We explored response to single-dose praziquantel therapy in a cohort of 33 women with Schistosoma haematobium infection in rural Mwanza, Tanzania. Women with S. haematobium infection confirmed both by eggs in urine and by polymerase chain reaction (PCR) received single-dose praziquantel and treatment of concomitant sexually transmitted infections. Macroscopic cervical abnormalities were also quantified. After 6 months, microscopically detectable egg excretion was eliminated, but 8 of 33 women (24%) were persistently positive for S. haematobium by PCR, and 11 (33%) had cervical abnormalities potentially attributable to schistosomiasis. This suggests that praziquantel treatment more frequently than every 6 months may be necessary for complete elimination of the parasite and prevention of genital tissue pathology. This aggressive therapy may in turn play a key role decreasing HIV susceptibility in millions of people living in regions in which S. haematobium is endemic.
Nous avons exploré la réponse au traitement par une dose unique de praziquantel dans une cohorte de 33 femmes présentant une infection à Schistosoma haematobium dans les régions rurales de Mwanza, en Tanzanie. Les femmes avec une infection de S. haematobium confirmée à la fois par des œufs dans les urines et par la réaction en chaîne de la polymérase (PCR) ont reçu une dose unique praziquantel et un traitement pour les infections sexuellement transmissibles concomitantes. Les anomalies cervicales macroscopiques ont été également quantifiées. Après six mois, l'excrétion d’œufs détectables par microscopie a été éliminée, mais 8 des 33 femmes (24%) sont restées constamment positives pour S. haematobium par PCR et 11 (33%) présentaient des anomalies cervicales potentiellement attribuables à la schistosomiase. Ceci suggère qu'un traitement par le praziquantel plus fréquemment que tous les six mois serait nécessaire pour l’élimination complète du parasite et pour la prévention de la pathologie des tissus génitaux. Cette thérapie agressive peut à son tour jouer un rôle clé dans la diminution de la sensibilité au VIH chez des millions de personnes vivant dans des régions où S. haematobium est endémique.
Hemos explorado la respuesta a una terapia con una dosis única de praziquantel en un estudio de cohortes de 33 mujeres con infección por Schistosoma haematobium en zonas rurales de Mwanza, Tanzania. Las mujeres con infección por S. haematobium, confirmadas tanto por presencia de huevos en la orina como mediante PCR, recibieron una dosis única de praziquantel y tratamiento para infecciones de transmisión sexual concomitantes. Las anormalidad macroscópicas del cérvix también se cuantificaron. Tras seis meses, se eliminaron las excreciones de huevos detectables mediante microscopía, pero 8 de 33 mujeres (24%) continuaban siendo positivas para S. haematobium mediante PCR y 11 (33%) tenían anormalidades en el cérvix potencialmente atribuibles a la esquistosomiasis. Esto sugiere que podría ser necesario un tratamiento con praziquantel con una frecuencia mayor a cada seis meses para eliminar completamente el parásito y prevenir la patología del tejido genital. Esta terapia agresiva podría a su vez jugar un papel clave en la disminución de la susceptibilidad al VIH en millones de personas viviendo en regiones en las cuales S. haematobium es endémico.
Schistosoma haematobium is a parasitic worm infection acquired through contact with infested freshwater that affects an estimated 112 million people in sub-Saharan Africa (Van der Werf et al. 2003). Adult S. haematobium worms live for years in host pelvic venous plexi and lay hundreds of eggs per day (Maguire 2010). The schistosome miracidium within the egg secretes proteolytic enzymes through the egg's shell that facilitate the migration of the egg through the vessel wall and into the surrounding tissues towards the mucosal tissue and lumen of the genitourinary tract (Maguire 2010). Some eggs reach the lumen and are excreted in urine or genital secretions (Cheever et al. 1994), but a significant fraction become trapped in the tissues of urogenital organs, where they cause acute and chronic inflammation. Histopathological and autopsy studies of women with S. haematobium infection demonstrate eggs throughout the upper and lower genital tract (Edington et al. 1975; Poggensee & Feldmeier 2001; Kjetland et al. 2012). S. haematobium infection in women has been associated with HIV infection in cross-sectional studies (Kjetland et al. 2006a; Ndhlovu et al. 2007; Downs et al. 2011) and is postulated to increase HIV susceptibility both by damaging the mucosal integrity of the female genital tract and by causing local mucosal and systemic immune responses (Feldmeier et al. 1994; Kjetland et al. 2006a; Chenine et al. 2008; Hotez et al. 2009; Obeng et al. 2008; Siddappa et al. 2011).
The WHO recommends empiric antischistosome treatment with praziquantel for at-risk individuals in sub-Saharan Africa to reduce disease morbidity and with the possibility of preventing HIV (World Health Organization 2010, 2012). In highest prevalence areas (>50%), annual single-dose praziquantel is recommended for both children and adults (World Health Organization 2006). Self-report of even one antischistosome treatment before the age of 20 years in girls has been associated with significantly lower rates of genital manifestations of schistosome infection such as sandy patches and contact bleeding (Kjetland et al. 2008).
Two longitudinal studies have suggested that single-dose praziquantel significantly reduces urinary egg excretion but does not completely cure S. haematobium infection, particularly its gynaecological manifestations. The smaller study reported regression of genital sandy patches in 2/4 women re-examined 9 weeks after treatment but persistence in the other two (Richter et al. 1996). In the larger study of 338 women seen after 12 months, praziquantel treatment was not associated with improvements in gynaecological abnormalities or contact bleeding (Kjetland et al. 2006b). In order to explore and extend these findings, we added molecular diagnostics to investigate response to single-dose praziquantel in a cohort of women with S. haematobium infection in rural Tanzania.
We conducted a six-month cohort study in which women with Schistosoma haematobium infection were recruited from the S. haematobium-endemic villages of Lubiri and Nyamilama, treated with praziquantel and retested after 6 months. This was a proof-of-concept study to explore the utility of PCR for monitoring treatment response in urogenital schistosomiasis. We enrolled women only for this preliminary study because our work was incorporated into an ongoing women's health programme administered by the Tanzanian Ministry of Health.
As previously described, women aged 18–50 had been screened for cervical cancer using visual inspection with acetic acid (the standard of care in Tanzania) in partnership with a governmental outreach programme (Downs et al. 2011). We enrolled women who screened negative for cervical cancer and who had S. haematobium infection confirmed by visualisation of eggs in urine plus were PCR-positive for schistosome DNA in urine. We additionally examined cervical smears for eggs and tested cervicovaginal lavages for schistosome DNA. A nurse collected socio-demographic data using a structured questionnaire in Kiswahili, the local language.
Women received free observed treatment with single-dose praziquantel (40mg/kg) at enrolment. We included pregnant and lactating women as praziquantel is recommended for these groups by WHO (2006). Women and their sexual partners also received free treatment for the sexually transmitted infections (STIs) for which testing can be routinely carried out in Tanzania, including syphilis, trichomoniasis, Chlamydia, gonorrhoea, candidiasis and bacterial vaginosis if indicated. Women participated in an educational seminar about schistosomiasis, STIs and possible associations with HIV infection on the day of study enrolment. Study participants were re-evaluated for S. haematobium and gynaecological infections at the six-month follow-up visit and again received indicated treatment without cost.
Gynaecological examinations were performed at enrolment and 6 months to document any macroscopic cervical abnormalities including sandy patches, irregular blood vessels, contact bleeding, condylomata and abnormal discharge. Prior to any sample collection, the cervix was visually inspected using a flashlight and magnifying glass. Macroscopic abnormalities were identified by the physician and recorded and drawn to scale on a standardised data collection form. Abnormal discharge, defined as discharge of any colour other than white or white discharge that was thick and/or copious, was also noted and described on the form. The physician was blinded to the prior examination findings at the time of follow-up data collection.
The first sample collected was a wet preparation for candidiasis, trichomoniasis and bacterial vaginosis, followed by an endocervical swab for Chlamydia and gonorrhoea as previously described (Downs et al. 2011, 2012). Next, cervical smears were taken from abnormal-appearing regions if present and otherwise from the transformation zone. Lastly, cervicovaginal lavage samples were prepared using four millilitres of normal saline to wash the face of the cervix three times.
Single urine samples at enrolment and follow-up were collected between 10 am and 2 pm, filtered and examined microscopically for schistosome ova by a trained microscopist. Serum was tested by rapid plasma reagin (RPR), with Treponema pallidum particle agglutination (TPPA) testing performed on positive samples for diagnosis of syphilis.
DNA isolation and PCR was performed as described previously (Obeng et al. 2008). Briefly, DNA was isolated from well-mixed pre- and post-treatment urine and cervicovaginal lavages using QIAamp Tissue kit spin columns (QIAgen, Hilden, Germany). Quantitative real-time PCR testing was performed using Schistosoma genus-specific primers and detection probe. Phocine Herpes Virus 1 (PhHV-1) was used as an internal control in each sample along with PhHV-1-specific primers and detection probe to detect any inhibition. Amplification, detection and analysis were performed with the CFX96 real-time detection system (Bio-Rad laboratories). The PCR output from this system consists of a cycle threshold (Ct) value, representing the amplification cycle in which the level of fluorescent signal exceeds the background fluorescence and reflecting the parasite-specific DNA load in the sample tested. Negative and positive control samples are included in each amplification run.
Continuous variables were summarised by median and interquartile range (IQR), and categorical variables were summarised by frequency and percentage. Proportions were compared using Fisher's exact test and continuous variables by the Wilcoxon rank-sum test. Pre/post-treatment values were compared with the Wilcoxon matched-pairs signed-rank test. Two-sided hypotheses/tests were assumed for all confidence intervals and P-values. Data were analysed using Stata version 11 (College Station, Texas).
Written informed consent was obtained by a trained study nurse fluent in the local language who read the consent form to all study participants. Women who decided to participate signed their names or made their mark/thumbprint (if illiterate) on the consent form. Ethical permission was granted by Bugando Medical Centre, the Tanzanian National Institute for Medical Research and Weill Cornell Medical College.
Between November 2009 and May 2010, we enrolled 39 women with S. haematobium infection detected by both microscopy and PCR. Of these, 33 (85%) were re-examined a median of 188 (IQR, 183–188) days after treatment and included in the analysis. There were no significant differences between age or number of urinary ova between women who were lost to follow-up (n = 6) and those who were not.
Socio-demographic information is detailed in Table 1. Patients were young, with a median age of 22 (20–28) years, and 6 patients were aged 18–19. The large majority (90%) were married. Sexual risk factors were assessed indirectly. Nearly one-third of women had children with at least two different men, and more than half of women stated that their husbands had children with another woman during the marriage. Ten women (30%) reported going to bed hungry in the past month because their household did not have enough food, a measure of food insecurity (Nelson et al. 1998; Tappis et al. 2013).
Table 1. Demographic characteristics of 33 adult women (>18 years) treated for urogenital schistosomiasis and followed up after 6 months
Non-missing data were included for each calculation.
Age in years—median (interquartile range)
Married, n (%)
Number of children
Number of fathers of a woman's children
Husband had children outside of marriage during the marriage, n (%)
Went to bed hungry in past month, n (%)
Reported no prior treatment for schistosomiasis, n (%)
Microscopy and PCR pre- and post-treatment
At enrolment, all women had urinary ova with a median of 8 (4–53) per 10 ml. Three women additionally had viable S. haematobium ova in cervical smears. All 33 women had positive urine PCR results. Urine Ct values ranged from 20.6 to 36.3 (median 27.8). Cervicovaginal lavage PCRs were positive in 9 women (27%), ranging from 19.9 to 35.1 (median 28.3). Two of the three women with eggs in cervical smears had PCR-positive lavages.
Six months after treatment, ova were no longer detected in any urine or cervical sample. Eight women (24%) still had detectable schistosome DNA by PCR with Ct values <35 (six urines and two lavages, Figure 1). In women with persistently positive PCR findings, the DNA quantity remained approximately unchanged. Urine PCRs became zero in the two women with persistently positive lavages.
Gynaecological abnormalities pre- and post-treatment
STIs diagnosed included: four Chlamydia infections (12%) and one each (3%) of syphilis, gonorrhoea, trichomoniasis and multiple cervical Condylomata. At enrolment, 9/33 women (27%) were negative for STIs but had at least one macroscopic gynaecological abnormality that, in the absence of STIs or malignancy, could be consistent with Schistosoma involvement of the genital tract (Table 2). These included sandy patches, contact bleeding, convoluted blood vessels and abnormal discharge.
Table 2. Findings in 33 adult women (>18 years) treated for Schistosoma haematobium infection and retested after 6 months
Six months post-treatment
Urine PCR (Ct value)
Positive urine PCR, n (%)
Interquartile range (IQR)
Urinary ova (per 10 ml)
Positive urinary ova, n (%)
Lavage PCR (Ct value)
Positive lavage PCR, n (%)
Positive cervical smear ova, n (%)
Convoluted blood vessels, n (%)
Sandy patches, n (%)
Contact bleeding, n (%)
Abnormal discharge, n (%)
Any gynaecological abnormality potentially consistent with genital S. haematobium infection, n (%)
After treatment, the same macroscopic abnormalities persisted in 7 of these 9 women, and four others had new abnormalities that could be consistent with cervical Schistosoma infection in the absence of STIs and malignancy. Thus, we observed, in total, 11 women (33%) with gynaecological abnormalities that were possibly consistent with schistosomiasis 6 months after praziquantel treatment. Four of these 11 (36%) were persistently PCR-positive (3 urines and one lavage) compared with 4 (18%) of the 22 who had no gynaecological abnormalities (P = 0.25). Ages were not significantly different between groups of women who remained positive by PCR or who had persistent gynaecological abnormalities and those who did not.
In total, 10 women had evidence of S. haematobium involvement of the genital tract by eggs or PCR at the time of enrolment, and 6 months after treatment, 2 (20%) remained PCR-positive in cervicovaginal lavages. Schistosome DNA additionally was detectable in urine of 6 other women at 6 months. Of the 25 women who were negative for both schistosome eggs and DNA at 6 months, 7 (28%) had gynaecological abnormalities in the absence of STIs or cervical cancer, suggestive of S. haematobium-induced pathology.
In this cohort of 33 young adult women with Schistosoma haematobium infection who received single-dose praziquantel therapy, Schistosoma DNA was detectable 6 months later in the genital tract of 2/10 women with initial genital tract involvement and in urine of 6 additional women. Moreover, nearly one-third had ongoing cervical pathology potentially attributable to schistosome infection. Our findings suggest that, after single-dose praziquantel, which is WHO's current recommendation for schistosomiasis (World Health Organization 2013), parasite remnants and genital tract tissue pathology persist in a substantial minority of adult women. Other groups have advocated higher and/or repeated praziquantel doses (Alonso et al. 2006; Gryseels et al. 2006; Centers for Disease Control 2013). Our work supports this concept and raises serious concern given recent findings that schistosomiasis may be a risk factor for HIV acquisition (Kjetland et al. 2006a; Downs et al. 2011, 2012).
Schistosoma real-time PCR has been evaluated as an indicator of genital S. haematobium infection (Kjetland et al. 2009) but not as a marker of effectiveness of praziquantel treatment. Our work documents that Schistosoma DNA was detectable 6 months after treatment in 8/33 women whose negative egg counts appeared to indicate resolved infection. Previous studies in Zimbabwe and Malawi have similarly reported unresolved macroscopic gynaecological pathology (sandy patches, blood vessel abnormalities and/or contact bleeding) in women treated for genital S. haematobium infection and any STIs and re-examined up to a year later (Richter et al. 1996; Kjetland et al. 2006b). Our work now substantiates these clinical reports by showing that, after praziquantel treatment, a sizeable percentage of women have not only ongoing cervical pathology but also detectable parasite DNA in urine and cervicovaginal lavage fluid.
The persistence of gynaecological lesions potentially caused by S. haematobium suggests that S. haematobium-induced pathology may be irreversible. Liver fibrosis due to S. mansoni infection is often permanent even after parasite death (Richter 2000). Similarly, gynaecological lesions are believed to begin as inflammatory tissue reactions to migrating ova (Smith & Christie 1986; Helling-Giese et al. 1996) which, over time, become fibrotic ‘sandy patches’ (Smith & Christie 1986; Kjetland et al. 2008) that may not be curable. In support of this hypothesis, women from an S. haematobium-endemic region who reported receiving praziquantel before age 20 were less likely to have gynaecological contact bleeding and sandy patches than women treated after age 20 (Kjetland et al. 2008). Thus, regular treatment beginning in childhood appears optimal to prevent potentially irreversible genital lesions that may predispose to HIV infection (Kjetland et al. 2008; Hotez et al. 2009; World Health Organization 2010). Establishment of the most effective treatments for genital S. haematobium infection remains an urgent research priority.
A limitation of this study was our inability to determine whether Schistosoma DNA detected at 6 months was due to re-infection, sample contamination by semen of infected sexual partners or by stool, partial treatment failure, post-treatment maturation of juvenile worms which are less susceptible to praziquantel (Utzinger et al. 2003) or lingering ova and parasite products in tissue and blood (Wichmann et al. 2009). We believe that the latter three explanations are the most likely sources of DNA in urine samples, which were clean catch and showed no ova. Cervicovaginal lavages are inherently not sterile and could more easily have been contaminated. Second, neither testing for herpes simplex virus and human papillomavirus infections, nor exclusion of premalignant lesions was possible. Lastly, the sensitivities of cervical smears (Poggensee et al. 2001), PCR of cervicovaginal lavages (particularly in women older than age 25) (Kjetland et al. 2009) and single rather than double urine samples are all suboptimal for detection of S. haematobium infection. Lower sensitivity would bias our study towards finding less residual infection in patients with confirmed S. haematobium at the study outset. Thus, our detection of persistent schistosome DNA in 8/33 women may be low. Either way, our findings suggest that treatment may need to be repeated more frequently than the WHO recommends (annually) and more frequently than even the six-month duration of this study in order to maximise parasite elimination and minimise tissue pathology.
In conclusion, our results suggest that single-dose praziquantel does not completely eliminate S. haematobium DNA in a substantial portion of women. Moreover, one dose of praziquantel appears not to reverse cervical pathology. Our data supports using praziquantel early and often in girls and women living in endemic areas. Given that S. haematobium may be a risk factor for HIV acquisition, this low-cost ($0.32) (Hotez et al. 2009) intervention may have the potential not only to lessen schistosome-related morbidity but also to prevent new HIV infections among millions of women at risk in Africa.
We thank Eric A.T. Brienan and Pystje Hoekstra for performing the PCR assays and the Tanzanian women for their enthusiastic participation in this study. This work was supported by the 2009 Merle A. Sande/Pfizer Fellowship Award in International Infectious Diseases, which is awarded annually by the Infectious Diseases Society of America Education & Research Foundation and by the National Foundation for Infectious Diseases. This work was also supported by grants T32 HS000066 from the Agency for Healthcare Research and Quality, T32 AI007613 from the National Institute for Allergy and Infectious Diseases and CTSC grant UL1-RR024996.