Mortality rates in people dually infected with HIV-1/2 and those infected with either HIV-1 or HIV-2: a systematic review and meta-analysis

Prince, Puck D.a,*; Matser, Amya,b,*; van Tienen, Carlac; Whittle, Hilton C.d; Schim van der Loeff, Maarten F.a,e

doi: 10.1097/01.SPC.0000432532.87841.78
Epidemiology and Social

Objective: As compared to HIV-1 infection, HIV-2 is less transmissible, disease progression is slower, and the mortality risk is lower. It has been suggested that HIV-2 infection inhibits the progression of HIV-1 in individuals dually infected by HIV-1 and HIV-2 (HIV-D). We examined whether the mortality rates in dually infected individuals differ from those in persons infected with either HIV-1 or HIV-2.

Design: We conducted a systematic review and meta-analysis.

Methods: Medline and Embase databases were searched for studies that reported the number of deaths and person-years of observation (PY) for at least two of the three HIV groups (i.e. HIV-1, HIV-2, and HIV-D). Meta-analyses were then performed with random-effects models, estimating combined mortality rate ratios (MRRs).

Results: Of the 631 identified titles, six articles were included in the meta-analysis of HIV-D-infected individuals versus HIV-mono-infected persons, and seven were included in the analysis of HIV-1-mono-infected versus HIV-2-mono-infected individuals. The overall MRR of those infected with HIV-D versus HIV-1 was 1.11 [95% confidence interval (CI) 0.95–1.30]. The overall MRR of those infected with HIV-D versus HIV-2 was 1.81 (95% CI 1.43–2.30) and the MRR of those infected with HIV-1 versus HIV-2 was 1.86 (95% CI 1.44–2.39).

Conclusion: HIV-2-mono-infected persons have a lower mortality rate than those mono-infected with HIV-1 and those with HIV-D. There is no evidence that HIV-2 delays progression to death in HIV-D-infected individuals.

Author Information

aCluster of Infectious Diseases, Amsterdam Public Health Service, Amsterdam

bJulius Centre for Health Sciences & Primary Care, University Medical Centre Utrecht, Utrecht

cMedical Microbiology and Infectious Diseases, Erasmus Medical Center, Rotterdam, the Netherlands

dLondon School of Hygiene and Tropical Medicine, London, UK

eCenter for Infection and Immunology Amsterdam, Academic Medical Center, Amsterdam, the Netherlands.

*Puck D. Prince and Amy Matser contributed equally to the writing of this article.

Correspondence to Amy Matser, Public Health Service of Amsterdam, Cluster of Infectious Diseases, Department of Research, Postbox 2200, 1000 CE Amsterdam, The Netherlands. Tel: +31 020 5555362; e-mail.

Received 26 April, 2013

Revised 21 June, 2013

Accepted 27 June, 2013

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Website (

Article Outline
Back to Top | Article Outline


According to the World Health Organization, an estimated 34 million individuals were living with HIV at the end of 2010 [1]. Most infections are caused by HIV-1, whereas HIV-2 is almost exclusively found in West Africa. In the absence of HIV-1, the prevalence of HIV-2 among adults in Guinea-Bissau was 8–10% [2–4]. Since the 1990s, the prevalence of HIV-2 in West Africa has decreased, whereas the prevalence of HIV-1 has increased [5–8]. HIV-1 and HIV-2 share the same transmission routes, but HIV-2 is less transmissible [9–11], and it has a longer median time from infection to AIDS [12]. The mortality rate among HIV-1 infected individuals is substantially higher than that among HIV-2-infected individuals [13,18,19]. As the genetic structure of HIV-1 and HIV-2 are nearly identical [14] and the pathogenesis of disease induced by the viruses is very similar [15], HIV-2 is regarded as an important model for HIV-1 pathogenesis, and studying immune responses to HIV-2 infection may provide clues for HIV-1 vaccine development [16].

In West Africa, dual infections with both HIV-1 and HIV-2 (HIV-D) are relatively common. In The Gambia, the prevalence among clinic patients was relatively stable over time and was 0.9% in 2003 [8]. In Senegal, the prevalence among sex workers increased from 0.2% in 1985 to 3.2% in 1995 and then decreased to 1.8% in 2003 [17].

A recent study in an occupational cohort of police officers in Guinea-Bissau compared disease progression between individuals infected with HIV-D and those infected with only HIV-1. Esbjörnsson et al.[20] concluded that individuals with HIV-D had a significantly slower disease progression, defined as time to AIDS, than people with only HIV-1 infection. If this is true, HIV-D-infected individuals would be expected to have a lower mortality rate than HIV-1-mono-infected individuals [21]. However, several large comparative studies from West Africa did not report reduced mortality rates among the HIV-D-infected individuals [13,18,22].

We conducted a systematic review and meta-analysis to summarize and combine the available data on HIV-D, HIV-1 and HIV-2 mortality rates reported by studies with data on at least two of the HIV groups. We compared the mortality rates of individuals infected with either HIV-1 or HIV-2 and those infected with HIV-D, and we compared individuals mono-infected with HIV-1 and those mono-infected with HIV-2.

Back to Top | Article Outline


Protocol and inclusion criteria

A protocol was prepared and piloted (Appendix 1,, in which search strategy, inclusion criteria, methods of data extraction and an analysis plan were documented. We included longitudinal studies that reported the number of deaths for at least two of the following groups from the same population: HIV-1-mono-infected individuals, HIV-2-mono-infected individuals, and HIV-D-infected individuals. There were no restrictions regarding the study population, language or publication dates. In writing this article, we used the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [23].

Back to Top | Article Outline
Search and study selection

A systematic search was conducted on September 17, 2012, in Medline (1966–Present) and Embase (1980–Present) databases. The following search terms were used: HIV-1, HIV-2, mortality, death, fatality, survival, disease progression, outcome assessment, and, if possible, their index terms. The exact search strategies for Medline and Embase are provided in Appendix 1, The study selection was performed by two independent researchers. Articles were first screened by title and then by abstract. The full text was screened when articles potentially met the inclusion criteria or when it was not clear from the abstract whether the criteria were met. References for articles included in the full-text screening were also screened. Full-text articles were selected if mortality data for at least two HIV groups (i.e., HIV-D and HIV-1, HIV-D and HIV-2, or HIV-1 and HIV-2) were reported (Fig. 1). Any disagreement was solved by consensus. If consensus was not reached, a third researcher made the final decision.

Back to Top | Article Outline
Data extraction

During the process of data extraction, three researchers extracted the data independently, using a standardized, piloted form. Differences were resolved by consensus. The extracted information included study characteristics (i.e. first author, article title, country where study originated, journal, and year of publication), characteristics of the study population (i.e. inclusion criteria, inclusion dates, follow-up dates, source of patients, reported comorbidities, use of ART, prophylaxis and whether HIV cases were prevalent or incident), and the methods used for HIV diagnosis. Furthermore, we extracted by HIV group the total number of persons included, age and sex, follow-up times, number of deaths, the number lost to follow-up, CD4+ count at baseline and viral load at baseline.

In case several studies described the same cohort population and reported overlapping inclusion periods the study that contained the largest population or had the longest follow-up was included. Authors were contacted to provide extra data when needed.

Back to Top | Article Outline
Risk of bias in individual studies

We developed and piloted a quality assessment checklist to assess the risk of bias in the studies. This checklist was based on the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) checklist [24] and the Newcastle-Ottawa Scale checklist [25] and was adapted for this review (Appendix 1, Checklist items were scored by three independent researchers using ‘Good’ (+), ‘Moderate’ (±), and ‘Poor’ (-). If information was not available in the study, this was indicated by (x). If consensus was not reached, a fourth researcher made the final decision.

Back to Top | Article Outline
Statistical analysis

We performed meta-analyses using random-effect models to estimate the mortality rate ratio (MRR) and 95% confidence intervals (CI) of individuals infected with HIV-D versus those infected with HIV-1, of those infected with HIV-D versus HIV-2 and of those infected with HIV-1 versus HIV-2. If differences in disease progression and mortality rates between HIV groups exist, they would be more easily detectable in cohorts of asymptomatic individuals than in cohorts with advanced disease [13,15]. To account for this, we performed two a priori defined subanalyses. First, we distinguished community-based studies from studies in which patients were recruited in a hospital or clinic. Second, a distinction was made on the basis of the patients’ disease progression. Early disease progression was assumed for individuals recruited in a community survey. Advanced disease progression was assumed for individuals with HIV-related comorbidities who were recruited in hospitals, and intermediate disease progression was assumed for individuals with HIV-related comorbidities recruited through a community survey or at health clinics.

In a third subanalysis, we accounted for quality of the study as this might have affected the outcome. In the quality assessment, items were scored as (+), (±), (−), or (x), corresponding to the values 3 to 0. Sum scores were used to qualify the studies on the basis of the median sum score as either low or high. Heterogeneity was estimated with use of a measure of inconsistency (I2) [26]. In case of heterogeneity, a leave-one-out sensitivity analysis was performed to assess the individual contribution of a study to the combined estimate. We used Stata 11.2 (Stata Corp., College Station, Texas, USA) and the user-written metan package [27].

Back to Top | Article Outline


Study selection

We identified 631 potentially eligible studies (Fig. 1). After screening by abstract and title, 50 studies remained for full-text screening, and after full-text screening, 20 studies were considered eligible. After considering cohort overlap, eight studies were excluded. Six of the remaining were potentially eligible according to the study design, but specific data to calculate the MRR were not reported. Authors of these studies were contacted. After the data request, the person-years of observation were obtained from van Tienen et al.[18]. This study was included, and the other five were excluded. Eventually, seven were selected for data extraction. All seven reported information on mortality rates among individuals infected with HIV-1 and HIV-2, and six of the seven studies reported on mortality rates among HIV-D-infected and HIV-mono-infected individuals (Fig. 1).

Back to Top | Article Outline
Study characteristics

The included studies were all cohort studies from West Africa, published in English (Table 1), including adults (age ≥15 years) living in the study area. Van Tienen et al.[18] and Holmgren et al.[22] recruited participants from the general population. Schim van der Loeff et al.[13] and Gustafson et al.[28] recruited participants at a clinic. Norrgren et al.[29] and Kassim et al.[30] recruited hospital patients with tuberculosis who were followed during tuberculosis treatment, and Hansmann et al.[31] recruited pregnant women. Follow-up times varied between studies; it ranged from 8 months [28,29] to 20.4 years [18]. The number of participants ranged from 103 [29] to 1519 [13]. HIV diagnoses were made either by enzyme-linked immunosorbent assay (ELISA) and polymerase chain reaction (PCR) [13,18,31] or by a combination of ELISA and a confirmation antibody test [22,28–30]. Most studies included prevalent cases of HIV. Holmgren et al.[22] and Schim van der Loeff et al.[13] reported a few incident HIV cases.

Back to Top | Article Outline
Mortality rate ratio of HIV-D versus HIV-1

Six studies were included in the analysis of HIV-D versus HIV-1 mortality rates (Fig. 2a). The analysis involved 512 individuals infected with HIV-D and 1169 individuals with HIV-1; of those, 200 persons with HIV-D and 554 with HIV-1 died during follow-up. None of the individual studies had an MRR that significantly differed from 1. Individual estimates ranged from 0.82 (95% CI 0.48–1.40) [28] to 1.31 (95% CI 0.91–1.88) [30]. The combined MRR was 1.11 (95% CI 0.95–1.30), so the mortality rates for HIV-D and HIV-1 did not differ significantly. There was no indication of heterogeneity between studies (I2 = 0%, P = 0.80).

Back to Top | Article Outline
Mortality rate ratio of HIV-D versus HIV-2

Six studies were included in the analysis of HIV-D versus HIV-2 mortality rates (Fig. 2b). The analysis involved 512 individuals infected with HIV-D and 1706 individuals with HIV-2; of those, 200 persons with HIV-D and 643 with HIV-2 died during follow-up. All studies found higher mortality rates for HIV-D than for HIV-2. The studies with the highest weights demonstrated significant effects, whereas the studies with a lower weight showed nonsignificant effects. The combined MRR was 1.81 (95% CI 1.43–2.30), which showed that HIV-D-infected individuals had a substantially higher death rate than HIV-2-mono-infected individuals. There was moderate heterogeneity (I2 = 39.2%; P = 0.14).

Back to Top | Article Outline
Mortality rate ratio of HIV-1 versus HIV-2

Seven studies were included in the analysis of mortality rates for HIV-1 versus HIV-2 (Fig. 2c). The analysis involved 1270 individuals infected with HIV-1 and 1949 individuals with HIV-2; of those, 586 persons with HIV-1 and 666 with HIV-2 died during follow-up. Five of the seven studies found significantly increased MRRs for HIV-1 compared with HIV-2, ranging from 1.45 (95% CI 1.28–1.64) [13] to 3.45 (95% CI 2.03–5.84) [31]. The two other studies showed a nonsignificantly increased MRR [22,29]. The combined MRR was 1.86 (95% CI 1.44–2.39). Heterogeneity was moderately high (I2 = 62.1%; P = 0.015). The leave-one-out analyses showed that most heterogeneity was caused by Hansmann et al.[31] in which only pregnant women were included [31]; I2 dropped to 32.1% (P = 0.195) when this study was excluded.

Back to Top | Article Outline

We conducted three predefined subanalyses (Table 3). First, we examined the MRRs by patient setting. Norrgren et al.[29], Schim van der Loeff et al.[13], Kassim et al.[30] and Gustafson et al.[28] described hospital or clinic populations, and Holmgren et al.[22], van Tienen et al.[18] and Hansmann et al.[31] described community populations. The combined MRR of the hospital and clinic studies comparing the mortality rate of HIV-D to that of HIV-1 was 1.13 (95% CI 0.95–1.34) and did not differ significantly from the combined MRR of 1.04 (95% CI 0.71–1.51) in the community studies. Second, we examined the MRRs of HIV-D versus HIV-1 by stage of disease of the population. In this analysis, the combined MRR of HIV-D versus those of HIV-1 in a population that was, on average, infected early, was 1.07 (95% CI 0.71–1.60) [18] and did not differ significantly from the combined MRR among individuals who had, on average, an intermediate HIV infection (1.11; 95% CI 0.89–1.39) [13,22] or advanced HIV infection (1.12; 95% CI 0.85–1.48) [28–30]. Third, we examined the MRRs of those infected with HIV-D versus HIV-1 by study quality. The median sum score calculated from the quality assessment was 25 (Table 2); therefore, we considered the quality of a study low when the sum score was less than 25 and high when sum score was at least 25. Three studies were considered of low quality [22,28,30] and four of high quality [13,18,29,31]. The MRRs were 1.08 (95% CI 0.78–1.50) for low and 1.11 (95% CI 0.92–1.34) for high quality, showing no significant difference between low-quality and high-quality studies.

Back to Top | Article Outline


In this comparative review and meta-analysis, we found that the mortality rates of both HIV-1-mono-infected individuals and those infected with HIV-D were significantly higher than the rate among HIV-2-mono-infected individuals. The mortality rate of HIV-D-infected individuals did not significantly differ from the mortality rate of HIV-1-mono-infected individuals. Even when the analysis was limited to studies that included patients at relatively earlier stages of HIV infection, there was no indication of a difference in mortality rate between HIV-D-infected and HIV-1-mono-infected individuals.

It is known that the mortality rate for HIV-2 is lower than that for HIV-1. The first large study on HIV-2 epidemiology suggested this [2], with confirmation by later studies [19,32,33]. The best explanation for this difference is the overall much lower plasma viral load in HIV-2-infected individuals [31,34,35]. It has been suggested that HIV-2 infection might protect against HIV-1 infection [36], but this effect could not be shown in subsequent studies [37–39]. In contrast, a meta-analysis showed that HIV-2-infected individuals are more likely to acquire HIV-1 infection than HIV-negative individuals [38]. Several laboratory studies suggested that in dually infected individuals, HIV-2 exerted a possible inhibitory effect on HIV-1 disease progression [20,40–43], but other studies found no effect of HIV-2 on the viral load of HIV-1 [34,44]. In this meta-analysis, we did not find evidence for an inhibitory effect, as there was no significant difference between mortality rates for HIV-D and HIV-1. We also did not find a significant effect in the subanalyses that included individuals in earlier stages of the disease from those in advanced stages. Because some of the studies reported mortality rates of patients treated for tuberculosis, which is one of the main causes of death in HIV-infected individuals, mortality rates were probably higher in those patients. Thus, it can be argued that differences in mortality rates might not be as evident in cases of advanced disease. Adjusting for disease stage by CD4+ cell count could have provided further insight, but too few studies provided this information. In the study by Schim van der Loeff et al.[13] it was shown that differences in mortality rates between HIV-1 and HIV-2 infection are more pronounced when CD4+ cell counts are high. In the patient group with high CD4+ cell counts (>500/μl), the authors did not find a significant difference between mortality rates of HIV-D and HIV-1 infection, which agrees with our findings.

An ideal study design would consist of sero-incident HIV cases followed for several years. Unfortunately, such studies do not exist. This meta-analysis included only studies in which most of the populations had prevalent HIV infections with unknown seroconversion dates, which could have led to bias in several ways. In view of the history of the two epidemics in West Africa, we may assume that the large majority of HIV-D-infected individuals were first infected with HIV-2 and later acquired HIV-1 infection [5,39,45,46]. Individuals more prone to disease progression, comorbidities or mortality, with only HIV-2 or HIV-D infection might have died before the start of studies. In this case, the HIV-D-infected population might be a population of survivors with a more favorable prognosis and a lower mortality rate, and the observed MRR of HIV-D versus HIV-1 infection might be an underestimate of the true effect. On the contrary, the observed MRR of HIV-D versus HIV-1 infection might be an overestimate if individuals infected with HIV-D had a longer history of HIV-1 infection than HIV-1-mono-infected individuals.

Differential loss to follow-up between HIV groups might have led to underestimation or overestimation of the reported MRRs if this loss was related to mortality. One might expect that healthier individuals, who are less likely to die, are more likely to be lost, which would lead to a higher observed mortality rate of a population. There was no clear pattern of loss to follow-up by HIV-status: the HIV-group with the highest percentage of individuals who were lost differed between studies.

In the studies of Gustafson et al.[28], Kassim et al.[30], Hansmann et al.[31] and Norrgren et al., [29] individuals were assigned an HIV status on the basis of tests performed at baseline, but some mono-infected individuals may have acquired an additional HIV infection during follow-up. This occurrence would have led to misclassification of those individuals as mono-infected, rather than dually infected. Three studies allowed for HIV seroconversion during follow-up by permitting these individuals to contribute person-time to two HIV groups [13,18,22]. In the other studies, potential misclassification would have led to individuals being misclassified as HIV-2-infected instead of HIV-D-infected; it might also have led to a lower observed MRR of HIV-1 infection compared with HIV-2, but no change for the MRR of HIV-D infection compared with HIV-1. Two studies reported some seroconverters during follow-up, but the outcomes of these cases were not reported separately [13,22]. Furthermore, most studies used ELISA to differentiate HIV infections. This approach is not always adequate in differentiating between HIV monoinfection and HIV-D infection, owing to cross-reactivity of antibodies. Even when monospecific ELISA and confirmatory western blot testing are used, cross-reactivity can still be observed [47–51].

In two studies, co-trimoxazole prophylaxis was provided, which reduces morbidity and the mortality rate [18,28]. In these studies, the mortality rates might be slightly reduced due to co-trimoxazole use and the MRRs might be slight underestimates or overestimates depending on the HIV group with the highest number of individuals in advanced disease stage.

Co-infections, for example with human T-lymphotropic virus (HTLV-1), might influence mortality rates. Where measured, the prevalence of HTLV-1 was highest in the HIV-D and lowest in the HIV-1 group [18,29]. Assuming an independent effect of HTLV-1 on mortality, this would have led to an overestimate of the MRRs comparing HIV-D to HIV-1 and HIV-D to HIV-2, and to an underestimate of the MRR of HIV-1 versus HIV-2.

Potential publication bias was not formally assessed, because the number of included studies was low [52]. In case of publication bias, it is most likely that studies that found a nonsignificant effect would not have been published. Still, publication bias would not have affected the MRR of HIV-D infection versus HIV-1, but it could have affected the MRR of HIV-D infection versus HIV-2 and HIV-1 versus HIV-2, which might be overestimated in this circumstance.

Heterogeneity was low-to-moderate in the analyses comparing the mortality rates of HIV-D-infected persons with those of HIV-mono-infected individuals, but it was relatively high in the analysis of HIV-1 and HIV-2 monoinfection. The leave-one-out sensitivity analysis showed that the high heterogeneity was caused mainly by one study in which only pregnant women were included who were probably diagnosed early in their infection [31]. The MRR of HIV-1 versus HIV-2 infection was higher in this study than in the other studies.

The conclusion drawn from this meta-analysis contradicts with the previous findings of Esbjörnsson et al.[20]. In that study, it was shown that HIV-1 disease progression was inhibited by HIV-2 in dually infected individuals. Population characteristics (e.g. age and sex) cannot explain the different findings. Unfortunately, the study could not be included in the meta-analysis, because it did not report mortality data, but used progression to AIDS as endpoint. In the study the HIV-1 seroconversion date was known, but not the (preceding) HIV-2 seroconversion date; thus survivor bias might have caused the observed protective effect of HIV-2 on HIV-1 disease progression [20]. The study was rather small, and therefore it is unlikely that adding this study would have changed the conclusion of the meta-analysis.

The current literature is not sufficient to draw definite conclusions about the interaction between HIV-2 and HIV-1 infection on the risk of disease progression as measured by mortality rates, because it was not possible to adjust for time since seroconversion or markers such as CD4+ count. It is unlikely that this problem will be solved in the future. The number of cohort studies in West Africa is limited, and available data have probably already been published. In addition, further studies comparing the natural history of the infections, including mortality rates, are impossible, because we have entered the ART era, with effective treatment against HIV-1 infection and, to a lesser degree, against HIV-2, which alters the natural history of HIV infection. Endpoints such as time to AIDS, time to CD4+ count less than 350 cells/μl, or time to the start of ART may still be possible, but these endpoints are more disputable than mortality.

In this meta-analysis, we have confirmed that the mortality rates among HIV-1-mono-infected individuals and HIV-D-infected individuals are higher than among HIV-2-infected individuals, but we did not find a difference between the mortality rates for HIV-D infection and HIV-1 infection. Therefore, by measuring mortality rates, we conclude that there is no evidence in HIV-D-infected persons that HIV-2 infection impedes the progression of HIV-1 infection.

Back to Top | Article Outline


The authors thank I. Nagel (AMC library) for the assistance with the search strategy, Dr P. Scott for the helpful comments during the design of the study, Dr R. Geskus for the statistical help and S. H. Ebeling for editing the article. We also thank the authors Prof. K. de Cock, Prof. S. Lucas, and Prof. J.E. Malkin who kindly responded to our request for data. We also thank two anonymous reviewers for their useful comments.

Author contributions: P.P., A.M., C.T., M.S. contributed to the design of the study and protocol and performed the systematic review. P.P., A.M., M.S. performed the analyses. P.P., A.M., C.T., H.W., M.S. drafted and made a substantive intellectual contribution to the manuscript.

Back to Top | Article Outline
Conflicts of interest

The authors declare that they have no conflicts of interest.

Back to Top | Article Outline


1. WHO, UNAIDS, UNICEF. Global HIV/AIDS response. Epidemic update and health sector progress towards Universal Access. Progress report 2011. Geneva, Switzerland: WHO Press; 2011.
2. Poulsen AG, Kvinesdal B, Aaby P, Molbak K, Frederiksen K, Dias F, et al. Prevalence of and mortality from human immunodeficiency virus type 2 in Bissau, West Africa. Lancet 1989; 1:827–831.
3. Poulsen AG, Aaby P, Gottschau A, Kvinesdal BB, Dias F, Molbak K, et al. HIV-2 infection in Bissau, West Africa, 1987–1989: incidence, prevalences, and routes of transmission. J Acquir Immune Defic Syndr 1993; 6:941–948.
4. Wilkins A, Ricard D, Todd J, Whittle H, Dias F, Paulo Da SA. The epidemiology of HIV infection in a rural area of Guinea-Bissau. AIDS 1993; 7:1119–1122.
5. da Silva ZJ, Oliveira I, Andersen A, Dias F, Rodrigues A, Holmgren B, et al. Changes in prevalence and incidence of HIV-1, HIV-2 and dual infections in urban areas of Bissau, Guinea-Bissau: is HIV-2 disappearing?. AIDS 2008; 22:1195–1202.
6. Norrgren H, Andersson S, Biague AJ, da Silva ZJ, Dias F, Naucler A, et al. Trends and interaction of HIV-1 and HIV-2 in Guinea-Bissau, west Africa: no protection of HIV-2 against HIV-1 infection. AIDS 1999; 13:701–707.
7. Mansson F, Alves A, Silva ZJ, Dias F, Andersson S, Biberfeld G, et al. Trends of HIV-1 and HIV-2 prevalence among pregnant women in Guinea-Bissau, West Africa: possible effect of the civil war 1998 1999. Sex Transm Infect 2007; 83:463–467.
8. Schim van der Loeff M, Awasana A, Sarge-Njie R, van der Sande M, Jaye A, Sabally S, et al. Sixteen years of HIV surveillance in a West African research clinic reveals divergent epidemic trends of HIV-1 and HIV-2. Int J Epidemiol 2006; 35:1322–1328.
9. Adjorlolo-Johnson G, De Cock KM, Ekpini E, Vetter KM, Sibailly T, Brattegaard K, et al. Prospective comparison of mother-to-child transmission of HIV-1 and HIV-2 in Abidjan, Ivory Coast. JAMA 1994; 272:462–466.
10. Gilbert PB, McKeague IW, Eisen G, Mullins C, Gueye-Ndiaye A, Mboup S, et al. Comparison of HIV-1 and HIV-2 infectivity from a prospective cohort study in Senegal. Stat Med 2003; 22:573–593.
11. Kanki PJ, Travers KU, Mboup S, Hsieh CC, Marlink RG, Gueye-Ndiaye A, et al. Slower heterosexual spread of HIV-2 than HIV-1. Lancet 1994; 343:943–946.
12. Marlink R, Kanki P, Thior I, Travers K, Eisen G, Siby T, et al. Reduced rate of disease development after HIV-2 infection as compared to HIV-1. Science 1994; 265:1587–1590.
13. Schim van der Loeff M, Jaffar S, Aveika AA, Sabally S, Corrah T, Harding E, et al. Mortality of HIV-1, HIV-2 and HIV-1/HIV-2 dually infected patients in a clinic-based cohort in The Gambia. AIDS 2002; 16:1775–1783.
14. Guyader M, Emerman M, Sonigo P, Clavel F, Montagnier L, Alizon M. Genome organization and transactivation of the human immunodeficiency virus type 2. Nature 1987; 326:662–669.
15. Martinez-Steele E, Awasana AA, Corrah T, Sabally S, van der Sande M, Jaye A, et al. Is HIV-2- induced AIDS different from HIV-1-associated AIDS? Data from a West African clinic. AIDS 2007; 21:317–324.
16. Rowland-Jones S. A winding road towards an HIV vaccine. Eur J Immunol 2008; 38:13–14.
17. Hamel DJ, Sankale JL, Eisen G, Meloni ST, Mullins C, Gueye-Ndiaye A, et al. Twenty years of prospective molecular epidemiology in Senegal: changes in HIV diversity. AIDS Res Hum Retroviruses 2007; 23:1189–1196.
18. van Tienen C, Schim van der Loeff M, Peterson I, Cotten M, Andersson S, Holmgren B, et al. HTLV-1 and HIV-2 infection are associated with increased mortality in a rural West African community. PLoS One 2011; 6:e29026.
19. Poulsen AG, Aaby P, Larsen O, Jensen H, Naucler A, Lisse IM, et al. 9-year HIV-2-associated mortality in an urban community in Bissau, west Africa. Lancet 1997; 349:911–914.
20. Esbjörnsson J, Mansson F, Kvist A, Isberg PE, Nowroozalizadeh S, Biague AJ, et al. Inhibition of HIV-1 disease progression by contemporaneous HIV-2 infection. N Engl J Med 2012; 367:224–232.
21. van Tienen C, Schim van der Loeff M, Whittle H. Effect of HIV-2 infection on HIV-1 disease progression. N Engl J Med 2012; 367:1961–1963.
22. Holmgren B, da SZ, Vastrup P, Larsen O, Andersson S, Ravn H, et al. Mortality associated with HIV-1, HIV-2, and HTLV-I single and dual infections in a middle-aged and older population in Guinea-Bissau. Retrovirology 2007; 4:85.
23. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Int J Surg 2010; 8:336–341.
24. von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008; 61:344–349.
25. 2000; Wells G, Shea B, O’Connell D, Peterson J, Welch V, Losos M, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. 3rd Symposium on Systematic Reviews: beyond the basics. Improving quality and impact
26. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003; 327:557–560.
27. Harris RJ, Bradburn MJ, Deeks JJ, Harbord RM, Altman DG, Sterne JAC. Metan: fixed- and random-effects meta-analysis. Stata J 2008; 8:3–28.
28. Gustafson P, Gomes VF, Vieira CS, Samb B, Naucler A, Aaby P, et al. Clinical predictors for death in HIV-positive and HIV-negative tuberculosis patients in Guinea-Bissau. Infection 2007; 35:69–80.
29. Norrgren H, Bamba S, da Silva ZJ, Koivula T, Andersson S. Higher mortality in HIV-2/HTLV-1 co-infected patients with pulmonary tuberculosis in Guinea-Bissau, West Africa, compared to HIV-2-positive HTLV-1-negative patients. Int J Infect Dis 2010; 14 (Suppl 3):e142–e147.
30. Kassim S, Sassan-Morokro M, Ackah A, Abouya LY, Digbeu H, Yesso G, et al. Two-year follow-up of persons with HIV-1- and HIV-2-associated pulmonary tuberculosis treated with short-course chemotherapy in West Africa. AIDS 1995; 9:1185–1191.
31. Hansmann A, Schim van der Loeff MF, Kaye S, Awasana AA, Sarge-Njie R, O’Donovan D, et al. Baseline plasma viral load and CD4 cell percentage predict survival in HIV-1- and HIV-2-infected women in a community-based cohort in The Gambia. J Acquir Immune Defic Syndr 2005; 38:335–341.
32. Whittle H, Morris J, Todd J, Corrah T, Sabally S, Bangali J, et al. HIV-2-infected patients survive longer than HIV-1-infected patients. AIDS 1994; 8:1617–1620.
33. Schim van der Loeff MF, Larke N, Kaye S, Berry N, Ariyoshi K, Alabi A, et al. Undetectable plasma viral load predicts normal survival in HIV-2-infected people in a West African village. Retrovirology 2010; 7:46.
34. Alabi AS, Jaffar S, Ariyoshi K, Blanchard T, Schim van de Loeff MF, Awasana AA, et al. Plasma viral load, CD4 cell percentage, HLA and survival of HIV-1, HIV-2, and dually infected Gambian patients. AIDS 2003; 17:1513–1520.
35. O’Donovan D, Ariyoshi K, Milligan P, Ota M, Yamuah L, Sarge-Njie R, et al. Maternal plasma viral RNA levels determine marked differences in mother-to-child transmission rates of HIV-1 and HIV-2 in The Gambia. MRC/Gambia Government/University College London Medical School working group on mother-child transmission of HIV. AIDS 2000; 14:441–448.
36. Travers K, Mboup S, Marlink R, Gueye-Nidaye A, Siby T, Thior I, et al. Natural protection against HIV-1 infection provided by HIV-2. Science 1995; 268:1612–1615.
37. Ariyoshi K, Schim van der Loeff M, Sabally S, Cham F, Corrah T, Whittle H. Does HIV-2 infection provide cross-protection against HIV-1 infection?. AIDS 1997; 11:1053–1054.
38. Greenberg AE. Possible protective effect of HIV-2 against incident HIV-1 infection: review of available epidemiological and in vitro data. AIDS 2001; 15:2319–2321.
39. Schim van der Loeff M, Aaby P, Aryioshi K, Vincent T, Awasana AA, da CC, et al. HIV-2 does not protect against HIV-1 infection in a rural community in Guinea-Bissau. AIDS 2001; 15:2303–2310.
40. Al-Harthi L, Owais M, Arya SK. Molecular inhibition of HIV type 1 by HIV type 2: effectiveness in peripheral blood mononuclear cells. AIDS Res Hum Retroviruses 1998; 14:59–64.
41. Arya SK, Gallo RC. Human immunodeficiency virus (HIV) type 2-mediated inhibition of HIV type 1: a new approach to gene therapy of HIV-infection. Proc Natl Acad Sci U S A 1996; 93:4486–4491.
42. Kokkotou EG, Sankale JL, Mani I, Gueye-Ndiaye A, Schwartz D, Essex ME, et al. In vitro correlates of HIV-2-mediated HIV-1 protection. Proc Natl Acad Sci U S A 2000; 97:6797–6802.
43. Rappaport J, Arya SK, Richardson MW, Baier-Bitterlich G, Klotman PE. Inhibition of HIV-1 expression by HIV-2. J Mol Med (Berl) 1995; 73:583–589.
44. Nkengasong JN, Kestens L, Ghys PD, Koblavi-Deme S, Otten RA, Bile C, et al. Dual infection with human immunodeficiency virus type 1 and type 2: impact on HIV type 1 viral load and immune activation markers in HIV-seropositive female sex workers in Abidjan, Ivory Coast. AIDS Res Hum Retroviruses 2000; 16:1371–1378.
45. de Silva TI, van Tienen C, Rowland-Jones SL, Cotten M. Dual infection with HIV-1 & HIV-2: double trouble of destructive interference?. HIV Therapy 2010; 4:305–323.
46. van Tienen C, Schim van der Loeff MS, Zaman SM, Vincent T, Sarge-Njie R, Peterson I, et al. Two distinct epidemics: the rise of HIV-1 and decline of HIV-2 infection between 1990 and 2007 in rural Guinea-Bissau. J Acquir Immune Defic Syndr 2010; 53:640–647.
47. Ariyoshi K, Cham F, Berry N, Harding E, Sabally S, N’Gom PT, et al. Diagnosis of HIV-1/2 dual infection using dilution analysis of type-specific antibody. AIDS 1998; 12:2504–2505.
48. Tedder RS, O’Connor T, Hughes A, N’jie H, Corrah T, Whittle H. Envelope cross-reactivity in Western blot for HIV-1 and HIV-2 may not indicate dual infection. Lancet 1988; 2:927–930.
49. De Cock KM, Porter A, Kouadio J, Maran M, Lafontaine MF, Gershy-Damet GM, et al. Cross-reactivity on western blots in HIV-1 and HIV-2 infections. AIDS 1991; 5:859–863.
50. Andersson S, da SZ, Norrgren H, Dias F, Biberfeld G. Field evaluation of alternative testing strategies for diagnosis and differentiation of HIV-1 and HIV-2 infections in an HIV-1 and HIV-2-prevalent area. AIDS 1997; 11:1815–1822.
51. Nkengasong JN, Maurice C, Koblavi S, Kalou M, Bile C, Yavo D, et al. Field evaluation of a combination of monospecific enzyme-linked immunosorbent assays for type-specific diagnosis of human immunodeficiency virus type 1 (HIV-1) and HIV-2 infections in HIV-seropositive persons in Abidjan, Ivory Coast. J Clin Microbiol 1998; 36:123–127.
52. Egger M, Davey SG, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997; 315:629–634.

HIV-1; HIV-2; HIV-dual; meta-analysis; mortality; systematic review

Supplemental Digital Content

Back to Top | Article Outline
Copyright © 2014 Wolters Kluwer Health, Inc.