Skip Navigation

Natural History and Risk Factors Associated with Early and Established HIV Type 1 Infection among Reproductive-Age Women in Malawi

  1. Johnstone J. Kumwenda1,
  2. Bonus Makanani1,
  3. Frank Taulo1,
  4. Chiwawa Nkhoma2,
  5. George Kafulafula1,
  6. Qing Li3,
  7. Newton Kumwenda3, and
  8. Taha E. Taha3
  1. 1College of Medicine, University of Malawi, Chichiri, Maryland
  2. 2Johns Hopkins University–University of Malawi College of Medicine Research Project, Blantyre, Malawi, Maryland
  3. 3Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland
  1. Reprints or correspondence: Dr. Taha E. Taha, Bloomberg School of Public Health, Johns Hopkins University, 615 N. Wolfe St., Rm. E7138, Baltimore, MD 21205 (ttaha{at}jhsph.edu).
 
Next Section

Abstract

Background. Data evaluating the biological events and determinants of early human immunodeficiency virus type 1 (HIV-1) infection are limited in sub-Saharan Africa. We examined plasma viral levels and trends during early and established HIV-1 infection among reproductive-age women who participated in a randomized trial to treat genital tract infection in Malawi. We also assessed the association of injectable hormonal contraceptive use with HIV-1 infection.

Methods. We studied 3 groups of women who were infected or uninfected with HIV-1: seroconverters, seroprevalent women, and seronegative women. Questionnaires and blood samples were collected at baseline and every 3 months for 1 year. The virus set point in seroconverters and levels and trends of viral load over time were determined. The associations of injectable hormonal contraceptive use with HIV-1 infection and viral load were assessed using conditional logistic regression and mixed-effect models, respectively.

Results. In the original clinical trial, 844 women infected with HIV-1 and 842 women not infected with HIV-1 were enrolled. Of 31 women who experienced seroconversion during 12 months, 27 were matched with 54 seroprevalent and 54 seronegative women. The estimated median plasma virus set point was 4.45 log10 copies/mL (interquartile range, 4.32–5.14 log10 copies/mL). Injectable hormonal contraceptive use was significantly associated with HIV-1 seroconversion (adjusted odds ratio, 10.42; P=.03) but not with established HIV-1 infection. Among the seroconverters, a statistically significant interaction was found between the linear association of viral load and time of injectable hormonal contraceptive use (regression coefficient, −0.14; P=.02).

Conclusion. Knowledge of virus set point and trends of viral load in HIV-1 seroincident and seroprevalent asymptomatic women could assist in antiretroviral treatment management.

Natural history data from sub-Saharan Africa describing the trends of plasma viremia in women after seroconversion are scarce. The plasma viral loads are generally high in southern Africa, possibly because of HIV-1 subtype C—the most dominant clade in the southern region of Africa—and other factors, such as intercurrent infections. This sustained elevation of the viral level may lead to faster disease progression [1]. It is important to study these populations to understand determinants of infection and, consequently, to develop potential interventions.

During primary HIV-1 infection, high levels of virus replication are demonstrated by a steep increase in plasma HIV-1 RNA levels that often reach a peak of >106 copies/mL approximately 2–3 weeks after infection [2–4]. Although virologic and immunologic responses vary in HIV-1–infected individuals, mobilization of host defenses against the virus results in decreases in virus levels that reach a steady state or set point 2–6 months after infection [4, 5]. It is suggested that high steady-state levels of plasma HIV-1 RNA at 4–12 months after infection result in faster progression to AIDS [5–7]. On the basis of these initial events, HIV-1 becomes permanently established, and infected cells are constantly present in viral reservoirs in all patients [4].

HIV natural history studies among women in Africa and elsewhere have not yet confirmed the role of exposure to hormonal contraceptives. The hormonal contraceptive depot medroxyprogesterone acetate, a progestin-only injectable contraceptive, is widely used in Africa. The results of several studies that assessed the association of hormonal contraceptives and HIV-1 acquisition have been inconsistent, with some studies reporting increased risk [8] and others reporting no association [9]. Study of this factor is particularly important because globally a large number of women are advised to use hormonal contraceptives because of their benefits. This factor is also modifiable: if use of hormonal contraceptives is truly increasing the risk, then women could be counseled to use other methods of contraception.

In this article, we examine data collected from reproductive-age women in Malawi who participated in a clinical trial to treat genital tract infection [10]. We describe the patterns of the HIV-1 load over time in women who experienced seroconversion (hereafter, “seroconverters”) and in women with established or chronic HIV-1 infection (hereafter, “seroprevalent women”) and examine the association of injectable hormonal contraceptive use with HIV-1 infection.

Previous SectionNext Section

Methods

Study design and procedures. This study is based on data from an original clinical trial conducted in Malawi from January 2003 through May 2005. The details of the clinical trial were described elsewhere [10]. In brief, nonpregnant women of childbearing age, either infected or uninfected with HIV-1, were enrolled in the study at the Queen Elizabeth Central Hospital and 2 health centers in Blantyre, Malawi. These women were attending general reproductive health services, including family planning. HIV-1–infected and HIV-1–uninfected women were separately randomized to use either an active intravaginal gel (metronidazole gel) or a placebo gel. Women used these gels for 5 consecutive nights every 3 months for 1 year. Women were observed at 3-month intervals from enrollment to 12 months.

At each visit, questionnaires on clinical, sexual, and behavioral history were completed, and physical examinations were conducted. Laboratory samples (blood and cervicovaginal fluid or swabs) were collected at each visit. Blood samples were tested to determine HIV-1 infection status at baseline (Determine HIV-1/2 [Abbott Laboratories]; Wellcozyme ELISA [Wellcome Diagnostics]; and Genetic Systems EIA [Genetic Systems]) and seroconversion among HIV-1–uninfected women at subsequent visits (confirmed by Western blot test [BioRad]). HIV serologic tests were conducted while the study was ongoing, and seroconverters were appropriately counseled. Plasma viral load was measured after study closure and after seroconverters and matched seroprevalent samples were identified. HIV-1 RNA level in samples collected at each visit in seroconverters and individuals with established HIV-1 infection was measured using the Roche HIV-1 RNA Amplicor Monitor, version 1.5. Vaginal samples were tested for Trichomonas vaginalis and Candida species on the basis of standard microscopy. Gram-stained vaginal smear specimens were examined to classify vaginal flora into normal, intermediate, or bacterial vaginosis on the basis of the Nugent scoring method [11]. All tests were performed locally at the Johns Hopkins College of Medicine research laboratory, with the exception of viral load testing, which was conducted at a US reference laboratory (University of North Carolina, Chapel Hill).

In the current investigation, we performed a series of nested case-control studies. Three matched groups of women were studied. Group A contained seroconverters, who were women who entered the clinical trial uninfected with HIV-1 and became infected during the 12-month period of the trial follow-up (i.e., seroincident women). Group B contained women with established HIV-1 infection; these women were known to be HIV-1 infected at entry into the clinical trial (i.e., seroprevalent women). Group C contained women uninfected with HIV-1 from the outset to the end of the trial (i.e., seronegative women). The group A women were matched with women in groups B and C at a ratio of 1:2 on 2 factors: treatment arm of the clinical trial (either active or placebo gel) and date of enrollment (±1 week). Therefore, for each group A woman, 2 group B and 2 group C women were matched on these 2 factors. Selection of women in groups B and C was random among those who were eligible according to the matching restrictions.

In the nested case-control studies, the cases were women in groups A and B and the controls were women in group C (seronegative). After closure of the clinical trial, a limited number of the seroconverters were observed for ∼12 additional months to collect blood samples for measurements of plasma viral load and CD4+ cell counts (FACSCount; Becton Dickinson Immunocytometry Systems). Samples were collected at ∼3-month intervals in this limited study. Consent for this extended follow-up was separately obtained at the last visit; therefore, not all seroconverters were available.

All women in this study were antiretroviral naive. An antiretroviral treatment program has been recently initiated, and all women who are eligible for treatment have been referred to these services. Women received continuous counseling on HIV-1 infection, routine clinical care, and treatment of sexually transmitted infections at no cost according to the country's treatment guidelines. The study protocol was approved by the University of Malawi College of Medicine Research and Ethics Committee and the Committee on Human Research of the Johns Hopkins University Bloomberg School of Public Health.

Statistical analysis. The levels and trends of viral load for women in both groups A and B were examined after log10 transformation. In addition to calculating mean (±SD) and median (interquartile ranges) plasma viral loads over time, natural splines were used to interpolate the time trend in log10 viral load for women in groups A and B. These curves use a piecewise cubic polynomial function to smooth nonlinear changes in time series data [12]. The virus set point for each converter was estimated as the first viral load value available within the interval starting at 3 months and up to 12 months (91–365 days) since seroconversion [13].

Conditional logistic regression analyses were used to examine association of various risk factors with seroconversion or established HIV-1 infection in this matched data set. Univariate and adjusted ORs and 95% CIs were estimated. We adjusted only for age, bacterial vaginosis, number of sex partners, and use of injectable hormonal contraceptives because the sample size was small. The association of injectable hormonal contraceptive use with plasma viral load was further examined in women in groups A and B using mixed-effect models [14]. These data are not independent because of the repeated measures on the same individuals over time. The mixed-effect models include a mixture of fixed and random factors and accommodate the correlation inherent in these data. The interaction of time and hormonal contraceptive use was examined in these fixed-effect models. The Proc Mixed procedure of SAS software, version 9.1 (SAS Institute), was used for the analysis. Variance components were estimated using the restricted maximum likelihood method.

Previous SectionNext Section

Results

Overall, 842 women uninfected with HIV-1 and 844 women infected with HIV-1 were randomized at baseline in the original clinical trial. Of 842 women uninfected with HIV-1, 787 (93%) returned for follow-up visits. Of these, 31 women experienced seroconversion during 1 year of follow-up; the overall incidence of HIV-1 infection was 4.51 cases per 100 person-years (i.e., 31 infections/686.96 person-years; 95% CI, 2.96–6.06). Of these 31 HIV-1 seroconverters (group A), 27 were matched with 54 women infected with HIV-1 (group B) and 54 HIV-1–seronegative women (group C). In 4 (1.3%) of the 31 seroconverters, serologic test results were negative and HIV-1 RNA was detectable (>400 copies/mL); these were identified as acute infections, and the subjects were matched with appropriate control subjects.

Table 1 gives the characteristics of the women in groups A, B, and C at baseline and for some variables at the visit immediately before seroconversion. Seroconverting women in group A were significantly younger than seroprevalent women in group B (P<.001). Group A women also tended to use injectable hormonal contraceptives more than women in groups B and C at baseline and at the visit before seroconversion. The frequency of bacterial vaginosis and T. vaginalis infection was significantly higher among women in group A, compared with those in group C. Women in groups B and C were comparable, with the exception of number of lifetime sexual partners being significantly higher among group B women.

Figure 1
View larger version:
Figure 1

A, Spline curve showing plasma viral load (log10) in women from Malawi who seroconverted. For the acute cases (!0 months), calendar time shown does not necessarily reflect exact time of conversion because visits were scheduled at every 3 months (including 1 acute case patient who missed the visit before the visit when the first positive ELISA test result was reactive). B, Spline curve showing plasma viral load (log10) in women from Malawi with established (seroprevalent) HIV-1 infection.

Figure 2
View larger version:
Figure 2

Viral load (log10) trends over time in seroconverters (A) and seroprevalent women (B) who reported use or not use of injectable hormonal contraceptives (mixed-effects models). Numbers in parentheses are the numbers of measurements of the viral load. Solid dots, Viral load measurements in women who used hormonal contraceptives (HCs); +, viral load measurements in women who did not use HCs.

Table 1
View larger version:
Table 1

Characteristics of case patients and matched control subjects.

Throughout the study, the general health status was reported either as good or excellent in 83%–93% of women in group A, 89%–96% of those in group B, and 78%–90% of those in group C. The differences among the groups were not statistically significant at each visit. The median frequency of sex acts per week reported at each visit was similar among all groups of women (2–3 times per week).

The median and mean plasma viral loads of group A and the matched group B women are given in table 2. The median (likewise the mean) plasma viral load decreased over time among the seroconverters from as high as 5.49 to 4.16 log10 copies/mL from 0 to ⩾18 months (table 2). Figure 1A also shows that the highest values of plasma viral load among women who seroconverted were seen early (range, 5.86–6.50 log10 copies/mL) and the minimum values were seen at ∼18 months (range, 2.86–3.06 log10 copies/mL). Among the seroprevalent women, the viral load values remained stable, with little variability (median range, 4.7–4.9 log10 copies/mL for 0–12 months) (table 2 and figure 1B). The estimated median plasma virus set point among 19 of 31 seroconverters was 4.45 log10 copies/mL (interquartile range, 4.32–5.14 log10 copies/mL). Among 12 seroconverting women in whom at least 2 CD4 cell count measurements were available during the extended follow-up period, the CD4 cell counts after conversion decreased in 8 women from a median of 393 cells/µL (range, 217–566 cells/µL) to 327 cells/µL (range, 181–537 cells/µL) and increased in 4 women from a median of 452 cells/µL (range, 192–682 cells/µL) to 583 cells/µL (range, 363–945 cells/µL).

Table 2
View larger version:
Table 2

Mean and median plasma viral load among women of reproductive age in Malawi.

In the conditional logistic regression analysis, comparing seroconverters to seronegative women, use of injectable hormonal contraceptives increased the odds of HIV-1 acquisition by ∼10-fold after adjusting for other factors (P=.03) (table 3). Comparing women with established HIV-1 infection and HIV-1–seronegative women, use of injectable hormonal contraceptives at baseline was not significantly associated with prevalent HIV-1 infection; only having >1 sexual partner was significantly associated with increased odds of HIV-1 prevalent infection (P=.03) (table 3).

Table 3
View larger version:
Table 3

Association of HIV-1 infection with selected risk factors.

With use of mixed-effect models, we examined the association of injectable hormonal contraceptive use with viral load in the seroconverters and seroprevalent women. These models included an interaction term for use of hormonal contraceptives and time (either from seroconversion or enrollment). In the seroconverters model, the linear regression coefficients (±SE) were 0.49±0.280 (P=.09) for hormonal contraceptive use at the last visit before conversion (compared with nonuse), 0.02±0.053 (P=.70) for time since seroconversion (3-month intervals), and −0.14±0.061 (P=.02) for the interaction of time and hormonal contraceptive use. In the seroprevalent model, the regression coefficients were −0.16±0.215 (P=.46) for injectable hormonal contraceptive use at baseline (compared with nonuse), 0.01±0.025 (P=.71) for time since enrollment (3-month intervals), and 0.01±0.038 (P=.70) for the interaction of time and hormonal contraceptive use. A statistically significant interaction was found between time to seroconversion and use of injectable hormonal contraceptives (P=.02), and no significant interaction was found among seroprevalent women (P=.70). These results are shown in figure 2.

Previous SectionNext Section

Discussion

In this study, we described some of the biological events that accompanied HIV-1 acquisition and examined factors associated with early and established HIV-1 infection. As reported by others, few studies in sub-Saharan Africa have examined the natural history of HIV-1 infection [15]. Even more limited is the availability of data on the natural history of HIV-1 infection among women at low risk as in the general population of women. The few studies reported from Africa were among female sex workers (e.g., in Kenya [13] and South Africa [16]). This study was conducted among relatively healthy, asymptomatic, Malawian women. They were not at high risk according to a median number of sex acts of ∼8 per month, which was similar to estimates from population-based studies in Uganda [17].

As expected [2], the viral load was very high during acute infection (∼1 million copies/mL) and subsequently decreased. After the early months following seroconversion, the viral load became closer to what was observed among women with established HIV-1 infection. The range of viral load in the seroprevalent women during a follow-up of 12 months (4.7–4.9 log10 copies/mL) was comparable to the virus set point in the seroconverters (4.5 log10 copies/mL) and remained stable, similar to reported trends among asymptomatic individuals [18].

The high plasma virus set point in these reproductive age women in Malawi is similar to results from Kenya among sex workers (4.67 log10 copies/mL) and those from India (another HIV-1 subtype C region) among women attending sexually transmitted disease clinics (4.4 log10 copies/mL) [1]. The high virus set point in seroconverters—and, likewise, the high viral load among women with established HIV-1 infection—may suggest that African women maintain these high virus levels during the early stages of HIV-1 acquisition, with no substantial decreases occurring thereafter. A potential biological explanation may relate to infection with HIV-1 subtype C in Malawi [19]. Viruses transmitted during acute infection are usually macrophage tropic, non–syncytium inducing and use mostly CCR5 coreceptors [2]. After early infection and during progression to disease, T cell–tropic viruses (syncytium inducing and mostly using the CXCR4 coreceptor) usually predominate. However, studies of African individuals infected with HIV-1 subtype C have shown lower frequency of the syncytium-inducing phenotype and preponderance of the CCR5 coreceptor among individuals with established HIV-1 infection [20, 21]. This implies that individuals with persistent non–syncytium-inducing viruses may also maintain the high viral load that accompanied seroconversion. This speculation is based on observations from other studies.

Injectable hormonal contraceptive use was significantly associated with HIV-1 acquisition after adjusting for other factors (table 3). Because of the small numbers included in the analysis, however, the 95% CIs were wide. This is similar to observations from other studies [8, 22], and potential biological explanations have been reported [9], including local cervical changes [23]. We note, however, that findings of an association of HIV-1 acquisition with hormonal contraceptive use have not been consistent, and a recent large multisite study reported no association [9]. In our own analysis hormonal contraceptive use was not associated with established HIV-1 infection (table 3).

An interesting finding of our study is the effect modification of use of injectable hormonal contraceptives and time on the linear level of viral load. This interaction was statistically significant among seroconverters but not among seroprevalent cases. This result implies that the effect of injectable hormonal contraceptive use on viral load depends on the time since seroconversion. This is apparent in figure 2: among seroconverters, most events occurred early, and on average, the viral load was very high among hormonal contraceptive users before declining over time. A study from Kenya among sex workers on a much larger sample of seroconverters also found use of injectable hormonal contraceptives at the time of HIV-1 infection to be associated with a higher virus set point [24]. In our study, the viral load trend among hormonal contraceptive nonusers showed limited increase over time. Among seroprevalent cases, and similar to another study from Kenya among postpartum women [25], no change was seen in viral load among hormonal contraceptive users and nonusers (i.e., no interaction because the effect was constant over time). These findings of the interaction of time with hormonal contraceptive use and their effect on viral load may explain the inconsistent associations between hormonal contraceptive use and HIV-1 acquisition reported in multiple studies. Depending on the timing of ascertainment of exposure (hormonal contraceptive use), an association is most likely to be positive only in the early stages (e.g., acute infection) after HIV-1 acquisition. Whether early time in itself is virologically important or whether it is a surrogate for another factor needs further investigation. Such information may have implications for prevention, treatments, and vaccines [26, 27].

In this study, which was mostly descriptive, the limitation of making inferences on a small number of observations is acknowledged. Acute HIV-1 infections are rare events (1%–2% [28]), and timing of seroconversion requires more frequent follow-up and HIV-1 testing. Our results, however, were consistent with findings of other studies; therefore, for purposes of treatment and management of complications, the prognostic utility of these findings could be the same as in other settings. Viral load is the strongest predictor of disease progression [29], and in Africa, CD4 cell count is the means of diagnosis used to initiate antiretroviral treatment.

The comparability of virus set point in this low-risk population with results from high-risk groups within Africa suggests that these biologic values do not substantially vary by risk category. The estimated steady state of virus level (a state of equilibrium resulting from a balance between viral replication and host defense mechanisms) was nonetheless high in our study. Another important observation is that high viral levels in asymptomatic individuals, irrespective of their risk category, should imply acute infection. Individuals with acute infection should be promptly treated for their own benefit and to protect others; these individuals could contribute to ∼50% of cases of HIV-1 transmission [17, 30]. For women of reproductive age, the role of injectable hormonal contraceptive use in HIV-1 infection remains unclear. Studies that target the period of acute HIV-1 infection may be more important in confirming this association and providing biological explanations.

Previous SectionNext Section

Acknowledgments

We thank the women who participated in this study and the study staff of the College of Medicine–Johns Hopkins University Research Project in Blantyre, Malawi.

Financial support. The Bill and Melinda Gates Institute for Population and Reproductive Health and Johns Hopkins University Bloomberg School of Public Health.

Potential conflicts of interest. All authors: no conflicts.

  • Received September 26, 2007.
  • Accepted January 23, 2008.
Previous Section
 

References

  1. ↵
    1. Mehendale SM,
    2. Bollinger RC,
    3. Kulkarni SS,
    4. et al
    . Rapid disease progression in human immunodeficiency virus type 1-infected seroconverters in India. AIDS Res Hum Retroviruses 2002;18:1175-9.
    CrossRefMedlineWeb of Science
  2. ↵
    1. Khan JO,
    2. Walker BD
    . Acute human immunodeficiency virus type 1 infection. N Engl J Med 1998;339:33-9.
    CrossRefMedlineWeb of Science
    1. Kaufmann GR,
    2. Cunningham P,
    3. Kelleher AD,
    4. et al
    . Patterns of viral dynamics during primary human immunodeficiency virus type 1 infection. J Infect Dis 1998;178:1812-5.
    Abstract/FREE Full Text
  4. ↵
    1. Johnston MI,
    2. Fauci AS
    . An HIV vaccine: evolving concepts. N Engl J Med 2007;356:2073-81.
    CrossRefMedlineWeb of Science
  5. ↵
    1. Schacker TW,
    2. Hughes JP,
    3. Shea T,
    4. Coombs RW,
    5. Corey L
    . Biological and virologic characteristics of primary HIV infection. Ann Intern Med 1998;128:613-20.
    Abstract/FREE Full Text
    1. Sterling TR,
    2. Vlahov D,
    3. Astemborski J,
    4. Hoover DR,
    5. Margolick JB,
    6. Quinn TC
    . Initial plasma HIV-1 RNA levels and progression to AIDS in women and men. N Engl J Med 2001;344:720-5.
    CrossRefMedlineWeb of Science
  7. ↵
    1. Lefrere J-J,
    2. Roudot-Thoraval F,
    3. Mariotti M,
    4. et al
    . The risk of disease progression is determined during the first year of human immunodeficiency virus type 1 infection. J Infect Dis 1998;177:1541-8.
    Abstract/FREE Full Text
  8. ↵
    1. Lavreys L,
    2. Baeten JM,
    3. Martin HL,
    4. et al
    . Hormonal contraception and risk of HIV-1 acquisition: results of a 10-year prospective study. AIDS 2004;18:695-7.
    CrossRefMedlineWeb of Science
  9. ↵
    1. Morrison CS,
    2. Richardson BA,
    3. Mmiro F,
    4. et al
    . Hormonal contraception and the risk of HIV acquisition. AIDS 2007;21:85-95.
    Medline
  10. ↵
    1. Taha TE,
    2. Kumwenda K,
    3. Kafulafula G,
    4. et al
    . Intermittent intravaginal antibiotic treatment of bacterial vaginosis in HIV-uninfected and infected women: a randomized clinical trial. PLoS Clin Trials 2007;2:e10.
    CrossRefMedline
  11. ↵
    1. Nugent RP,
    2. Krohn MA,
    3. Hillier SL
    . Reliability of diagnosing bacterial vaginosis is improved by a standardized method of gram stain interpretation. J Clin Microbiol 1991;29:297-301.
    Abstract/FREE Full Text
  12. ↵
    1. Chambers JM,
    2. Hastie TJ
    , editors. Statistical models in S. Pacific Grove, California: Wadsworth & Brooks/Cole Advanced Books and Software; 1992. Generalized additive models; p. 249-308.
  13. ↵
    1. Lavreys L,
    2. Baeten JM,
    3. Chohan V,
    4. et al
    . Higher set point plasma viral load and more-severe acute HIV type 1 (HIV-1) illness predict mortality among high-risk HIV-1-infected African women. Clin Infect Dis 2006;42:1333-1339.
    Abstract/FREE Full Text
  14. ↵
    1. Demidenko E
    . Mixed models: theory and applications. Hoboken, NJ: John Wiley & Sons. 2004.
  15. ↵
    1. Jaffar S,
    2. Grant AD,
    3. Whitworth J,
    4. Smith PG,
    5. Whittle H
    . The natural history of HIV-1 and HIV-2 infections in adults in Africa: a literature review. Bull World Health Organ 2004;82:462-9.
    MedlineWeb of Science
  16. ↵
    1. Grobler J,
    2. Gray CM,
    3. Rademeyer C,
    4. et al
    . Incidence of HIV-1 dual infection and its association with increased viral load set point in a cohort of HIV-1 subtype C-infected female sex workers. J Infect Dis 2004;190:1355-9.
    Abstract/FREE Full Text
  17. ↵
    1. Wawer MJ,
    2. Gray RH,
    3. Sewankambo NK,
    4. et al
    . Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis 2005;191:1403-9.
    Abstract/FREE Full Text
  18. ↵
    1. Garcia F,
    2. Vidal C,
    3. Gatell JM,
    4. Miro JM,
    5. Soriano A,
    6. Pumarola T
    . Viral load in asymptomatic patients with CD4+ lymphocyte counts above 500×106/l. AIDS 1997;11:53-7.
    CrossRefMedlineWeb of Science
  19. ↵
    1. Ping L-H,
    2. Nelson JAE,
    3. Hoffman if,
    4. et al
    . Characterization of V3 sequence heterogeneity in subtype C human immunodeficiency virus type 1 isolates from Malawi: underrepresentation of X4 variants. J Virol 1999;73:6271-81.
    Abstract/FREE Full Text
  20. ↵
    1. Abebe A,
    2. Demissie D,
    3. Goudsmit J,
    4. et al
    . HIV-1 subtype C syncytium-and non-syncytium inducing phenotypes and coreceptor usage among Ethiopian patients with AIDS. AIDS 1999;13:1305-11.
    CrossRefMedlineWeb of Science
  21. ↵
    1. Thomson MM,
    2. Rerez-alvarez L,
    3. Najera R
    . Molecular epidemiology of HIV-1 genetic forms and its significance for vaccine development and therapy. Lancet Infect Dis 2002;2:461-71.
    CrossRefMedlineWeb of Science
  22. ↵
    1. Ungchusak K,
    2. Rehle T,
    3. Thammapornpilap P,
    4. Spiegelman D,
    5. Brinkmann U,
    6. Siraprapasiri T
    . Determinants of HIV-1 infection among female commercial sex workers in northeastern Thailand: results from a longitudinal study. J Acquir Immune Defic Syndr Hum Retrovirol 1996;12:500-7.
    MedlineWeb of Science
  23. ↵
    1. Bulterys M,
    2. Smith D,
    3. Chao A,
    4. Jaffe H
    . Hormonal contraception and incident HIV-1 infection: new insight and continuing challenges. AIDS 2007;21:97-9.
    CrossRefMedlineWeb of Science
  24. ↵
    1. Lavreys L,
    2. Baeten JM,
    3. Kreiss J,
    4. et al
    . Injectable contraceptive use and genital ulcer disease during the early phase of HIV-1 infection increase plasma virus load in women. J Infect Dis 2004;189:303-11.
    Abstract/FREE Full Text
  25. ↵
    1. Richardson BA,
    2. Otieno PA,
    3. Mbori-Ngacha D,
    4. Overbaugh J,
    5. Farquhar C,
    6. John-Stewart GC
    . Hormonal contraception and HIV-1 disease progression among postpartum Kenyan women. AIDS 2007;21:749-53.
    MedlineWeb of Science
  26. ↵
    1. Gupta S,
    2. Jacobson LP,
    3. Margolick JB,
    4. et al
    . Estimating the benefit of an HIV-1 vaccine that reduces viral load set point. J Infect Dis 2007;195:546-50.
    Abstract/FREE Full Text
  27. ↵
    1. Gilbert PB,
    2. DeGruttola VG,
    3. Hudgens MG,
    4. Self SG,
    5. Hammer SM,
    6. Corey L
    . What constitutes efficacy for a human immunodeficiency virus vaccine that ameliorates viremia: issues involving surrogate end points in phase 3 trials. J Infect Dis 2003;188:179-93.
    Abstract/FREE Full Text
  28. ↵
    1. Pilcher CD,
    2. Price MA,
    3. Hoffman if,
    4. et al
    . Frequent detection of acute primary HIV infection in men in Malawi. AIDS 2004;18:517-24.
    CrossRefMedlineWeb of Science
  29. ↵
    1. Mellors JW,
    2. Munoz A,
    3. Giorgi JV,
    4. et al
    . Plasma viral load and CD4+ lymphocytes as prognostic markers of HIV-1 infection. Ann Intern Med 1997;126:946-54.
    Abstract/FREE Full Text
  30. ↵
    1. Cohen MS,
    2. Gay C,
    3. Kashuba ADM,
    4. Blower S,
    5. Paxton L
    . Narrative review: antiretroviral therapy to prevent the sexual transmission of HIV-1. Ann Intern Med 2007;146:591-601.
    Abstract/FREE Full Text
| Table of Contents

This Article

  1. Clin Infect Dis. (2008) 46 (12): 1913-1920. doi: 10.1086/588478
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most