The impact of zinc supplementation on Schistosoma mansoni ... - Nature

33±37 All rights reserved 0954±3007/97 $12.00

European Journal of Clinical Nutrition (1997) 51,

ß 1997 Stockton Press.

The impact of zinc supplementation on Schistosoma mansoni reinfection rate and intensities: A randomized, controlled trial among rural Zimbabwean schoolchildren H Friis1,3, P Ndhlovu2, T Mduluza2, K Kaondera2, B Sandstro¨m3, KF Michaelsen3, BJ Vennervald1 and NO Christensen1 1

Danish Bilharziasis Laboratory, Denmark; 2 Blair Research Laboratory, Ministry of Health, Zimbabwe; and 3 Research Department of Human Nutrition, Royal Veterinary and Agricultural University, Denmark

Objectives: To assess the effect of zinc supplementation on susceptibility to S. mansoni reinfections among schoolchildren. Design: Randomized, double-blind, placebo-controlled trial. Setting and Subjects: 313 rural Zimbabwean schoolchildren (144 boys and 169 girls), 11–17 y). Interventions: Supplementation with zinc (30 or 50 mg) or placebo on schooldays for 12 months. Due to drought, a food programme was in operation during the last eight months of the study. Outcome measures: S. mansoni and S. haematobium reinfection rates and intensities. Results: There was no difference in reinfection rates between the zinc and placebo groups (25 vs 29%, P 0.46). However, the median intensity of S. mansoni reinfection, although low in both groups, was significantly lower in the zinc than in the placebo group (7 vs 13 eggs per gram of faeces, P 0.048). No difference in either S. haematobium reinfection rates or intensities were seen. Conclusions: Zinc supplementation reduced the intensity of S. mansoni reinfections. Although the intensities of reinfection were very low, the finding probably reflects a biological effect of zinc that could be of public health importance in settings with higher transmission. Sponsorship: Danish International Development Assistance. Descriptors: Randomized, controlled trial; zinc deficiency; zinc supplementation; S. mansoni; S. haematobium; reinfection; schoolchildren; Zimbabwe

Introduction Interactions of nutrition and infection have received considerable interest for many years (Scrimshaw et al, 1968). While most infections have the potential to impair the nutritional status of the host, single nutrient deficiencies may either aggravate or alleviate a specific infection (Scrimshaw et al, 1968). Accordingly, repletion of single nutrient deficiencies may either hinder or facilitate control of a specific infection, for which reason knowledge about the synergistic or antagonistic nature of such interactions is important prior to commencement of food or single nutrient supplementation programmes. The effects of micronutrient deficiencies on susceptibility to specific infections have recently been focused on (Bundy et al, 1987; Tomkins et al, 1989). In particular, zinc deficiency could be of importance for susceptibility to infections, since it is essential for numerous immunefunctions (Gross et al, 1980; Beisel, 1982) and may affect parasite kinetics (Bundy et al, 1987). However, studies are impeded by the lack of a satisfactory measure of zinc status. Correspondence: Dr H Friis, c/o Danish Bilharziasis Laboratory, Jaegersborg Alle´ 1D, DK-2920 Charlottenlund, Denmark. Received 10 February 1996; revised 11 September 1996; accepted 1 October 1996

In order to assess the role of zinc deficiency in susceptibility to S. mansoni infections, a randomized controlled zinc supplementation trial, based on the classic schistosomiasis reinfection study, was carried out among a study population in rural Zimbabwe with growth-limiting zinc deficiency (Friis, 1996).

Subjects and methods Study area and population A subsistence farming community in Sabi Valley, 65 km east of Chiredzi in south-eastern Zimbabwe, with low nutritional status and high prevalence of S. mansoni infection among schoolchildren, was identified. The climate in the area was hot and dry, with an annual rainfall below 400 mm, and the vegetation dry savanna woodland and scrub. The staple crops were maize and millet. Cattle and goats were kept, but mainly used as cash products. The Mkwasini River, a tributary of the Save River, courses between Mareya and Muteo Primary Schools, from which subjects were recruited for this study. Children attending grade three to six were eligible for inclusion in the study. Accordingly, 198 (84.3%) of 235 and 115 (92.0%) of 125 children from grade three to six at Muteo and Mareya Primary School, respectively, were included in the study.

Zinc supplementation and s. mansoni reinfections

H Friis et al

34

Study design and interventions After baseline examinations, all children found infected with S. haematobium and/or S. mansoni were treated with a single dose of praziquantel (40 mg per kg body weight). The study design was a randomized, placebo-controlled, double-blind zinc supplementation study. Allocation to supplementation with either enterocoated (Vernix enterosolubile) zinc sulphate tablets or identical-looking placebo tablets (The Hospital Pharmacy, Copenhagen County, Denmark) was done by simple randomization. Children weighing below 29.5 kg were given tablets containing 30 mg of elementary zinc, and children weighing 29.5 kg or above were given 50 mg. Each child was allocated one container, labelled with class, name and study number, for each term, with tablets in known excess. The containers were kept at the school, and the tablets were administered by the schoolteachers on all school days throughout the one year study period. The number of tablets taken by each child was calculated by subtracting the number of tablets remaining after completion of the 12-month follow-up examination from the number doled out. Compliance was expressed as the number of tablets taken as percentage of the maximum number of tablets that could be taken. The code was not broken before the data entry and cleaning were completed. Due to an alarming food situation in the study area, a school-based food supplementation programme was introduced after four months of the study. During the last eight months of the study, the programme provided the children with a meal of maize, beans and fish on all schooldays, as previously described (Friis, et al. 1996). S. mansoni and S. haematobium A urine and a stool sample were collected from each child between 9 am and noon at three different days at the baseline examination and at two different days after six weeks as well as after three, six, nine, and 12 months. A single 50 mg cellophane faecal thick smear (Peters et al, 1980) was prepared from each stool sample and examined for eggs of S. mansoni after at least 24 h clearing. The mean egg output was expressed as eggs per gram faeces (epg). From each urine sample filtration of 10 ml of stirred urine was done (Peters et al, 1976) using Nytrel (TM) filters. The filters were examined for eggs of S. haematobium and mean egg output was expressed as eggs per 10 ml of urine (ep 10 ml). Assessment of cure rate was done by parasitological reexamination six weeks and three months after the baseline treatment. A child was considered cured if found egg positive at the baseline examination, but egg negative at the six-week parasitological reexamination. However, if found egg positive six weeks after treatment, but not after three months, then the infection was considered to have been inactive at the six-week examination, and the child to have been cured. A child not found infected at baseline was only considered uninfected if eggs were still not found at the six-week as well as at the three-month examination. If found primarily uninfected or successfully treated, as defined above, but found infected at any of the later followup examinations, then the child was considered to have been reinfected. The intensity of reinfection was calculated for each schistosome species separately as the mean egg output of the first parasitological follow-up examination revealing schistosome eggs and all the following examinations.

Biochemistry Blood samples were taken from the cubital vein between 8 am and noon as part of the baseline examinations. Serum was collected and the concentration of zinc determined by atomic absorption spectrophotometry (Perkin Elmer Model 360, Perkin Elmer, Norwalk, CT, USA). Anthropometry Height and weight were related to references and heightfor-age (HA) and weight-for-age (WA) Z-scores computed using the nutritional anthropometry module of Epi Info version 5 software (Dean et al, 1990). Based on the measurements of right mid upper arm circumference and triceps skinfold thickness, arm muscle area-for-age (AMAA) and arm fat area-for-age (AFAA) Z-scores were computed based on the NHANES data (Frisancho, 1990). Permissions Permission to conduct the study was obtained from the Medical Research Council of Zimbabwe. The study protocol was also approved by the Danish Central Medical Ethics Committee. The headmasters and teachers of the two schools, and the local Environmental Health Technician (EHT) were informed about the project and agreed to participate. The village headmen and parents were informed about the study and gave their consent. Statistical analysis Normality of the variables was assessed by the WilkShapiro/rankit plot procedure computed in Statistix Version 4.0 (Analytical Software, Tallahassee, USA). The Chisquare test was used to test for differences in proportions between the two supplementation groups, whereas the Wilcoxon two-sample test was used to test for differences in medians. Results Of the 313 children in the cohort, 291 (93.0%)were found infected with either S. mansoni (82.1%) or S. haematobium (70.3%) at the baseline examination. From the six-week and three-month re-examination 280 were found cured and 14 primarily uninfected according to the definitions given above. Of these 294 (93.9%) thus considered uninfected, 261 (88.8%) were followed-up parasitologically for the full 12-month study period. The 52 children either not found uninfected or lost to full parasitological follow-up were equally distributed between the zinc and placebo groups (27 vs 25, P 0.81). Baseline equivalence was achieved through the simple randomization carried out as seen from Table 1, where baseline characteristics for the 261 children followed-up for reinfections are shown. Reinfections Among the 261 children followed-up for 12 months, 71 (27.2%) children became S. mansoni reinfected and 54 (20.7%) became S. haematobium reinfected. There was no difference in S. mansoni reinfection rate between children allocated to zinc and placebo (33 (25.2%) vs 38 (29.2%), P 0.46). However, the intensity of reinfection was significantly lower among those receiving zinc than among those receiving placebo (median 7 (interquartile range (IQR) 4–20) vs 13 (IQR 7–28) epg, P 0.048) (Table 2). For S. haematobium, neither the reinfection rate (23 (17.7%) vs 31 (23.7%), P 0.23) nor median intensity (1 (IQR 0.2–2) vs 1 (IQR 0.2–3) ep 10 ml, P 0.46) were


H Friis et al

Table 1

35

Baseline characteristics of children allocated to placebo or zinc groups Placebo group (n 130)

Female/male Age in yearsa Height-for-age Z-scorea Weight-for-age Z-scorea Arm muscle area-for-age Z-scorea Arm fat area-for-age Z-scorea Serum zinc (mmol/L)a S. mansoni Prevalence (%)b Intensity (epg)c S. haematobium Prevalence (%)b Intensity (ep 10 ml)c

67/63 11.1 (1.8) 1.18 (0.95) 1.26 (0.73) 0.88 (0.57) 1.02 (0.22) 12.0 (2.3)

Zinc group (n 131) 74/57 10.9 (1.9) 1.12 (0.88) 1.27 (0.68) 0.92 (0.57) 1.04 (0.22) 11.9 (2.5)

7 7 7 7

7 7 7 7

83.9 (77.4–90.3) 66 (27–140)

86.3 (80.3–92.2) 50 (27–100)

73.9 (66.2–81.5) 9 (4–35)

71.0 (63.1–78.9) 9 (3–27)

a

Mean (s.d.). Prevalence (95% confidence interval). Median (interquartile range) eggs per gram of faeces (S. mansoni) and eggs per 10 ml of urine (S. haematobium) among infected.

b c

Table 2

S. mansoni and S. haematobium reinfection rate and intensity in the placebo and zinc group

Schistosome reinfection S. mansoni Rate (%)a Intensityb S. haematobium Rate (%)a Intensityc a b c

Placebo group (n 130)

Zinc group (n 131)

P

29.2 (21.3–37.2) 13 (7–27)

25.2 (17.7–32.7) 7 (4–20)

0.46 0.048

17.7 (11.0–24.3) 1 (0.2–3)

23.7 (16.3–31.0) 1 (0.2–2)

0.23 0.46

Reinfection rate (95% confidence interval). Median (interquartile range) eggs per gram of faeces among 71 S. mansoni reinfected children. Median (interquartile range) eggs per 10 ml of urine among 54 S. haematobium reinfected children.

Table 3 Baseline parasitological data of children according to S. mansoni reinfection status Schistosome infection S. mansoni Prevalence (%)a Intensity (epg)b S. haematobium Prevalence (%)a Intensity (ep 10 ml)b

Reinfected (n 71)

Not reinfected (n 190)

84.5 (75.9–93.1) 50 (25–152)

85.3 (80.2–90.4) 53 (27–115)

74.7 (64.3–85.0) 9 (3–22)

71.6 (65.1–78.0) 10 (4–41)

a

Prevalence (95% confidence interval). Median (interquartile range) eggs per gram of faeces (S. mansoni) and eggs per 10 ml of urine (S. haematobium) among infected. b

significantly different between the placebo and zinc group (Table 3). As seen from Table 3, baseline prevalence and intensity of S. mansoni and S. haematobium infection were similar among the 71 children later found S. mansoni reinfected compared to the 190 children remaining uninfected. Compliance Among the 261 children in whom reinfections could be assessed, the median compliance was 47% (IQR 31–68), corresponding to a tablet every other schoolday, or every four days. There was no difference in compliance between children in the zinc and placebo group (median 47% (IQR 31–70) vs 48% (IQR 32–65)). Compliance could not be assessed separately for the last nine months where the reinfections took place.

Discussion Associations between S. mansoni infection and low serum concentrations of zinc have been reported from crosssectional hospital-based (Mikhail et al, 1982a; Mikhail et al, 1982b) and field studies (Prasad et al, 1963). However, serum zinc is not a satisfactory measure of zinc status (Golden, 1989), and even if a valid and causal association exists, the direction of the cause and effect relationship remains to be determined. These methodological problems can only be overcome by carrying out a randomized, controlled zinc supplementation trial in a study population with suboptimal zinc status. The presence of growth-limiting zinc deficiency in the study population has been documented (Friis et al., 1996). In the randomized, placebo-controlled, double-blind study presented here, zinc supplementation had neither effect on the rate of S. mansoni reinfection, nor the rate or intensity of S. haematobium reinfection. For both S. mansoni and S. haematobium the intensities of reinfections were very low. Nevertheless, the intensities of S. mansoni reinfection were significantly lower among the children allocated to zinc supplementation than among children allocated to placebo. Baseline equivalence between the two groups was achieved through simple randomization, whereby known as well as unknown potential confounding factors should be controlled for (Table 1). In particular, it is not conceivable that the slightly higher median intensity of S. mansoni infection in the placebo group compared to the zinc group reflects a difference in predisposition to infection, since baseline prevalence and intensity of infection were similar


H Friis et al

36

among those later found S. mansoni reinfected and those not found reinfected (Table 3). Although egg output is considered to reflect wormburden, and as such used as a measure of intensity of infection, the reduction in egg output found in this study could be due to an effect of zinc on stool volume or parasite kinetics. Anorexia is a prominent symptom in zinc deficiency, and zinc supplementation may stimulate appetite in zinc deficient individuals. Nonetheless, an increased stool volume is not a likely explanation of the difference in egg output since the weight gain was similar in the two groups over the 12 months study period (Friis et al., 1996). Effective formation of granulomata has been associated with facilitated excretion of the eggs (Doenhoff et al, 1986), and zinc deficiency has been shown to impair granuloma formation (Nawar et al, 1992). Hence, if this is the case in human infections, then it could have led to an underestimation of the true effect of zinc. If the reduction in egg output among zinc supplemented children was mediated by an effect of zinc on fecundity of the worms, it would be equally interesting from a public health point of view, since the egg is the key parasite pathogenetic factor. However, no information is available concerning the possible role of zinc in fecundity of the schistosome worms. Moderate zinc deficiency has been shown to reduce spontaneous motor activity in prepubertal rhesus monkeys (Golub et al, 1994). Thus, a potential bias not accounted for in this study is that zinc supplementation could have led to an increase in exploratory activity, with an increase in water contact and reinfections. Again, if this was the case, then the effect of zinc supplementation would be underestimated. Only a crude measure of compliance for the full 12-month study period was obtained. This may have led to a misclassification of compliance with respect to the last nine months where the reinfections took place, which could explain why no dose-response effect was demonstrated in the zinc group. While Mansour et al (1983) found that intraperitoneal zinc to apparently zinc sufficient hamsters reduced the worm burden following exposure to S. mansoni cercariae, no animal studies on the effect of zinc deficiency on susceptibility to schistosome infection have been published. However, zinc deficiency has been shown to increase the susceptibility to other helminths or impair the expulsion (Fenwick et al, 1990b; Fenwick et al, 1990a; Laubach, 1990) in laboratory animal studies. Whether the effect of zinc on intensity of S. mansoni reinfections found in our study was immune-mediated or due to high tissue concentrations of zinc reducing the infectivity of the parasite (Dresden et al, 1974; Asch et al, 1977) is not clear, but a very low dietary intake of zinc in the study population (Friis et al., 1996) supports the former explanation. The role of zinc nutrition in susceptibility to schistosome infection has, to the best of our knowledge, not previously been studied in a longitudinal, human study. The effect of zinc plus multimicronutrient supplementation, compared to multimicronutrient supplementation without zinc, on parasitic reinfection was studied in a randomized, controlled trial among Guatemalan primary schoolchildren (Grazioso et al 1993). No effects were seen on either total helminth or specific helminth (Ascaris lumbricoides, Trichuris trichiura, Ancylostoma duodenale, Enterobius vermicularis, Hymenolepis nana) reinfection rates, but the power of the study was very low due to small sample size and low reinfection rates.

The fact that an effect of zinc was seen in our study on intensities of S. mansoni reinfection, but not on rates, may not be surprising. Exposure is obviously the major determinant of risk of reinfection, whereas nutritional factors are more likely to modulate than to eliminate susceptibility to reinfection. Nevertheless, with the shift in strategy of schistosomiasis control which took place in the early eighties, from control of transmission to control of morbidity (World Health Organization, 1983), a reduction in intensities could be of public health interest. Although the effect of zinc supplementation found in our study is not likely to be of clinical significance, it probably reflects a biological effect of zinc on susceptibility to S. mansoni infection, that may be of public health importance in settings with higher transmission. Acknowledgements—We thank the schoolchildren and teachers at Mareya and Muteo Primary Schools for their participation and cooperation. Furthermore, we thank the staff at the De Beers Laboratory, Chiredzi, and at the Blair Research Laboratory for assistance in conducting the study, and the Secretary for Health, Zimbabwe, for permission to publish. Thanks also to Mette Knop for review of the manuscript.

References Asch HL, Dresden MH (1977). Schistosoma mansoni: effects of zinc on cercarial and shistosomule viability. J Parasitol. 63: 80–86. Beisel WR (1982). Single nutrients and immunity. Am J Clin Nutr. 35: 417–468. Bundy DAP, Golden MHN (1987). The impact of host nutrition on gastrointestinal helminth population. Parasitology. 95: 623–635. Dean AG, Dean JA, Burton AH, Dicker RC (1990). Epi Info, Version 5: a word processing, database, and statistics program for epidemiology on microcomputers. USD, Incorporated, Stone Mountain, Georgia. Doenhoff MJ et al (1986). The schistosome egg granuloma: immunopathology in the cause of host protection or parasite survival? Trans R Soc Trop Med Hyg. 80: 503–514. Dresden MH, Edlin EM (1974). Schistosoma mansoni: Effects of some cations on the proteolytic enzymes of cercaria. Exp Parasitol. 35: 299– 303. Fenwick PK et al (1990a). Zinc deficiency and zinc repletion: effect on the response of rats to infection with Trichinella spiralis. Am J Clin Nutr. 52: 166–172. Fenwick PK et al (1990b). Zinc deprivation and zinc repletion: effect on the response of rats to infection with Strongyloides ratti. Am J Clin Nutr. 52: 173–177. Friis H et al (1996). The impact of zinc supplementation on growth and body composition: A randomised, controlled trial among rural Zimbabwean schoolchildren. Eur J Clin Nutr. : 118/96. Frisancho AR (1990). Anthropometric standards for the assessment of growth and nutritional status. The University of Michigan Press: Ann Arbor. Golden MHN (1989). The diagnosis of zinc deficiency. In: Mills CF (ed). Zinc in Human Biology. Springer-Verlag: Berlin, pp 323–333. Golub MS et al (1994). Modulation of behavioral performance of prepubertal monkeys by moderate dietary zinc deprivation. Am J Clin Nutr. 60: 238–243. Grazioso CF et al (1993). The effect of zinc supplementation on parasitic reinfestation of Guatemalan schoolchildren. Am J Clin Nutr. 57: 673– 678. Gross RL, Newberne PM (1980). Role of nutrition in immunological function. Physiol Rev. 60: 188–302. Laubach HE (1990). Effect of dietary zinc on larval burdens, tissue eosinophil numbers, and lysophospholipase activity of Ascaris suum infected mice. Acta Trop. 47: 205–211. Mansour MM, Mikhail MM, Guirgis NI (1983). Effect of zinc supplementation on S. mansoni-infected hamsters. Ann Trop Med Parasitol. 77: 517–521. Mikhail MM, Mansour MM (1982a). The interaction of zinc and vitamin A in human schistosomiasis. Eur J Clin Inv. 12: 345–350. Mikhail MM, Mansour MM (1982b). Complications of human schistosomiasis and their effect on levels of plasma copper, zinc and serum vitamin A. Hum Nut Clin Nutr. 36C: 289–296.


H Friis et al

Nawar O et al (1992). The effect of zinc deficiency on granuloma formation, liver fibrosis, and antibody responses in experimental schistosomiasis. Am J Trop Med Hyg. 47: 383–389. Peters PA et al (1976). Field studies of a rapid, accurate means of quantifying Schistosoma haematobium eggs in urine samples. Bull WHO. 54: 159–162. Peters PA, El Alamy MA, Warren KS, Mahmoud AAF (1980). Quick Kato-smear for field quantification of Schistosoma mansoni eggs. Am J Trop Med Hyg. 29: 217–219.

Prasad AS et al (1963). Zinc metabolism in patients with the syndrome of iron deficiency anemia, hepatosplenomegaly, dwarfism, and hypogonadism. J Lab Clin Med. 61 (4): 537–549. Scrimshaw NS, Taylor CE, Gordon JE (1968). Interactions of nutrition and infection. World Health Organization, Geneva. Tomkins A, Watson F (1989). Malnutrition and infection. A review. United Nations, Geneva. World Health Organization (1983). The strategy of morbidity reduction in schistosomiasis. WHO/SCHISTO/83.68.

37