, we e-mailed hepatitis experts in the Maghreb region. All reports selected were grouped according to the target population general population, blood donors (1st blood donation screening), pregnant women and the haemodialyzed patients (HD) for each country.
For data extraction and selection, we catalogued documents with Endnote. Two authors systematically screened search results. Other references were reviewed if the title or abstract suggested that the document had relevant information about the prevalence of HBV in Maghreb region. Data were regarded as eligible when the number, prevalence, viral strains in the Maghreb region, in a given country or a region were mentioned. For HBV, prevalence is based predominantly on serological testing for HBsAg.
We extracted information about study design (eligibility criteria, recruitment and enrolment dates, and recruitment methods and locations), participant characteristics (age range and sex), and hepatitis reports (number of participants tested, number and proportion of patients who tested positive for HBsAg). The same methodology was followed to extract molecular characteristics of HBV. The prevalence of HBV in the general population, blood donors, pregnant women and HD were estimated on the average prevalence, calculated and weighted by study size in the case of multiple estimates from comparable studies. Microsoft Excel software was used to calculate prevalence.
Chronic hepatitis B epidemiology in the Maghreb
The endemicity of chronic hepatitis B, influenced primarily by the age at which infection occurs, is known to vary greatly throughout the world. In Africa, the level of endemicity is generally high and increases from North to South with the Sahara representing a real epidemiological frontier with regard to HBV carriage. The data reported from the Maghreb are heterogeneous, and even occasionally inconsistent within the same country, plausibly reflecting diversity of environmental, socio-economic or cultural factors. However, it may be as well a mere consequence of inappropriately designed studies. Overall, in the Maghreb region, the estimated HBsAg prevalence range is 1.8–4.9% with a number 2.7 million persons persistently infected (Table 1).
Table 1. Estimates for the prevalence of hepatitis B (HBsAg) (%)
NA, non-available data.
In Algeria, to date, there is a conspicuous absence of figures estimating the overall prevalence of HBsAg in populations. Level of HBV endemicity is, however, suspected to be intermediate according to the WHO standards (2–7%). The first epidemiological survey, published in 1984, showed a prevalence of HBsAg carriers within a range of 1.8–2.8% . Regional variations were observed with higher levels of infection in Eastern and Saharian regions of Algeria. The authors observed that HBsAg positivity increases with age and low socio-economic status . A decade later, a survey conducted to assess hepatitis B seroprevalence in 1112 apparently healthy blood donors and 715 pregnant women in different regions in Algeria, detected HBsAg seropositivity in 3.6% and 1.6% of cases respectively . In 1998, the estimated national prevalence of HBsAg carrier state was 2.2% . The authors observed that for the majority of the cases, diagnosis was fortuitous resulting from a blood donation, a pre-operative evaluation, a premarital or a prenatal check-up. The number of intra-familial cases was very high suggesting an important horizontal transmission in Algerian populations .
In Libya, hepatitis B remains an important communicable disease because of an intermediate prevalence of the carriage, a strong incidence of new cases, accompanied by a high rate of hospitalization for liver diseases. A national population-based sero-epidemiological survey between October 2004 and March 2005 was conducted on more than 67 000 participants chosen to obtain a representative sampling of the general population from all parts of Libya. This survey reported an overall prevalence of 2.2% with 120 000–150 000 chronic HBsAg carriers . Interestingly, the prevalence of HBs carriage remains lower in Libyan blood donors (1.3%) underlining the efforts produced by the health system to recruit safe donors . In the country, HBV chronic infection is primarily related to male gender, skin scarification practices, unsafe injections, history of blood transfusion and familial history of hepatitis B. It is generally acknowledged that a majority of HBV infections are acquired horizontally [9, 11]. However, it was observed that infected pregnant women represent a significant source of viral spread in the community. A recent study conducted on 1500 pregnant women and 1500 cord blood samples of their neonates delivered at Tripoli Medical Centre showed a prevalence of 1.5% in pregnant women and 0.9% in neonates. All HBsAg-positive neonates were born from HBs Ag-positive mothers with a rate of maternal transmission of 61% . Universal new-born hepatitis B vaccination was introduced in 1993. The vaccine was also recommended for household contacts of HBV-infected individuals and healthcare workers. Finally, the prevalence of chronic hepatitis B in high-risk population, as prison inmates of western Libya, rises to 7% .
In Morocco, systematic nationwide screening of blood donors for HBsAg was introduced in 1985 and vaccination of neonates started in 1999. In absence of a national survey, it is, however, difficult to estimate precisely the overall prevalence rate of chronic hepatitis B. Overall, according to available data, the prevalence is considered as low to intermediate; ranging from 1 to 3% [14-18]. In our previous study, we reported that the rate of HBs antigen carriage in the general population was 1.5%, a prevalence consistent with that reported in healthcare workers (1%) [14-16, 19]. It appears that the modes of contamination in the Moroccan Kingdom are diverse either vertical, intra-familial, sexual or parenteral . In a recent study conducted to assess HBsAg prevalence on more than 16 000 sexually active subjects, the marker was detected in 1.7% of the subjects and an at-risk sexual behaviour was identified as the main cause of contamination (44% of cases) [14, 16]. Another study conducted in barbers and their clients showed a congruent HBsAg prevalence of 1.7–1.9%. In this study, the dominant risk factors for HBV carriage is older age, low education level, urban dwelling, married life and history of transfusion . In blood banks, an early XXIst century survey found that HBsAg prevalence reaches 2.5–2.8% . More recently, surveys were conducted to assess the prevalence of serological markers of hepatitis B at the Blood Transfusion Centre of Military Teaching Hospital Mohammed V in Rabat (n =19 800 donors), in the blood bank of Casablanca (n =169 600) or in the blood bank of a southern Moroccan hospital (Guelmim). The observed HBsAg prevalences were ranging from 0.8 to 1.3% [17, 21]. Authors concluded that from 1991 to 2010, HBV seropositivity underwent a significant decline in Morocco . Besides, studies conducted on high-risk populations showed a prevalence of HBsAg ranging from 36% in new-borns of HBV-chronically infected mothers to 6% in prison inmates , whereas in pregnant women, the prevalence of HBsAg was 1.3%, close from the levels reported in general population . An early study conducted at the sexually transmitted diseases (STD) clinic of the Pasteur Institute of Morocco showed a seroprevalence of 3.0% for HBsAg confirming the tight link between chronic infection and at-risk sexual practices . Recently, a retrospective study carried out at Department of Gastroenterology at the University Hospital of Rabat on 12 000 hospitalized individuals for any admission excluding liver cases between 1990 and 2004, produced a 15.8% prevalence of HBsAg . Finally, Boulaajaj et al. conducted a study on a group of haemodialyzed patients previously vaccinated against hepatitis B. A proportion of 96% of the subjects presents a titre of anti-HBs greater than 10 UI/ml, but the prevalence of HBsAg was 2% . Multivariate statistical analysis showed that a history of surgical treatment, blood transfusion or injection drug use were the main risk factors of chronic hepatitis B in Morocco .
Mauritania, in close contact with sub-Saharan Africa, has, without surprise, the highest prevalence of HBV carriage in the Maghreb region. Early studies assessing HBsAg seroprevalence were performed in 1992 on blood donors by the National Hospital laboratory. At that time, a very high prevalence of HBsAg was observed (20.3%) . At the same period, a serological survey was conducted in different ethnic groups of Mauritania. Authors concluded that poor hygienic conditions related to nomadic lifestyle, favour contamination with hepatitis B virus . Recently, other factors such as polygamy, low educational level and history of transfusion were correlated with a risk of HBV infection . A survey searching for different markers of HBV infection in 267 primary and secondary schoolchildren from two Mauritanian localities in the south and south-east of the country, produced figures as high as 17% HBsAg carriers . The rates of HBsAg in pregnant women was ranging from 11 to 16%  . More recently, another study conducted from October 2008 to December 2009 in 11 000 blood donors reported a 15% prevalence of HBsAg . These data, remarkably stable over three decades, confirmed the serious problem posed to Public Health by hepatitis B and should prompt a programme to combat it. An expanded programme of immunization including hepatitis B has been, fortunately, introduced in Mauritania in 2000. Finally, it is worth mentioning that most Mauritanian references are ancient and novel surveys should be launched to monitor the trends of hepatitis B epidemiology in the population.
For Tunisia, hepatitis B represents historically a major public health problem with its inherent high morbidity and mortality. Seroprevalence studies were, thus, conducted quite early and, as a consequence, it is in Tunisia that we get the most detailed knowledge about the trends affecting HBV endemicity. According to early works, prevalence of HBsAg varies throughout the country, ranging from 3 to 13% with higher prevalence in the south and central-western regions. A prevalence of the chronic carriage exceeding 15% in some villages was even observed [31-33]. A large study conducted in more than 33 000 healthy people found HBsAg in 6.5% of individuals . These prevalences positioned the country close to the upper threshold of intermediate endemicity. At the same period, the carriage rate of HBs antigen in blood donors was about 5% . At that time, the predominant modes of contamination in Tunisia were identified as vertical, intra-familial or sexual. Tunisian men, having plausibly more sex partners than the women, were shown to sustain higher HBV infection rates (current and/or past) than women [31-33]. Not surprisingly, the youngest population subset (under 20 years old) was consistently shown to be at higher risk of HBsAg carriage than the adult population [31-33]. The magnitude of vertical and perinatal transmission of HBV was addressed in a study conducted on a cohort of 2300 pregnant Tunisian women among whom 4% were HBsAg positive and 1.4% were previously vaccinated against hepatitis B . Vertical/perinatal transmission of HBV in the first 3 months of life occurred in only 0.4% of the 177 mothers and child pairs. In contrast, in the general population, HBsAg positivity which was 11% in infants under 5 years old, increased rapidly with age till 25 years of age and then more slowly in mature adulthood . Thus, it was concluded that vertical and perinatal transmission of HBV does not play a major role in Tunisia; contrasting with horizontal transmission, mainly in childhood and adolescence, that appears to be the major mode of infection. To address this preoccupying situation, the systematic nationwide screening of blood donors for HBsAg was introduced in 1985 and vaccination of neonates started in 1995. Lately, a population-based sero-epidemiological study enrolled around 10 000 volunteers in the governorates of Béja in the north and Tataouine in the south of Tunisia. It showed that the overall prevalence of HBsAg was 5.3% albeit with significant differences observed between the two governorates (4.2% in Béja and 5.6% in Tataouine; P =0.001) . At the individual level, the presence of a family member infected with HBV, scarification practices, injectable medication and male sex significantly increased the risk of HBsAg positivity . In Tunisia, the prevalence of HBsAg in high risk groups is not drastically different from that observed in the general population underlining the high standards of the medical practice in the country. Actually, the prevalence of chronic hepatitis B was only 8.4% and 7.1% in polytransfused and haemophiliacs respectively [37, 38].
HBV Genotypes/subtypes in the Maghreb
HBV has evolved in multiple genetic strains that are differentially distributed among human populations. HBsAg subtype is determined by amino acids residues at positions 122, 127, 134, 159 and 160 enabling virus allocation to one of the nine immunological subtypes: adw (adw2 and adw4), ayw (ayw1, ayw2, ayw3 and ayw4), adr (adrq+, adrq−) and ayr .
In 134 Moroccan patients with HBV, the majority of strains belonged to HBsAg subtype ayw2 (82.1%, n =110) followed by adw2 (10.4%, n =14), ayw3 (3%, n =4) and ayw4 (3%, n =4) . A similar distribution was found in Algerian HBV strains with the predominant ayw2 subtype (73%).
In Tunisia, there was little information regarding the distribution of immunological subtype. A study conducted by Borchani–Chabchoub and colleagues on five hepatitis B virus strains isolated from plasma samples of patients showed, however, that all sequences belonged to subtype ayw2 .
In Libya, no data about the distribution of HBV subtypes is available and a seminal study is still eagerly expected.
In Mauritania, a survey conducted in sera from 515 black and 499 white individuals living in 8 villages showed two main subtypes, ayw2 (34.7%) and ayw4 (63%). The subtypes ayw2 was more prevalent in North of the country and ayw4 in the South. Analysis of the subtype distribution in each village indicates that there is no relationship between HBsAg subtype and ethnic groups, but there is a correlation between HBsAg subtype and the geographical location of the villages . This result is in agreement with the reported predominance of subtype ayw2 in Mediterranean countries and ayw 4 in West Africa .
Based on a minimum divergence of 8% of the complete genome sequences, HBV has been classified in different genotypes consecutively identified as genotypes A-H [44-47]. Within some HBV genotypes, subgenotype diversity was also described, with a minimum genetic distance of 4%. Recently, a ninth genotype isolated in Laos  and in Vietnam  and tentatively termed ‘I’, though it is still subject to debate . Finally, a tenth genotype provisionally assigned to genotype ‘J’ was isolated from a Japanese patient .
In Algeria, little remains known about molecular diversity of HBV. The only study conducted on 75 chronic hepatitis B patients from north-east Algeria, showed that the genotype D was predominant (93%, n =70) followed by genotype A (5%, n =4) and E (1.3%, n =1). Moreover, according to the authors, Algerian strains clustered independently from other genotype D sequences, suggesting the possible emergence of a new subtype . This result seems in keeping with the data obtained on Moroccan patients [14, 40] (Figure 1). In Libya, data about molecular features of HBV are scarce. The only result available showed a predominance of genotype D like other countries in North Africa . In Mauritania, few information about the molecular characteristics of HBV, but genotypes prevailing were genotype D (53%, n =43), genotype E (35%, n =28) and genotype A (12%, n =10)  determined in 81 samples (Figure 1).
In Morocco, four surveys were published about the distribution of HBV genotypes reporting the predominance of genotype D [14, 40, 53, 54]. In addition, phylogenetic analysis based on pre-S/S sequences suggested that Moroccan HBV strains drifted on the Western margins of the genotype D distribution area to produce isolates differing subtly from other D strains . Later on, in 134 patients, we showed that among the genotype D strains (90%, n =120), the majority (70.8%, n =85) belonged to subgenotype D7, 25.8% to subgenotype D1 (n =31) and 0.9% to subgenotype D2 (n =1), whereas all genotype A (10%, n =14) strains belonged to subgenotype A2 .
In Tunisia, the diversity of HBV was assessed by four studies. The first study published in 2006 in 79 Tunisian patients with chronic HBV infection reported predominance of genotype D (80%, n =66) followed by genotype A (9%, n =7) and genotype E (8%, n =6). No significant difference was observed between genotypes with regard to the clinical status of infected patients . At the same period, Ayed and co-workers conduct another study on 164 patients confirming the predominance of genotype D (84.75%, n =139), as well as a marginal presence of genotypes A, B and C  (Figure 1). These various molecular studies were performed essentially in the northern part of the country. Another survey conducted on 130 HBV-infected patients originating from the central part of Tunisia reported an even higher prevalence of genotype D (96%, n =125) followed by genotype A (4%, n =5) . Two reports described HBV subtypes with a novel variant named D7 [57, 58]. Hannachi et al. concluded that subgenotype D7 is endemic to northern Africa and is progressively replaced by subgenotype D1 towards the East of the continent .
Precore and core promoter mutants of hepatitis B
During the course of chronic HBV infection, HBe antigen (HBeAg) is considered a marker of active viral replication. Seroconversion to anti-HBe is usually accompanied by a decreased HBV replication and often coincides with clinical remission of liver disease. In some patients, however, detectable serum levels of HBV DNA, persistently elevated alanine aminotransferase (ALT), as well as continued hepatic necrosis and inflammation, are still noticed after anti-HBe seroconversion. Most of these patients are infected with HBV variants that decrease or abolish the production of HBeAg. The search for a molecular basis to such anomaly led to the discovery of precore (PC, nt1814–1900) and the basal core promoter (BCP, nt 1613–1849) mutations that abolish or decrease HBeAg production [59-61]. A classical mutation, G1896A, within PC gene would introduce a stop codon at residue 28 (W→Stop), terminating the translation of the precore protein .
In Algeria, a single study, published in 2008, showed that HBV PC mutants were present in 87% of patients. BCP mutants were observed in 60% of cases (n =39). They were frequently characterized by the concomitance (80% of cases, n =51) of BCP double mutation and stop-codon mutation at nucleotide 1896 . No correlation was found between the presence of PC and BCP and viral load .
In Libya, the natural history of hepatitis B has not been completely studied in details and, therefore, the steps and rates leading from an HBe proficient to an HBe deficient strain are poorly described. El zouki and colleagues reported a high prevalence rate of HBeAg-negative/anti-HBe-positive in 150 chronic hepatitis B patients, which accounted for 80% (n =120). This study indicated that patients suffering from a HBeAg-negative chronic hepatitis B were older, display a higher frequency of family history of hepatitis and a higher prevalence of fibrosis and cirrhosis than HBeAg-positive patients .
In Mauritania, the only survey searching for different markers of HBV infection in 267 primary and secondary school children in two Mauritanian sites located at the southern part of the country. Data indicated a high prevalence rate of HBeAg-negative/anti-HBe-positive (71%, n =190) among chronically infected patients .
In Morocco, several reports published from 1993 to 2011 showed a constant predominance of PC stop codon G1896A mutation detected in 84–86% of chronic hepatitis B [14, 64]. More recently, other mutations were described in the BCP: an A to T substitution at nucleotide 1762 and a G to A substitution at nucleotide 1764 and G1613A [65, 66]. The BCP A1762T/G1764A double mutation that diminishes HBe Ag production was found in 22.9–24.4% of Moroccan patients [54, 67]. The higher prevalence of G1896A compared with A1762T/G1764A mutations may be attributed to the predominance of HBV genotype D in Moroccan patients [14, 67]. G1896A is restricted to genotypes that have T at nt 1858 (non-A genotypes), since its stabilizes the encapsidation signal, essential for viral replication, whereas CP A1762T/G1764A mutation occurred preferentially in genotypes with C1858 (genotype A). The presence of G1896A was associated with a decreased risk of advanced liver disease (AdLD) compared with the wild-type virus. In contrast, the prevalence of double mutation A1762T/G1764A increased significantly with the progression of disease from inactive carrier to AdLD state with high viral load and elevated ALT levels compared to wild-type virus, suggesting that BCP mutations determine, at least partially, disease severity .
In Tunisia, the frequency of HBV precore mutants is high. The team of the Institut Pasteur of Tunis observed that the rate of HBV mutants at precore region (position 1896) and/or in the basal core promoter were detected in 33 of HBeAg-negative patients was 57.5% (n = 19), . Ayed and co-workers confirmed this result in 73 HBeAg-negative patients (57.7%, n =42) . In another report, the premature precore stop codon at G1896A was twice as common in HBV/D (87/102, 85%) than in HBV/A (3/7, 43%) . In patients with HBV mutants, viral loads were either moderate (49%) or high (>2000UI/ml) (51%) . In addition, a study of HBV in Tunisian HBsAg-positive blood donors indicated that the basal core promoter mutations A1762T and G1764A are found in 86% of HBsAg-positive individuals .
Hepatitis D in Maghreb region
In the Maghreb region, data about the prevalence of hepatitis delta virus (HDV) infection in HBV carriers are scarce. In Morocco, for example, a single study performed two decades ago on 142 HBsAg carriers, 1.4% had anti-Delta antibody . It was concluded at that time that delta infection is not a problem in Morocco . In contrast, Algeria, Mauritania and Tunisia were countries where hepatitis Delta was an important issue for Public Health authorities. Actually, 20–30 years ago, HDV was found in 17–33% of Tunisian HBsAg carriers, in 6–20% of HBsAg-positive patients with chronic liver diseases in Algiers, whereas it was found in 20% of blood donors and 15% of pregnant women from Mauritania [27, 33, 68-70]. Its prevalence was more important in the Southern part of Tunisia in connection with high HBV endemicity and cultural practices (scarification, traditional circumcision). Current studies indicate that, at least for Tunisia, the prevalence of HDV decreased in the last decades to reach levels ranging from 3 to 10% of HBsAg carriers . The molecular characterization of HDV strains showed that genotype 1 was present in Tunisia  and genotypes 1 (90%) and 5 (10%) circulating strains in Mauritania [27, 68].
Overall, a significant effort should be made to clarify the current epidemiological status of hepatitis D in Algeria, Libya and Morocco.
Co-infection hepatitis B in HIV group in Maghreb region
As a result of the sharing of transmission routes, infection with HBV is common among patients with human immunodeficiency virus (HIV) infection. In these patients, chronic co-infection with HBV is associated with a significant excess of morbidity and mortality [72, 73]. As a consequence, liver disease, resulting from chronic hepatitis or other aetiologies, is nowadays one of the most frequent causes of death among HIV-infected people . The prevalence of co-infection with HBV is variable and depends on the route of HIV acquisition . In Algeria, the prevalence of HBsAg is 6% in HIV patients . In Libya and in Mauritania, there is no data regarding the prevalence of co-infection HBV/HIV. In Morocco, a study conducted among HIV-positive patients in Military hospital in Rabat detected a prevalence of 6.7% for HBsAg . In Tunisia, the first study to assess the co-infection rate of viral hepatitis B in HIV-positive patients reported a prevalence of 5.6% for HBsAg .
Co-infection with hepatitis C in the Maghreb region
Hepatitis C is affection of global distribution that still represents worldwide major causes of severe liver disease, including cirrhosis and hepatocellular carcinoma (HCC). The prevalence of anti-HCV antibodies in the Maghreb region varies between 0.4% and 1.9% . The prevalence HBsAg in hepatitis C group was estimated at 1.9% in Libya  and 5% in Morocco  and Tunisia . In Algeria and in Mauritania, there is no data regarding the prevalence of co-infection HBV/HCV.
HBV in hepatocellular carcinoma in the Maghreb
Chronic infection with HBV is still the most important causes of HCC in the world . The relative importance of HBV infection in HCC aetiology is known to vary greatly from one part of the world to another , and can change over time . However, despite intermediate levels of chronic hepatitis B in the general population (1–5%, except in Mauritania where HBV is high), HCC incidence in the Maghreb region is particularly low (1.3/105 ASR, according to Globocan 2008) ranging far below incidences observed in most European countries, where hepatitis viruses are far less prevalent. To explain this situation, we conducted in collaboration with the Institutes Pasteur of Algiers and Tunis, a multicentre case-control study in 164 patients with HCC. We observed that 60% of HCC patients were positive for anti-HCV and 18% for HBsAg. By opposition, the HBsAg prevalence in HCC was lower than in Middle−Eastern countries, like Saudi Arabia and Lebanon [85-87]. Thus, our data suggest that the low incidence of HCC in the region is may be linked to a weaker tumorigenic potential of circulating hepatitis viruses strains (genotype D7 vs. other D for HBV), to an absence of usual tumorigenic cofactors as aflatoxin B1 and also the life expectancy in each country. It would be of great interest to perform a similar study on Mauritanian (genotype A of HBV) to figure out whether changes in viral and environmental conditions are capable to induce relevant alterations in the molecular epidemiology of HCC and stimulate an increased HCC incidence.
Conclusions and needs for future research
Chronic Hepatitis B infection remains an important Public Health problem in the Maghreb, as in so many countries of the world. Based on published evidence presented above, striking epidemiological similarities exist between most Maghreb countries even if Mauritania stays apart because of its stronger links with Sub-Saharan Africa and its high endemicity of HBV. Nevertheless, in all countries, targeted interventions should be considered and notably the intensification of the immunization programme. In addition, the quality of epidemiological data about hepatitis B virus in the Maghreb remains rather disparate, whereas knowledge about HDV is almost nil except in Tunisia and Mauritania. Novel large-scale epidemiological studies aiming at prevalence and risk factors identification should be launched on cooperative bases. Accurate molecular methods to identify subtypes, variants and recombinant genomes, are required to estimate current and future burdens of HBV infection to implement effective preventive measures.