HIV and Parasitic Infection and the Effect of Treatment among Adult Outpatients in Malawi

BRIEF REPORT HIV and Parasitic Infection and the Effect of Treatment among Adult Outpatients in Malawi Mina C. Hosseinipour, 1,2 Sonia Napravnik, 2 George Joaki, 1 Syze Gama, 1 Nyanyiwe Mbeye, 1 Blessings Banda, 1 Francis Martinson, 1 Irving Hoffman, 2 and Myron S. Cohen 2 1 University of North Carolina Project, Lilongwe, Malawi; 2 University of North Carolina at Chapel Hill, Chapel Hill We measured enteric parasitic infection prevalence and the effect of treatment on human immunodeficiency virus (HIV) RNA levels to assess their importance to HIV primary care in resource-limited settings. Adults in Lilongwe, Malawi, were evaluated, treated, and followed-up for parasitic and HIV infections. Of 389 patients, 266 (68%) were HIV infected. Helminth infections were more common in HIV-uninfected than in HIV-infected patients (39% vs. 17%). Among HIV-infected patients, helminth infections were associated with higher CD4 cell counts but not with higher HIV RNA levels. Successful treatment of parasitic infections had no effect on HIV RNA levels. Although common, parasitic infections did not impact HIV RNA levels. Understanding the role played by endemic infections on HIV transmission and disease progression in developing countries is essential to effective provision of care for HIV infection. Although antiretroviral therapy (ART) is being introduced in resource-limited settings, basic medical services remain limited, and preventable, endemic illnesses account for substantial morbidity among HIV-infected patients [1]. Studies of the interaction between HIV infection and intestinal parasitic infections have suggested that HIV-infected patients who are coinfected with helminths experience a shift from Received 4 August 2006; accepted 24 November 2006; electronically published 20 March 2007. Potential conflicts of interest: none reported. Financial support: National Institutes of Health (NIH; grant P30 AI 50410 to fund the Center for AIDS Research program at the University of North Carolina at Chapel Hill); NIH Fogarty Center Career Development Award (National Institute of Allergy and Infectious Diseases grant 1 K01 TW06677-01 to M.C.H.). Reprints or correspondence: Dr. Mina Hosseinipour, Div. of Infectious Diseases, Bioinformatics Bldg., 130 Mason Farm Rd., Campus Box 7030, Chapel Hill, NC 27599-7030 (Mina_Hosseinipour@med.unc.edu). The Journal of Infectious Diseases 2007; 195:1278 82 2007 by the Infectious Diseases Society of America. All rights reserved. 0022-1899/2007/19509-0008$15.00 DOI: 10.1086/513274 a Th1 response (or a type 1 immune system) to a predominantly Th2 response (or a type 2 immune system), as well as an increase in immune system activation [2 4]. Chronic immune activation and the presence of a dominant Th2 cytokine environment may increase the risk of acquiring HIV infection, and concomitant infection with HIV and intestinal parasites may potentiate the virulence of both within the infected patient [3, 4]. Although helminth infections are ubiquitous in developing countries, their effect on the epidemiology of HIV infection, including the risk of HIV transmission and disease progression and management, remains uncertain [3, 4]. Treatment of helminth infections may reduce plasma HIV RNA levels [5], but not all studies confirm this [6 8], and mortality in HIV-infected patients may be similar by helminth infection status [7]. In the present study, we determined the adult prevalence of enteric and urinary parasitic infections in Lilongwe, Malawi, and estimated prevalence differences by HIV infection status. We also studied the effect of parasitic treatment on HIV RNA levels among HIV-infected ART-naive individuals. Participants, materials, and methods. Individuals attending HIV counseling and testing services or outpatient clinics at Kamuzu Central Hospital were eligible. At enrollment, patients received HIV counseling, blood samples were collected for HIV testing and HIV RNA quantification, stool and urine samples were collected for parasitologic testing, and a demographic and clinical history questionnaire was completed. All patients returned 1 week after enrollment to receive treatment for identified infections. HIV-infected and pathogenic parasite coinfected patients were also asked to return 4 weeks after therapy. At that visit, blood was drawn for determination of follow-up HIV RNA levels, stool and urine samples were collected for parasitologic testing, and a questionnaire was completed. Samples were tested for helminths and pathologic and nonpathologic protozoal infections. Schistosomiasis was defined as infection with Schistosoma mansoni or Schistosoma haematobium. Helminth was used to refer to both geohelminths and schistosomiasis. Treatment was based on the guidelines of the Malawi Ministry of Health, including (1) albendazole (single 400-mg dose) for infection with Enterobius vermicularis, Ascaris lumbricoides, Trichuris trichiura, Trichostrongylus species, and hookworm; (2) praziquantel (single 40-mg/kg dose) for infection with Schistosoma mansoni and Schistosoma haematobium; (3) albendazole (400 mg once daily for 3 days) for infection with Strongyloi- 1278 JID 2007:195 (1 May) BRIEF REPORT

Table 1. Spectrum of parasitic infections observed in Lilongwe, Malawi. No. (%) Infection type HIV infected HIV uninfected All 266 123 Any parasitic infection (protozoan or helminth) 99 (37.2) 68 (55.3)!.001 Helminth infections a,b Any helminth infection 46 (17.3) 48 (39.0)!.001 Any geohelminth infection c 37 (13.9) 39 (31.7)!.001 Schistosomiasis 11 (4.1) 13 (10.6).014 Hookworm 27 (10.2) 35 (28.5)!.001 Schistosoma mansoni 11 (4.1) 13 (10.6).014 Ascaris lumbricoides 8 (3.0) 3 (2.4) NC Hymenolepis nana 2 (0.8) 1 (0.8) NC Strongyloides stercoralis 2 (0.8) 0 (0.0) NC Schistosoma haematobium 1 (0.4) 1 (0.8) NC Protozoan infections a,d Any protozoan infection 71 (26.7) 41 (33.3).179 Pathogenic protozoan infections 34 (12.8) 17 (13.8).778 Entamoeba histolytica/dispar 31 (11.7) 13 (10.6).753 Giardia lamblia 3 (1.1) 3 (2.4) NC Blastocystis hominis 1 (0.4) 1 (0.8) NC Nonpathogenic protozoan infections 44 (16.5) 25 (20.3).364 Escherichia coli 41 (15.4) 20 (16.3).831 Iodamoeba buetschlii 2 (0.8) 5 (4.1) NC Chilomastix mesnili 0 (0.0) 1 (0.8) NC Endolimax nana 0 (0.0) 1 (0.8) NC Entamoeba hartmani 1 (0.4) 0 (0.0) NC Any mixed infection 27 (10.2) 23 (18.7).019 NOTE. NC, not calculated (because of small sample size). a Categories are not mutually exclusive. b Enterobius vermicularis, Hymenolepsis diminuta, Taenia saginata, Taenia solium, Trichostrongylus species, and Trichuris trichiura helminth infections were not detected. c Geohelminth category excludes schistosomiasis. d Balantidium coli, Dientamoeba fragilis, and Entamoeba polecki protozoal infections were not detected. des stercoralis; and (4) metronidazole (800 mg 3 times daily) for pathogenic protozoal infections, including with Entamoeba histolytica/dispar, Giardia lamblia, and Blastocystis hominis. Treatment was not given for infections with nonpathogenic protozoa. HIV infection was assessed using 2 rapid HIV tests (Determine [Abbott] and UniGold [Trinity Biotech]). A patient was considered to be HIV infected if both rapid tests were positive. HIV ELISA was performed in the event of discordant rapidtest results. HIV RNA levels were measured using the Roche HIV RNA Amplicor Monitor test (version 1.5; Roche Molecular Diagnostics). CD4 cell counts were determined by flow cytometry (FACSCount system; Becton Dickinson). Stool analysis included a direct smear of formalin-fixed stool and concentrated smear using ethyl acetate sedimentation (Para- Pak Con-Trate System; Meridian Bioscience). Lugol s iodine was added to direct and concentrated specimens. The ZnPVA-fixed specimen was processed using trichrome staining. Two independent technicians reviewed all specimens. Urine was centrifuged, and the sediment was examined for red cells, white cells, helminth eggs, and other pathogens. Confirmed infections for schistosomiasis included identification of Schistosoma haematobium eggs. All specimens were processed in Lilongwe, Malawi, in laboratories certified by the College of American Pathologists and the UK National External Quality Assessment Service. Baseline characteristics between patients with and without HIV infection were compared using the Pearson x 2 test, the Wilcoxon-Mann-Whitney U test, and the t test, as appropriate. A Wilcoxon signed rank sum test was used to contrast the difference in log 10 -transformed HIV RNA levels from baseline to follow-up. SAS (version 9.1, SAS Institute) was used for analyses. Ethics approval was obtained from the University of North Carolina at Chapel Hill Committee on the Protection of the Rights of Human Subjects and from the Malawi National P BRIEF REPORT JID 2007:195 (1 May) 1279

Health Sciences Research Committee, and all participants provided written, informed consent Results. Of 389 participants, 266 (68%) were HIV infected. HIV-infected patients were more likely to be women (76% vs. 49%; P!.001) and older (median age, 32 [interquartile range {IQR}, 26 to 37] vs. 26 [IQR, 21 to 34] years; P!.001). Most patients used tap water ( n p 292; 75%) and a pit latrine ( n p 349; 90%), and these factors were comparable by HIV infection status. Sixty patients (15%) had been previously treated for a parasite, a median of 7 (IQR, 3 to 13) years ago; this was comparable by HIV infection status ( P p.37). At baseline, 167 patients (43%) had evidence of at least 1 parasitic infection (table 1). Of patients with a parasitic infection, 117 (70%) were infected by only 1 parasite, 41 (25%) by 2, 8 (5%) by 3, and 1 (1%) by 4. The most common helminth infection was hookworm. Overall, the HIV-uninfected patients were more likely to have at least 1 parasitic infection, including a helminth, geohelminth, schistosomiasis, hookworm, or mixed infection. However, both pathogenic and nonpathogenic protozoan infections were comparable by HIV infection status. Baseline HIV RNA levels and CD4 cell counts were available for 264 patients. The median HIV RNA level at baseline was 4.92 (IQR, 4.37 to 5.38) log 10 copies/ml, and patients with a parasitic coinfection had levels that were similar to those in patients without a parasitic coinfection (median, 4.89 [IQR, 4.40 to 5.38] and 4.92 [IQR, 4.35 to 5.39] log 10 copies/ml, respectively; P p.95). HIV RNA levels were also comparable and were not significantly different by type of parasitic infection. The median baseline CD4 cell count was 248 (IQR, 127 to 405) cells/mm 3. Patients without a parasitic infection had lower CD4 cell counts than did patients with a parasitic infection (median, 224 [IQR, 110 to 351] and 296 [IQR, 173 to 469] cells/mm 3, respectively; P!.001 ). Lower CD4 cell counts were also observed among patients without a helminth or protozoan infection (for no helminth vs. helminth infection, 235 [IQR, 110 to 392] vs. 320 [IQR, 251 to 558] cells/mm 3 [ P!.001]; for no protozoan vs. protozoan infection, 235 [IQR, 122 to 358] vs. 310 [IQR, 173 to 493] cells/mm 3 [ P p.002]). Of the 73 HIV-infected patients who had at least 1 pathogenic parasitic infection at baseline, 71 received treatment and 63 returned for follow-up evaluation. Of the patients who received treatment, baseline and 4-week posttreatment HIV RNA levels were available for 57 (80%). The median follow-up HIV RNA level among these 57 patients was 4.98 (IQR, 4.49 to 5.48) log 10 copies/ml, compared with their baseline median value of 4.86 (IQR, 4.47 to 5.35) log 10 copies/ml ( P p.89). The median change was 0.04 (IQR, 0.27 to 0.31) log 10 copies/ml, with 76% of patients experiencing a change of!0.50 log 10 copies/ ml in HIV RNA level from baseline. Among those with a baseline HIV RNA level 16.0 log 10 copies/ml ( n p 6), the level decreased by a median of 0.74 (IQR, 0.92 to 0014) log 10 copies/ml. There was no significant change in HIV RNA level by type of parasitic infection (figure 1) or by sex, age, or baseline CD4 cell count. Of the 57 patients with treated parasitic infections and available longitudinal HIV RNA values, 19 (33%) had evidence of at least 1 parasitic infection 4 weeks after receiving treatment. When we restricted the analysis to patients without evidence of parasitic infections at follow-up ( n p 38), we did not observe any differences in HIV RNA levels from baseline to follow-up (median, 4.91 [IQR, 4.48 to 5.41] and 4.97 [IQR, 4.29 to 5.48] log 10 copies/ml, respectively; P p.95). Figure 1. Distributions of HIV RNA level, represented by box plots at baseline (BL) and at follow-up (FU) and stratified by type of parasitic infection, among HIV-infected patients treated for pathogenic parasitic infections in Lilongwe, Malawi. 1280 JID 2007:195 (1 May) BRIEF REPORT

Discussion. As expected, we found that helminth and protozoan infections were common among individuals living in Lilongwe, Malawi. Although we did not observe a statistically significant difference in the prevalence of protozoan infections by HIV infection status, helminth infections were detected less frequently among HIV-infected than HIV-uninfected patients. Furthermore, among HIV-infected patients, the risk of detecting a helminth or protozoan infection decreased with declining CD4 cell counts. In the present study, the HIV-infected patients with a helminth infection had plasma HIV RNA levels similar to those in the patients without a helminth infection, and treatment for helminth and other parasitic infections did not affect HIV RNA levels. These findings are consistent with those of most [6 10], but not all [5, 11], prior studies. One study observed a shortterm increase in HIV RNA levels after helminth eradication, an increase that disappeared with longer follow-up [9]. In the 2 studies that have suggested a possible benefit for HIV RNA level [5, 11], the absolute changes were modest and unlikely to influence HIV disease progression or infectiousness. Subgroup analysis of our data in those with the highest baseline HIV RNA levels, although limited by a small sample size and subject to type II error, demonstrated decreasing HIV RNA levels after treatment, suggesting a potential benefit in those with the highest HIV RNA levels. Also, when interpreting the treatment studies that have been conducted to date, that variable lengths of follow-up have been used should be considered [5 11]. A short duration of follow-up, as in our study, potentially risks inadequate time for the evolution of favorable immune responses, whereas a longer follow-up can be influenced by disease progression. Although treatment for intestinal parasitic infection may decrease HIV replication by decreasing activated lymphocytes, it is possible that any reduction in immune activation is not sufficient to impact HIV replication in the presence of HIV infection, which itself causes persistent immune system activation [12]. Furthermore, although we failed to quantify worm burden, light infestations may not induce immune activation or a Th1/Th2 shift enough to influence HIV replication. Moreover, the relationship between HIV replication and the chronic immune activation present in persons with HIV and helminth infections is more involved than previously believed, in light of recent findings with respect to the effect of immune activation on T regulatory/suppressor cells [12]. It is also possible that an effect on plasma HIV RNA levels was masked by other infectious diseases that are common in resource-limited settings, including malaria and tuberculosis. Despite increasing evidence that enteric parasitic eradication in HIV-infected patients does not directly result in reduced HIV replication, there are other beneficial consequences to treating enteric infections in HIV-infected patients. For example, patients with HIV infection and schistosomiasis may be at increased risk of experiencing liver damage [13]. It is also possible that treatment of enteric or urinary parasitic infections results in reduced genital HIV RNA levels, which reduces the risk of transmitting HIV infection, as observed with sexually transmitted infections [14]. Among HIVuninfected individuals, the immune dysregulation observed with intestinal helminth infection may increase an individual s susceptibility to acquiring HIV infection [3], as may genital schistosomiasis [15]. The lower prevalence of helminth infections we found among HIV-infected patients and among patients with lower CD4 cell counts may represent a real difference in prevalence. However, it is also possible that we underestimated prevalence by relying on fecal smears rather than measuring circulating anodic antigen or circulating cathodic antigen (CCA) levels [13]. Prior studies have observed reduced egg excretion with decreasing CD4 cell counts among HIV-infected patients despite similar worm burdens, as measured on the basis of adult worm CCA levels [13]. As a consequence of a possible impact on general morbidity, the idea of presumptively deworming large populations has received some support [3]. Given the high prevalence of intestinal pathogens in areas of the world hardest hit by HIV infection, it would be ideal if treatment of such coinfections also resulted in a decrease in HIV burden such that HIV disease progression and transmission could be modified. Because treatment of intestinal pathogens is safe and efficacious, the treatment of identified infections is warranted on an individual basis; however, to date the literature does not support that treatment will modify HIV disease progression in antiretroviral-naive individuals. Further work in those at high risk for acquiring HIV infection is warranted to better define the impact of intestinal pathogens on HIV acquisition. References 1. Joint United Nations Program on HIV/AIDS (UNAIDS). 2006 report on the global AIDS epidemic. Geneva: UNAIDS, 2006. 2. Kalinkovich A, Weisman Z, Greenberg Z, et al. Decreased CD4 and increased CD8 counts with T cell activation is associated with chronic helminth infection. Clin Exp Immunol 1998; 114:414 21. 3. Bentwich Z, Kalinkovich A, Weisman Z, Borkow G, Beyers N, Beyers AD. Can eradication of helminthic infections change the face of AIDS and tuberculosis? Immunol Today 1999; 20:485 7. 4. Clerici M, Declich S, Rizzardini G. African enigma: key player in human immunodeficiency virus pathogenesis in developing countries? Clin Diagn Lab Immunol 2001; 8:864 6. 5. Wolday D, Mayaan S, Mariam ZG, et al. Treatment of intestinal worms is associated with decreased HIV plasma viral load. J Acquir Immune Defic Syndr 2002; 31:56 62. 6. 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 43. 7. Brown M, Kizza M, Watera C, et al. Helminth infection is not associated BRIEF REPORT JID 2007:195 (1 May) 1281

with faster progression of HIV disease in coinfected adults in Uganda. J Infect Dis 2004; 190:1869 79. 8. Modjarrad K, Zulu I, Redden DT, et al. Treatment of intestinal helminths does not reduce plasma concentrations of HIV-1 RNA in coinfected Zambian adults. J Infect Dis 2005; 192:1277 83. 9. 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 57. 10. Elliott AM, Mawa PA, Joseph S, et al. Associations between helminth infection and CD4+ T cell count, viral load and cytokine responses in HIV-1-infected Ugandan adults. Trans R Soc Trop Med Hyg 2003; 97: 103 8. 11. 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 61. 12. Deeks SG, Walker BD. The immune response to AIDS virus infection: good, bad, or both? J Clin Invest 2004; 113:808 10. 13. Karanja DM, Colley DG, Nahlen BL, Ouma JH, Secor WE. Studies on schistosomiasis in western Kenya. I. Evidence for immune-facilitated excretion of schistosome eggs from patients with Schistosoma mansoni and human immunodeficiency virus coinfections. Am J Trop Med Hyg 1997; 56:515 21. 14. Cohen MS, Hoffman IF, Royce RA, et al. Reduction of concentration of HIV-1 in semen after treatment of urethritis: implications for prevention of sexual transmission of HIV-1. Lancet 1997; 349:1868 73. 15. Kjetland EF, Ndhlovu PD, Gomo E, et al. Association between genital schistosomiasis and HIV in rural Zimbabwean women. AIDS 2006; 20:593 600. 1282 JID 2007:195 (1 May) BRIEF REPORT