Webb, Emily L. PhD*; Kyosiimire-Lugemwa, Jacqueline BBLT†; Kizito, Dennison MSc†; Nkurunziza, Peter MSc†; Lule, Swaib MBChB†; Muhangi, Lawrence MA†; Muwanga, Moses MBChB‡; Kaleebu, Pontiano PhD†; Elliott, Alison M. MD*,†
In 2010, 34 million people worldwide were estimated to be living with HIV, with 7000 new infections daily.1 Antiretroviral therapy (ART) slows disease progression in people living with HIV (PLHIV) by suppressing viral replication, and may prevent both horizontal and vertical transmission of the disease.2,3 Current World Health Organization guidelines are that ART be initiated in all PLHIV with a CD4 count ≤350 cells per cubic millimeter.4 In addition, the provision of ART for the prevention of mother-to-child transmission is recommended in all pregnant women with HIV, regardless of their symptoms.5 Although there have been dramatic improvements in treatment access over the past decade, 9 million individuals who are eligible to receive ART are still estimated not to do so.1 In parallel with continued efforts to improve ART coverage, other interventions that may impact on HIV disease progression and transmission should be considered.
Sub-Saharan Africa bears a disproportionate burden of the HIV epidemic, with ∼68% of infected individuals estimated to live in the region.6 Coinfection with other pathogens including helminths is common,7,8 and it has been suggested that persistent coinfections may have a detrimental impact on PLHIV. Helminths have profound effects on the host immune system, and these effects may spill over to impact on immune responses to other pathogens.9 Specifically, helminth infection induces a prominent type 2 response profile that inhibits the type 1 response profile needed to combat and control viral antigens such as HIV. It is hypothesized that removal of helminth infections, which can be achieved with cheap and widely available anthelmintic treatments, would reverse these associations.
Observational studies of the relationship between helminths and HIV have found conflicting results,10–13 but a recent systematic review14 and pooled data from 3 randomized controlled trials15–17 have concluded that there was evidence of a beneficial effect of anthelmintic treatment on markers of HIV disease progression. Level of HIV load during pregnancy is a key determinant of vertical transmission risk, and there is some observational evidence for a positive association between helminth infection and increased risk for mother-to-child transmission of HIV18; however, we have previously reported finding no benefit of anthelmintic treatment during pregnancy for vertical HIV transmission in a randomized controlled trial.19 To further elucidate the relationship between helminth coinfection and HIV, we report results from the same trial, investigating the impact of helminths and their treatment during pregnancy on viral load in HIV-infected women 6 weeks after treatment and after delivery.
Study Area and Participants
Participants were HIV-infected pregnant women enrolled in the Entebbe Mother and Baby Study (EMaBS; ISRCTN 32849447),20 a randomized controlled trial of anthelmintic treatment during pregnancy conducted in Entebbe Municipality and Katabi subcounty, a peninsula in Lake Victoria, Uganda, with a high burden of parasitic infections.21 Pregnant women presenting at the government-funded antenatal clinic at Entebbe General Hospital, who were resident in the study area, planning to deliver in the hospital, willing to know their HIV status, and in the second or third trimester of pregnancy, were eligible for inclusion in EMaBS. Exclusion criteria were evidence of possible helminth-induced pathology (hemoglobin <8 g/dL, clinically apparent severe liver disease, diarrhea with blood in stool), history of adverse reaction to anthelmintics, prior enrollment in EMaBS for an earlier pregnancy, or having an abnormal pregnancy as assessed by the midwife.
The primary outcomes for EMaBS were response to immunization and incidence of infectious diseases in the offspring, and vertical transmission of HIV; results for these have been reported elsewhere.19 In this analysis, we examine the effect of anthelmintic treatment on HIV load in the HIV-1–positive mothers in the cohort. The primary outcome for this analysis is viral load during pregnancy, measured at 6 weeks after treatment, with secondary outcome viral load at delivery. Women were eligible for inclusion in this analysis if they were HIV-1 positive and had viral load measured at enrollment and were excluded if known to be taking ART.
Study Design and Procedures
After obtaining written informed consent, demographic and clinical details were recorded, blood samples obtained, and women requested to return with a stool sample. Women were then randomized in a 1:1:1:1 ratio to single-dose albendazole 400 mg or matching placebo and praziquantel 40 mg/kg or matching placebo in a 2 × 2 factorial design. The randomization code was generated by the trial statistician in blocks of 100. Sealed envelopes containing the study intervention were prepared by colleagues at the Medical Research Council Unit in Entebbe with no other involvement in the trial. Treatments were allocated in numerical order by trained interviewer–counselors and taken under observation. All participants and staff were blinded to treatment allocation. After enrollment, women continued to receive standard antenatal care, including hematinics, tetanus immunization, and intermittent presumptive treatment for malaria (with sulfadoxine–pyrimethamine) after the first trimester.
Before the start of the project, a program for prevention of mother-to-child HIV transmission was established at Entebbe Hospital. In accordance with guidelines current at the time, HIV-positive women were counseled and given intrapartum and neonatal single-dose nevirapine for prevention of mother-to-child HIV transmission.22 From March 2005 onward, when ART provision became available, HIV-infected women were offered a repeat full blood count and CD4 T-cell count and referred to the Entebbe Hospital ART services if indicated. Those for whom ART was not yet indicated at enrollment had repeat CD4 counts performed annually (or earlier if clinically indicated), to allow referral when necessary.
Blood samples were collected 6 weeks after enrollment from mothers who had not yet delivered. Blood samples were also collected from the mother after delivery and from cord blood. Blood samples were taken from offspring of HIV-infected women at 6 weeks and 18 months of age to assess their HIV-1 status.
Stool samples were collected ∼6 weeks after delivery to assess effectiveness of the anthelmintic treatment; thereafter, all mothers were treated with both praziquantel and albendazole.
Parasitology and Virology
Stool samples were examined for helminth ova using the Kato-Katz method23 and by charcoal culture for Strongyloides24; 2 Kato-Katz slides were prepared from each sample and examined within 30 minutes for hookworm ova and the following day for other species. Urine examination for Schistosoma haematobium was not conducted due to the low prevalence of this species in the study area.
Hookworm and Schistosoma mansoni infections were classified into low, medium, and high intensities according to World Health Organization guidelines.25,26 Blood samples were examined by a modified Knott method for Mansonella perstans27 and by thick film for malaria parasites. Quality control for Kato-Katz analyses was provided by the Vector Control Division of the Ministry of Health, Uganda, and for malaria parasitology through the UK National External Quality Assessment Schemes.
HIV serology was performed for mothers by rapid test algorithm.28 For viral load assessment, plasma and whole blood cell pellet were separated by centrifugation and stored at −80°C until assays were performed. Plasma HIV load was measured using either Bayer Versant branched DNA assay version 3.0 (Bayer HealthCare, Leverkusen, Germany) or Roche Amplicor HIV-1 RNA Monitor test version 1.5 (Roche Molecular Systems Inc, Basel, Switzerland). Methods used to detect HIV-1 proviral DNA in infants at 6 weeks are described elsewhere.19 Eighteen-month samples were tested for HIV using rapid test algorithm.
Ethical approval was given by the Uganda Virus Research Institute, Uganda National Council for Science and Technology, and the London School of Hygiene and Tropical Medicine.
The sample size for EMaBS was determined for the primary outcomes of the full trial.20 Data were double entered into Microsoft Access (Microsoft Corp, Redmond, WA) and analyzed using Stata version 11 (StataCorp, College Station, TX). Viral load data were log10 transformed for analysis. Baseline characteristics of the mothers included in this study were summarized overall and by randomization arm.
An observational analysis of the association between infection with each helminth and HIV load was conducted by comparing log10 viral load enrollment samples from mothers with and without infection, using linear regression with adjustment for potential confounders (CD4 count, age, and asymptomatic malaria infection at enrollment) to calculate mean differences with 95% confidence intervals (CIs). Likelihood ratio tests were used to calculate P values.
Trial analysis was done by intention to treat. The effects of albendazole treatment and of praziquantel treatment on HIV load in mothers at 6 weeks after treatment and at delivery were examined separately using linear regression, including treatment group as a covariate and adjusting for baseline (enrollment) viral load, and any factors that showed imbalance between treatment arms and that were thought, a priori, likely to impact on viral load. Because the study was designed as a factorial trial, we checked for interaction between albendazole and praziquantel treatments on the 2 viral load outcomes by fitting interaction terms in regression models.
Because not all women were infected with all helminths at enrollment, a planned subgroup analysis was conducted, comparing the effect of albendazole treatment versus placebo in women who had hookworm (the most prevalent helminth treated by albendazole) at enrollment, and the effect of praziquantel treatment versus placebo in women who had S. mansoni at enrollment, using linear regression models adjusted for baseline HIV-1 load and any factors that showed imbalance between treatment arms. Differences between subgroups were examined by fitting interaction terms in regression models. All P values are 2 sided with no adjustment made for multiple comparisons.
Between April 2003 and November 2005, 2507 women were enrolled in EMaBS. Of these, 299 women (12%) tested positive for HIV. Five women on highly active antiretroviral therapy and 30 women for whom no viral load measurement was available at enrollment were excluded from the analysis, leaving 264 women suitable for inclusion (Fig. 1). Women who were HIV infected were on average older, less educated, and more likely to be widowed or divorced, to already have children, and to be infected with asymptomatic malaria, and less likely to be infected with hookworm, compared with HIV-negative women. At enrollment, 67% of the 264 women were infected with at least 1 helminth species, with individual prevalences: hookworm, 39%; M. perstans, 23%; S. mansoni, 18%; Strongyloides stercoralis, 11%; Trichuris trichiura, 9%; and Ascaris lumbricoides, 2%. The prevalences of all helminth species other than hookworm were comparable between HIV-infected and HIV-uninfected women.21 Helminth infections were generally mild among HIV-infected women: of hookworm infections, 92% were classified as light (<1000 eggs/g stool); of S. mansoni infections, 65% were light (<100 eggs/g stool). The characteristics of the 264 women at enrollment were broadly similar between the 4 randomization arms (Table 1), with the exception that women allocated to albendazole had lower mean HIV load and were less likely to have malaria parasitemia. In addition, women allocated to albendazole were more likely to have viral load quantified by the Bayer assay at 6 weeks after treatment. These chance imbalances were taken into account in the analysis. Numbers of serious adverse events are reported elsewhere29 and were distributed evenly between treatment groups.
Associations Between Helminth Infections and HIV-1 Load at Enrollment
The mean (SD) viral load at baseline was 4.09 log10 copies per milliliter (0.93). The mean viral load in women infected with hookworm was 0.22 log10 higher than in uninfected women (Table 2). After adjustment for age, CD4 count, and asymptomatic malaria infection, the difference in viral load was 0.24 log10 (95% CI: 0.01 to 0.47, P = 0.03). There was some evidence that women infected with T. trichiura had higher mean viral loads than those who were uninfected (adjusted mean difference 0.37 log10, CI: 0.00 to 0.74; P = 0.05). There was no evidence of a difference in log10 viral load for any other helminth infection (Table 2). For hookworm and S. mansoni, we found no evidence of an association between infection intensity and viral load.
Effect of Anthelmintic Treatment on HIV-1 Load During Pregnancy and After Delivery
At 6 weeks after enrollment, 41 women had given birth and 1 had miscarried. Of the remaining 222 women, HIV-1 load measurements were obtained for 166 (75%) at a median (interquartile range) of 42 days (41–45) after treatment; mean (SD) viral load was 4.06 log10 (0.91). Based on the single stool samples after delivery, single-dose albendazole and praziquantel treatments were estimated to be 81% and 63% effective in removing hookworm and S. mansoni, respectively.
There was no evidence of interaction between albendazole and praziquantel treatments; therefore, the effects of each treatment were examined independently. At 6 weeks after treatment, women who had received albendazole had lower mean viral load than those who received placebo (mean difference −0.40 log10, 95% CI: −0.68 to −0.13); however, after adjusting for baseline viral load, asymptomatic malaria, and assay used for viral load quantification at 6 weeks after treatment, this difference was reduced to −0.17 log10 (95% CI: −0.36 to 0.01, P = 0.07; Table 3). The effect of albendazole treatment was similar in women with hookworm infection at enrollment compared with those uninfected (interaction P = 0.44).
HIV load was measured in 234 women (89%) after delivery, at a median (interquartile range) time of 105 days (67–133) after treatment; this figure was comparable between treatment groups. Adjusting for viral load and malaria parasitemia at enrollment, the effect of albendazole treatment on viral load had diminished compared with that observed at 6 weeks after treatment during pregnancy (adjusted mean difference −0.11 log10, 95% CI: −0.28 to 0.07, P = 0.23). There were no effects of praziquantel treatment on viral load at either time point, nor any evidence of a differential effect of either anthelmintic treatment by susceptible helminth infection (Table 3).
Effect of Anthelmintic Treatment on HIV Vertical Transmission
Among the 294 highly active antiretroviral therapy–naive women included in this analysis, 16 were lost to follow-up before delivery, 5 had miscarriages, and 9 had stillbirths, leaving information available on 264 live deliveries. We have previously reported that 6-week blood samples were available from 211 infants of whom 39 (18%) were diagnosed with HIV infection and that there were no effects of treatment with albendazole or praziquantel on vertical HIV transmission.19 Two further HIV transmissions were detected in blood samples taken from children at 18 months; inclusion of these transmissions in the analysis had no impact on results.
There are limited data on the relationship between HIV load and helminth infection during pregnancy. We have previously shown, and confirmed in this analysis, that anthelmintic treatment during pregnancy does not significantly reduce the risk for vertical transmission; however, this finding is in the context of three quarters of women taking nevirapine at delivery, and power to detect small- to moderate-sized reductions in risk for vertical transmission was limited. Therefore, by analyzing the effect of anthelmintic treatment on viral load during pregnancy and at delivery, we sought to further understand the relationship between helminths during pregnancy and factors that could lead to vertical transmission of HIV.
We have shown that infection with the soil-transmitted intestinal helminths, hookworm and T. trichiura, during pregnancy is associated with higher HIV load. However, although there was some evidence that albendazole treatment led to a small reduction in viral load 6 weeks after treatment, this was not sustained to delivery. There was no evidence for a role of other helminth species in HIV load modulation, although power to detect associations for species that were rare in our study population, such as A. lumbricoides, was limited; therefore, we cannot discount a role for this species in viral load modulation.
The size of the effect of albendazole treatment was modest, conferring a reduction of 0.17 log10 in viral load. However, modeling studies have shown that even small reductions in HIV load at the population level could translate to reductions in disease incidence; for example, a decrease in viral load of 0.3 log10 has been estimated to decrease HIV transmission by 20% and HIV progression by 25%.30
We found no differential effect of albendazole treatment on HIV load among mothers who were diagnosed with hookworm infection at enrollment compared with those with no detectable hookworm, raising questions about the mechanism of the effect of albendazole. If the effect were mediated by hookworm removal, we would expect the benefit to be greatest in the hookworm-infected women and to persist to delivery (because stool samples at delivery showed a reduction in hookworm prevalence from 45% to 5% among women treated with albendazole29). One possible explanation is the low sensitivity of a single Kato-Katz stool sample31,32: women in the hookworm uninfected group may have had low-intensity hookworm infection or infections with other albendazole-susceptible helminth species. However, data available from a subgroup of women in the study who provided 3 stool samples indicate that sensitivity was high for hookworm. Albendazole treatment was not effective in the treatment of T. trichiura; therefore, the effect of albendazole treatment cannot be mediated through removal of this worm. Another possible explanation relates to the fact that albendazole has a broad spectrum of action and may clear other infections such as malaria,33,34 which may themselves impact on HIV load35; thus, an effect of anthelmintic treatment is seen in both the helminth “infected” and “uninfected” mothers; in this case, weakening of the effect by the time of delivery could be accounted for by the provision of intermittent presumptive treatment for malaria to all women in the latter part of pregnancy.
We found no evidence for a role of S. mansoni or M. perstans in HIV load modification, in contrast to previous randomized trials that have shown a reduction in viral load after treatment of schistosomiasis15 and filariasis,16 and our own observational study that showed a transient increase in viral load after treatment of schistosomiasis.13 For M. perstans, this was not unexpected because the single-dose anthelmintic treatments administered in this trial were not effective in its treatment; however, praziquantel treatment was 63% effective in treating S. mansoni; thus, we would have expected to see any important effects of the removal of this worm.
An association between S. haematobium infection and HIV prevalence has been documented,36,37 suggesting a role for this helminth species in HIV transmission dynamics.38 We could not examine this possibility because S. haematobium was not prevalent in the study area; similar studies conducted in regions where S. haematobium infection is endemic are required to elucidate this relationship.
In conclusion, soil-transmitted helminth coinfections may play a role in HIV load modulation during pregnancy, but we found no evidence for involvement of other helminth species. Treatment with albendazole caused a small decrease in HIV load; however, the observation that this decrease occurred regardless of whether the mother had an albendazole-susceptible worm taken together with the diminution of the effect of albendazole treatment by delivery indicates that any impact of albendazole treatment on HIV load may not be directly mediated through worm removal.
The authors thank all staff and participants of the EMaBS, the midwives of the Entebbe Hospital Maternity Department, the community field team in Entebbe and Katabi, and the staff of the Clinical Diagnostic Services Laboratory at the MRC/UVRI Uganda Research Unit on AIDS.
1. UNAIDS. AIDS at 30: Nations at the Crossroads. Geneva, Switzerland:World Health Organization; 2011.
2. Cohen MS, Chen YQ, McCauley M, et al.. Prevention of HIV-1 infection with early antiretroviral therapy. NEJM. 2011;365:493–505.
3. De Cock KM, Fowler MG, Mercier E, et al.. Prevention of mother-to-child HIV transmission in resource-poor countries: translating research into policy and practice. JAMA. 2000;283:1175–1182.
6. UNAIDS. AIDS Epidemic Update. Geneva, Switzerland: World Health Organization; 2009.
7. Brooker S. Estimating the global distribution and disease burden of intestinal nematode infections: adding up the numbers—a review. Int J Parasitol. 2010;40:1137–1144.
8. Fincham JE, Markus MB, Adams VJ. Could control of soil-transmitted helminthic infection influence the HIV/AIDS pandemic. Acta Trop. 2003;86:315–333.
9. Borkow G, Leng Q, Weisman Z, et al.. Chronic immune activation associated with intestinal helminth infections results in impaired signal transduction and anergy. J Clin Invest. 2000;106:1053–1060.
10. Brown M, Mawa PA, Joseph S, et al.. Treatment of Schistosoma mansoni infection increases helminth-specific type 2 cytokine responses and HIV-1 loads in coinfected Ugandan adults. J Infect Dis. 2005;191:1648–1657.
11. Hosseinipour MC, Napravnik S, Joaki G, et al.. HIV and parasitic infection and the effect of treatment among adult outpatients in Malawi. J Infect Dis. 2007;195:1278–1282.
12. Lawn SD, Karanja DM, Mwinzia P, et al.. The effect of treatment of schistosomiasis on blood plasma HIV-1 RNA concentration in coinfected individuals. AIDS. 2000;14:2437–2443.
13. Brown M, Kizza M, Watera C, et al.. Helminth infection is not associated with faster progression of HIV disease in coinfected adults in Uganda. J Infect Dis. 2004;190:1869–1879.
14. Walson JL, Herrin BR, John-Stewart G. Deworming helminth co-infected individuals for delaying HIV disease progression. Cochrane Database Syst Rev. 2009;(3):CD006419.
15. Kallestrup P, Zinyama R, Gomo E, et al.. Schistosomiasis and HIV-1 infection in rural Zimbabwe: effect of treatment of schistosomiasis on CD4 cell count and plasma HIV-1 RNA load. J Infect Dis. 2005;192:1956–1961.
16. Nielsen NO, Simonsen PE, Dalgaard P, et al.. Effect of diethylcarbamazine on HIV load, CD4%, and CD4/CD8 ratio in HIV-infected adult Tanzanians with or without lymphatic filariasis: randomized double-blind and placebo-controlled cross-over trial. Am J Trop Med Hyg. 2007;77:507–513.
17. Walson JL, Otieno PA, Mbuchi M, et al.. Albendazole treatment of HIV-1 and helminth co-infection: a randomized, double-blind, placebo-controlled trial. AIDS. 2008;22:1601–1609.
18. Gallagher M, Malhotra I, Mungai PL, et al.. The effects of maternal helminth and malaria infections on mother-to-child HIV transmission. AIDS. 2005;19:1849–1855.
19. Webb EL, Mawa PA, Ndibazza J, et al.. Effect of single-dose anthelmintic treatment during pregnancy on an infant's response to immunisation and on susceptibility to infectious diseases in infancy: a randomised, double-blind, placebo-controlled trial. Lancet. 2011;377:52–62.
20. Elliott AM, Kizza M, Quigley MA, et al.. The impact of helminths on the response to immunization and on the incidence of infection and disease in childhood in Uganda: design of a randomized, double-blind, placebo-controlled, factorial trial of deworming interventions delivered in pregnancy and early childhood [ISRCTN32849447]. Clin Trials. 2007;4:42–57.
21. Woodburn PW, Muhangi L, Hillier S, et al.. Risk factors for helminth, malaria, and HIV infection in iregnancy in Entebbe, Uganda. PLoS Negl Trop Dis. 2009;3:e473.
22. Guay LA, Musoke P, Fleming T, et al.. Intrapartum and neonatal single-dose nevirapine compared with zidovudine for prevention of mother-to-child transmission of HIV-1 in Kampala, Uganda: HIVNET 012 randomised trial. Lancet. 1999;354:795–802.
23. Katz N, Chaves A, Pellegrino J. A simple device for quantitative stool thick-smear technique in Schistosomiasis mansoni. Rev Inst Med Trop Sao Paulo. 1972;14:397–400.
24. Friend J. Helminths. In: Collee J, Frase A, Marmion B, et al., eds. Mackie and McCartney Practical Medical Microbiology. Edinburgh, United Kingdom: Churchill Livingstone;1996.
26. World Health Organization. Monitoring Helminth Control Programmes. Guidelines for Monitoring the Impact of Control Programmes Aimed at Reducing Morbidity Caused by Soil-Transmitted Helminths and Schistosomes, With Particular Reference to School-Age Children. Geneva, Switzerland: World Health Organization; 1999.
27. Melrose WD, Turner PF, Pisters P, et al.. An improved Knott's concentration test for the detection of microfilariae. Trans R Soc Trop Med Hyg. 2000;94:176.
28. Muhangi L, Woodburn P, Omara M, et al.. Associations between mild-to-moderate anaemia in pregnancy and helminth, malaria and HIV infection in Entebbe, Uganda. Trans R Soc Trop Med Hyg. 2007;101:899–907.
29. Ndibazza J, Muhangi L, Akishule D, et al.. Effects of deworming during pregnancy on maternal and perinatal outcomes in Entebbe, Uganda: a randomized controlled trial. Clin Infect Dis. 2010;50:531–540.
30. Modjarrad K, Chamot E, Vermund SH. Impact of small reductions in plasma HIV RNA levels on the risk of heterosexual transmission and disease progression. AIDS. 2008;22:2179–2185.
31. Knopp S, Rinaldi L, Khamis IS, et al.. A single FLOTAC is more sensitive than triplicate Kato-Katz for the diagnosis of low-intensity soil-transmitted helminth infections. Trans R Soc Trop Med Hyg. 2009;103:347–354.
32. Utzinger J, Booth M, N'Goran EK, et al.. Relative contribution of day-to-day and intra-specimen variation in faecal egg counts of Schistosoma mansoni before and after treatment with praziquantel. Parasitology. 2001;122(pt 5):537–544.
33. Lacey E. Mode of action of benzimidazoles. Parasitol Today. 1990;6:112–115.
34. Skinner-Adams TS, Davis TM, Manning LS, et al.. The efficacy of benzimidazole drugs against Plasmodium falciparum in vitro. Trans R Soc Trop Med Hyg. 1997;91:580–584.
35. Barnabas RV, Webb EL, Weiss HA, et al.. The role of co-infections in HIV epidemic trajectory and positive prevention: a systematic review and meta-analysis. AIDS. 2011;25:1559–1573.
36. Downs JA, Mguta C, Kaatano GM, et al.. Urogenital schistosomiasis in women of reproductive age in Tanzania's Lake Victoria region. Am J Trop Med Hyg. 2011;84:364–369.
37. Kjetland EF, Ndhlovu PD, Gomo E, et al.. Association between genital schistosomiasis and HIV in rural Zimbabwean women. AIDS. 2006;20:593–600.
38. Gibson LR, Li B, Remold SK. Treating cofactors can reverse the expansion of a primary disease epidemic. BMC Infect Dis. 2010;10:248.
© 2012 Lippincott Williams & Wilkins, Inc.