Rotavirus is an enteric pathogen that affects millions of children globally each year (1). No specific therapy is available for management of rotavirus diarrhea, and its current management involves prevention and or management of dehydration using oral rehydration solution (ORS) and continued feeding. There are limited studies demonstrating a beneficial effect of probiotics in rotavirus-induced diarrhea (2). The rhesus-human reassortant tetravalent rotavirus vaccine (RRV-TV) has been reported to provide a good (60–83%) but not a complete protection against rotavirus-induced severe gastroenteritis in children (3–5). A rotavirus vaccine recently approved for use was removed from the market because of complications, in particular intussusception associated with the vaccine (6,7). Further, the efficacy of this vaccine may be significantly reduced in developing countries, where the prevalence of malnutrition and associated immunodeficiency are high. Thus, there is a clear need to define improved, cost-effective alternatives for treatment of rotavirus diarrhea.
In a recent study, we have shown that immunoglobulins derived from colostrum of cows immunized against human strains of rotavirus (HRV) is effective in reducing stool output and in accelerating early recovery and clearance of virus from stools in infant and children with known rotavirus diarrhea (8). These observations suggest a role of oral anti-HRV antibodies in the management of HRV infections. However, the practical application of this immunotherapy using antibodies obtained from hyperimmunized animals may be limited because of difficulties in large-scale production of antibodies.
In hens, immunoglobulin IgY (yolk immunoglobulin) is passively transferred from their serum to egg yolk, which protects the chicks from various infectious diseases (9). In addition, the amount of IgY obtained from eggs of immunized hen is 18 times higher than can be obtained from serum of immunized rabbits (10). The amount of IgY produced by a hen, at least against HRV as measured by antigen neutralization titer, is 15 times higher for the Wa strain and 120 times higher for MO strains in a year, compared with that produced by a rabbit (11). It is much easier to immunize hens and produce desired antibodies in large scale within a limited time, because egg yolk contains 1% of IgY (9). It is also more hygienic to obtain IgY from egg yolk than from serum or colostrum of animals. It may also be possible to produce antibodies directed against epitopes other than human and bovine immunoglobulins to improve its therapeutic potential. Several animal studies have shown protective efficacy of oral IgY in the prevention of rotavirus diarrhea in mice (12–14), calves (15), and cats (16). Chicken immunoglobulins have also been used to treat other infections such as enterotoxigenic Escherichia coli infections in piglets (17), Campylobacter jejuni infections in chickens (18), and Streptococcus mutans infections (dental caries) in rats (19,20). Although a new concept of human medicine, passive immunization with IgY is an established practice in veterinary medicine. There are thus compelling reasons to hypothesize that this treatment modality is useful in the management of human rotavirus diarrhea. Yolken et al. (21) suggested a decade ago that, theoretically, chicken antibodies could be used as treatment for infants and children with rotavirus diarrhea . However, thus far, only one clinical study has been performed in humans with chicken immunoglobulins, in which IgY was used against S. mutans in adult volunteers (22). We are unaware of any study in which the therapeutic effect of IgY on HRV-induced diarrhea was assessed in humans. We therefore evaluated the efficacy of anti-rotavirus IgY in the treatment of infants and children with gastroenteritis due to rotavirus in a randomized, double-blind, placebo-controlled clinical trial.
PATIENTS AND METHODS
The study was conducted at the Clinical Research and Service Centre of the International Center for Diarrheal Disease Research, Bangladesh (ICDDR,B): Centre for Health and Population Research from January 1997 through June 1999. Male infants and children, aged 4 to 24 months, with history of acute watery diarrhea of less than 48 hours' duration and some degree of dehydration, who were brought to the short-stay ward at ICDDR, B, were eligible for the study. Children with severe malnutrition (marasmus or kwashiorkor), systemic infections requiring antibiotic therapy and a history of bloody mucoid diarrhea were not selected. Eligible children were taken to the study ward and were observed for 4 hours, during which rehydration with oral rehydration solution (ORS) was performed, and their stools were screened for rotavirus antigen by enzyme-linked immunosorbent assay (ELISA) and for cholera by dark-field (DF) microscopy. Those with a positive ELISA result, a negative DF for Vibrio cholerae, and a stool output of more than 20 mL/kg during the observation period were finally enrolled in the study. A written consent was required from parents or guardians for enrollment of each child in the study.
After enrollment in the study, children were randomly assigned to receive either hyperimmunized egg yolk (HEY) or a egg yolk powder from nonimmunized hens (placebo) in a dose of 10 g HEY or the placebo dissolved in 20 mL of water in four equally divided doses for 4 consecutive days.
Children were placed on a “cholera cot,” and pediatric urine collection (PUC) bags were applied for collection of urine and its separation from stools. Ongoing stool loss was replaced by equivalent amount of ORS; however, intravenous solution was used if ORS therapy failed to match ongoing losses (unscheduled intravenous therapy), because of frequent vomiting and purging and reappearance of signs of dehydration. However, the intravenous fluid was discontinued as soon as dehydration was corrected and oral rehydration was resumed. Exclusively breast-fed infants continued to receive mothers' milk, and weaned children received a milk-based formula in addition to breast milk. Non–breast-fed children were given milk formula or semisolid and solid foods appropriate for their age and food habits. The frequency of stool was recorded, and stool and urine outputs as well as ORS and food intake were measured using a balance with a sensitivity of 1 g.
A rectal swab culture was performed to detect Shigella spp., V. cholerae, Salmonella spp., Campylobacter jejuni, and diarrheagenic Escherichia coli, and other vibrios after final enrollment in the study. Stool specimens were also obtained daily for 4 days for detection of rotavirus antigen by ELISA. A venous blood sample for hematocrit, a complete blood count, and serum electrolytes were obtained after the observation period and initial rehydration.
The children remained in the hospital until cessation of diarrhea, which was defined as passage of last watery or loose stool before passage of two consecutive soft or formed stools, or no stool in more than two consecutive 8-hour periods.
The major outcome measures of the intervention includes stool frequency and output (in grams per kilogram per day), ORS intake (in milliliters per kilogram per day), time to fecal clearance of rotavirus, and the duration of diarrhea from initiation of therapy.
Production of Hyperimmunized Egg Yolk and Placebo
Hens, aged 150 days, were immunized intramuscularly with 1 × 107 FCFU of human rotavirus (Wa, RV5, RV3, and ST3) twice at an interval of 1 week using Freund's complete adjuvant. The method used for preparation of antigens, and the protocol used to immunize hens was in accordance with the previously described protocol of Otake et al. (20). Immunoglobulin Y obtained from nonimmunized hens was used for treatment of children in the placebo group.
Purification of Hyperimmunized Egg Yolk and Placebo
The HEY was purified from pooled yolks collected 4 to 22 weeks after immunization from immunized and nonimmunized hens, according to the previously described methods of Hatta et al. (23). In brief, egg yolk was treated with λ-carrageenan to remove other egg yolk proteins. The water-soluble proteins were then filtered trough filter paper. The supernatant obtained was further purified by chromatography on a diethylaminoethyl (DEAE)-Sephacel (Pharmacia, Uppsala, Sweden) column. The absorbed protein was eluted with phosphate buffer. The salting out procedure with anhydrous sodium sulphate was repeated three times, and the eluate was stirred for 30 minutes and centrifuged. The precipitate was dissolved in phosphate buffer to salt out residual contaminating proteins. The filtrate was freeze dried, and IgY powder was obtained.
The IgY content was 14.9% (immune) and 15.9% (placebo) weight of the powder. The protein content of the immune and placebo preparations, as determined by the Kjeldahl method, were 68.7% and 67.1% respectively, with an IgY content of 20%. Lipid, water, and ethanol contents of both IgY preparations were less than 3%, less than 5%, and less than 0.5%, respectively.
A master randomization chart was prepared by an experienced person using random permutated blocks of four, which was administered by a pharmacist holding the lyophilized HEY and nonimmunized IgY (placebo) preparations. Both of the HEY and placebo powders were dispensed in identical bottles that were serially numbered, and the numbers were preassigned to intervention and control IgY. The master random code was prepared and retained by a person who was not involved in the study. The codes were broken only at the end of acquisition and editing of data.
The desired sample size was determined on three major outcome measures: duration of diarrhea, stool output, and duration of fecal excretion of rotavirus. Assuming that the intervention would result in a 30% reduction in all these measures (based on our previous study (8), the sample size was determined to be 37, 22, and 25 in each group for duration of diarrhea, stool output, and fecal excretion of virus, respectively, at 5% significance and 80% power. The largest of these three figures, 37 patients in each group, was taken as the desired sample size. To adjust for an estimated 10% withdrawal rate, the total sample size was 80 in two groups.
Data were entered onto a personal computer, and analyzed by Statistical Package for Social Sciences, Version 8.0 Windows; (SPSS, Chicago, IL, U.S.A.). The treatment groups were compared with regard to their baseline (preintervention) characteristics. Quantitative outcome measures (stool output and ORS intake) were compared using Student's t-test for normally distributed data, and the Mann–Whitney test was used for comparison of data that were not normally distributed. Categorical outcome measures, (e.g. proportion of children with resolution of diarrhea within 4 days of treatment), was compared by χ2 test, and Fisher's exact test was used when appropriate.
Informed consents were obtained from the parents of the children. The study was approved by the local ethical and research review committees of the centre.
Two hundred thirty infants and children were initially screened for presence of rotavirus antigens in stools by ELISA, of whom 140 (71 in the intervention, and 69 in the control group) were positive for rotavirus, met the enrollment criteria, and thus finally enrolled in the study. Sixty-one children (34 in the intervention and 27 in the control groups) were subsequently excluded because of the presence of one or more copathogens in the stool culture (22 E. coli, 12 C. jejuni; 6 V. cholerae, 2 Salmonella spp, 2 Shigella spp., 6 Aeromonas spp., 9 multiple pathogens, and 2 others). Data for the remaining 79 children (37 HEY group; 42 control group) were analyzed. The neutralization titer, as measured in a type neutralization test against the serotype 3 was 1:100,000 compared with 1:1,600,000 for the bovine product used in our earlier rotavirus trial (8).
The demographic and clinical characteristics of the study children were similar in the two groups (Table 1). The mean ± SD of duration of diarrhea before enrollment in the study was 34.7 ± 10.4 and 31.5 ± 11.9 hours for children in the intervention and control groups, respectively (P = NS). Three children (one in treatment group and two in placebo group) needed unscheduled intravenous rehydration at some point during the study period (data not presented). Daily and cumulative stool outputs are shown in Figure 1. Compared with the placebo group, there was a definite trend for lower daily and cumulative stool output among children in the HEY group; however, the difference was statistically significant only on day 1 (P = 0.03). A similar trend of a lower requirement for ORS intake (data not presented) and frequency of stool (Fig. 2) was noted among children in the HEY group, and here again, the difference was statistically significant only on study day 1 (P = 0.008 for ORS intake and 0.03 for frequency of stool). The difference in duration of diarrhea (HEY vs. placebo; 94.5 ± 40.4 vs. 93.3 ± 44.6 hours) was not statistically significant. The probability of clearance of rotavirus from stools was significantly higher in children treated with HEY than in those treated with placebo (P = 0.02). More than 50% of the control children continued to have rotavirus in their stool on day 4, whereas, 74% children in the HEY group no longer had detectable virus (Fig. 3).
There is a great potential for large-scale production of hyperimmunized IgY by immunizing hens with appropriate antigen (9). However, its potential application as a therapeutic tool remains undetermined because of limited information from well-controlled studies. Results of a number of studies indicate therapeutic benefit of the egg yolk antibody IgY in the treatment of enteric infections in animals (15,20,24), and dental plaques (S. mutans) in human volunteers (22). We, for the first time to our knowledge, evaluated the clinical efficacy of immunoglobulin from hen's egg (IgY), specific to HRV, in the management of children with diagnosed rotavirus diarrhea. All the 79 children had relatively severe diarrhea, as indicated by a stool output of more than 20 mL/kg during the 4-hour observation period. Although the disease severity may reduce the possibility of influencing the course of the disease by an intervention, a positive therapeutic response would suggest its clinically useful role.
The stool output and stool frequencies were all less in children treated with HEY. However, these differences were only statistically significant on study day 1. The first day in the hospital is usually the most severe period of the illness, and therefore the beneficial effect of the product on the day signifies clinical relevance. As the children start to recover after rehydration and other measures, there perhaps is less potential for IgY to show a major effect. On study day 4, significantly higher proportion of children in the intervention group cleared virus from their stools. The excretion time of rotavirus from stool was also significantly reduced in HEY treated group compared with the placebo group (2.9 ± 1.5 day vs. 3.7 ± 1.7 day, P = 0.04. The finding remains that in acute rotavirus-induced gastroenteritis, viral excretion time can be influenced therapeutically by HEY and therefore, may have public health significance in limiting the potential spread of infection in hospitals and communities.
Although there was reduction in stool volume on day 1, no significant differences were observed on days 2 through 4 of illness. In our previously reported study, conducted in the same setting, hyperimmune bovine colostrum containing a high titer of anti-rotavirus antibody (1:1,600,000) significantly reduced the volume of stool on all study days and enhanced early clearance of rotavirus from stools (8).
There are several possible explanations for the reduced clinical benefit of IgY in the present study. First, the neutralization titer of the product was much lower than that of IgG in bovine colostrum (8), which may be related to the schedule followed for immunization of hens of only two doses. No additional immunization was used for preparing the HEY for this study. Theoretically, however, very high titers of antibodies can be achieved after a repeated immunization. A basic requirement for successfully treating rotavirus-induced gastroenteritis by passive immunotherapy is the high-titer neutralizing activity of the product, because product with a less potent activity gave negative results (25). The need for a high-titerd antibody preparation for therapy can also be deduced from previous failures to treat rotavirus gastroenteritis with preparation that showed prophylactic activity (26). In a challenge study with human volunteers using bovine anti-E. coli antibodies, Tacket et al. (27) previously demonstrated that using a product with a higher titer yielded better prophylaxis. Although titer was low, the product in our study demonstrated an overall 22% reduction of cumulative stool output and early clearance of virus, and therefore has proven the hypothesis. It is therefore, conceivable that a more powerful preparation with increasing specific titer would provide better clinical benefit in treating rotavirus-induced diarrhea. Future strategy for preparing such a HEY preparation as a therapeutic agent in rotavirus induced diarrhea is thus warranted.
The poor efficacy of the product may also be related to stability in the gastric and small intestinal environment. Although the in vivo stability of IgY is not known, an in vitro study has demonstrated that IgY is less resistant than mammalian IgG to proteolysis (28). In addition, the recovery of IgY was greatly diminished when incubated for more than 1 hour at a pH of 2.0 (9). Unfortunately, in this study we did not measure antibodies in the stool of the patients. These issues may be addressed in future studies in which a suitable formulation of IgY may be used to overcome the possibility of its degradation by gastric acid, such as microencapsulation of the IgY granules or addition of bicarbonate.
We did not determine the serotype(s) of infecting rotavirus in this study, and it is possible that some children may have been infected by a serotype(s) other than the four currently recognized serotypes, against which the product was developed. That this is a strong possibility is reflected in the results of another concurrent study at the ICDDR, B in which 35% of the children with acute diarrhea due to ELISA-positive rotavirus were infected with nontypeable strain(s) (29). However, nontypeable strains of rotavirus were rare at the beginning of our study (i.e., January 1997), and this problem was first noted in 1998. Yet, the possibility that at least a proportion of our study children may have been infected with such strains of rotavirus cannot be ignored. That there was a significant impact of the product on study day 1 argues against such possibilities.
It is also possible that the timing of administration HEY against HRV influences the protective efficacy. This is supported by the observation that the preventive effect of this product in suckling mice decreased with increasing time interval between IgY administration and challenge with HRV. However, administration of IgY within 24 hour of challenge was still effective in decreasing the proportion of mice with HRV-induced diarrhea after the challenge (11). In our study, treatment was initiated after a mean of 33 hours after onset of diarrhea, which may have been too late to influence diarrheal symptoms, because intestinal epithelial damage may have been initiated by that time.
There is limited information on the effective dose of IgY for an optimal outcome. In the only other study on humans that also used chicken antibodies against S. mutans (an organism associated with dental plaque formation), a topical application in the form of a 10-mL mouth rinse preparation containing 0.5 mg/mL of IgY was used (22). The selected dose was based on our earlier successful rotavirus trial with hyperimmune bovine colostrum. Furthermore, we considered the daily production of secretory IgA over the mucosal membranes that has been estimated to be 50 to 100 mg/kg (30). Therefore, the selected dose 10 g HEY (equivalent to 1.5 g of IgY) would be appropriate for children of this study (mean weight, 7.5 kg).
The preparation of HEY against human strains of rotavirus is simple, relatively inexpensive, safe, and hygienic. As high as 40 g of immunoglobulin can be produced from one hen per year (9) indicating a great potential for large scale and cost-effective production of IgY from hens' eggs. Because the effect of passive immunotherapy is dose dependent (25), the findings of the present study indicate that better result in rotavirus-induced diarrhea may be obtained using a preparation with a high-titer neutralizing activity.
The authors thank to Dr. Lennart Svensson at SMI, Sweden, for supplying the virus used for immunization and Ms. Shireen Ali, Ms. Afia Khatun, and Ms. Azmira Begum, and the nurses of the Clinical Study Ward of the Clinical Research and Service Centre of ICDDR, B for their hard work.
This study was supported by the ICDDR,B: Centre for Health and Population Research, the Swedish Medical Research Council, The European Community (Bio 4-CT97-2106), the Swedish Society of Medicine; the Jerring Foundation, Stockholm, Sweden; and the Samariten Foundation, Stockholm, Sweden. ICDDR, B is supported by countries and agencies that share their concern for the health problem in developing countries.
1. Cook SM, Glass RI, LeBaron CW, Ho MS. Global seasonality of rotavirus
infections. Bull World Heath Orga. 1990; 68:171–7.
2. Rautanen T, Isolauri E, Salo E, Vesikari T. Management of acute diarrhoea with low osmolarity oral rehydration solutions and Lactobacillus strain GG. Arch Dis Child. 1998; 79:157–60.
3. Perez-Schael I, Guntinas MJ, Perez M, et al. Efficacy of the rhesus-rotavirus
-based quadrivalent vaccine in infants and young children in Venezuela. N Engl J Med. 1998; 337:1181–7.
4. Joensuu J, Koskenniemi E, Pang XL, et al. Randomised placebo-controlled trial of rhesus-human reassortant rotavirus
vaccine for prevention of severe rotavirus
gastroenteritis. Lancet. 1997; 350:1205–9.
5. Lanata CF, Midthun K, Black RE, et al. Saftey, immunogenicity, and protective efficacy of one and three doses of the tetravalent rhesus rotavirus
vaccine in infants in Lima, Peru. J Infect Dis. 1996; 174:268–75.
6. Barnes G. Intussusception and rotavirus
vaccine. J Pediatr Gastroenterol Nutr. 1999; 29:375.
7. Gay N, Ramsey M, Waight P. Rotavirus
vaccination and intussusception. Lancet. 1999; 354:956.
8. Sarker SA, Casswall TH, Mahalanabis D, et al. Successful treatment of rotavirus
diarrhoea in children with immunoglobulin from hyperimmunized bovine colostrum. Pediatr Infect Dis J. 1998; 17:1149–54.
9. Hatta H, Ozeki M, Tsuda K. Egg yolk antibody IgY and its application. In: Yamamoto T, Juneja LR, Hatta H, et al., eds. Hen Eggs: Their Basic and Applied Science. New York: CRC Press, Inc.; 1997:151–78.
10. Gottstein B, Hemmeler E. Egg yolk immunoglobulin Y as an alternative antibody in the serology of echinococcosis. Z Parasitenkd. 1985; 71:273–6.
11. Hatta H, Tsuda K, Akachi S, Kim M, Yamamoto T. Productivity and some properties of egg yolk antibody (IgY) against human rotavirus
compared with rabbit IgG. Biosci Biotechnol Biochem. 1993; 57:450–4.
12. Bartz CR, Conklin RH, Tunstall CB, Steel JH. Prevention of murine rotavirus
infection with chicken egg yolk immunoglobulins. J Infect Dis. 1980; 42:439–41.
13. Ebina T. Prophylaxis of rotavirus
gastroenteritis using immunoglobin. Arch Virol Suppl 1996; 12:217–23.
14. Kuroki M, Ikimori Y, Yokoyama H, Peralta RE, Ieatlo Jr., FC, Kodana Y. Passive protection against bovine rotavirus
-induced diarrhoea in murine model by specific immunoglobulins from chicken egg yolk. Vet Microbiol. 1993; 37:135–46.
15. Kuroki M, Ohta M, Ikemori Y, Peralta RE, Yokoyama H, Kodana Y. Passive protection against bovine rotavirus
in calves by specific immunoglobulins from chicken egg yolk. Arch Virol. 1994; 138:143–8.
16. Hiraga C, Kodama Y, Sugiyama T, Ichikama Y. Prevention of human rotavirus
infection with chicken egg yolk immunoglobulins containing rotavirus
antibody in cat. J Jpn Assoc Infect Dis. 1990; 64:118–23.
17. Yokoyama H, Peralta RC, Diaz R, Sendor S, Ikemori Y, Kodama Y. Passive protective effect of chicken egg yolk immunoglobulins against experimental enterotoxigenic Escherichia coli infection in neonatal piglets. Infect Immun. 1992; 60:998–1007.
18. Tsubokura K, Berndtson E, Bogstedt A, et al. Oral administration of antibodies as prophylaxis and therapy in Campylobacter jejuni-infected chickens. Clin Exp Immunol. 1997; 108:451–5.
19. Hamada S, Horikoshi T, Minami T, et al. Oral passive immunization against dental caries in rats by use of hen egg yolk antibodies specific for cell-associated glucosyltransferase of Streptococcus mutans. Infect Immun. 1991; 59:4161–7.
20. Otake S, Nishihara Y, Makimura M, et al. Protection of rats against dental caries by passive immunization with hen-egg-yolk antibody (IgY). J Dent Res. 1991; 70:162–6.
21. Yolken RH, Leister F, Wee SB, Miskuff R, Vanderfecht S. Antibodies to rotavirus
in chickens' egg: a potential source for antiviral immunoglobulins suitable for human consumption. Pediatrics. 1988; 81:291–5.
22. Hatta H, Tsuda K, Ozeki M, et al. Passive immunization against dental plaque formation in humans: effect of a mouth rinse containing egg yolk antibodies (IgY) specific to Streptococcus mutans. Caries Res. 1997; 31:268–74.
23. Hatta H, Kim M, Yamamoto T. A novel isolation method for hen egg yolk antibody, “IgY”. Agric Biol Chem. 1990; 54:2531–5.
24. Ebina T, Tsukada K, Umezu K, et al. Gastroenteritis in suckling mice caused by human rotavirus
can be prevented with egg yolk immunoglobulin (IgY) and treated with a protein-bound polysaccharide preparation (PSK). Microbiol Immunol. 1990; 34 (7):617–29.
25. Hilpert H, Brussow H, Mietens C, Sidoti J, Lerner L, Werchau H. Use of bovine milk concentrate containing antibody to rotavirus
to treat rotavirus
gastroenteritis in infants. J Infect Dis. 1987; 156:158–66.
26. Ebina T, Sato A, Umezu K, et al. Prevention of ratvirus infection by oral administration of cow colostrum containing antibody against human rotavirus
antibody. Med Microbiol Immunol. 1985; 174:177–85.
27. Tacket C, Losonsky G, Link H, et al. Protection by milk immunoglobulin concentrate against oral challenge with enterotoxigenic Escherichia coli. N Engl J Med. 1988; 318:1240–3.
28. Shimizu M, Nagashima H, Sano K, et al. Molecular stability of chicken and rabbit immunoglobulin G. Biosci Biotech Biochem. 1992; 56:270–4.
29. Sarker SA, Fuchs GJ, Azin T, et al. High prevalence of untypeable rotavirus
in Bangladesh [abstract]. In:Current Trends of Gastroenterology and Nutrition
. Proceedings of the Sixth Congress of Asian Pan-Pacific Society of Paediatric Gastroenterology and Nutrition 1999 July 1–4;Singapore: Department of Pediatrics, National University Hospital, Singapore, 1999:38.
30. Conley ME, Delacroix DL. Intravascular and mucosal immunoglobulin A: two separate but related systems of immune defense. Ann Intern Med. 1987; 106:892–9.