Portal de Periódicos da CAPES

Options for the Prevention of Rotavirus Disease Other Than Vaccination

2008; Lippincott Williams & Wilkins; Volume: 46; Issue: S2 Linguagem: Inglês

10.1097/mpg.0b013e31816f79b0

1536-4801

Jacek Mrukowicz, Hania Szajewska, Timo Vesikari,

Infant Health and Development

High-quality clinical practice guidelines and recommendations are characterised by a full review of all of the available options for the prevention of a particular disease. Several interventions other than vaccination for the prevention of rotavirus (RV) gastroenteritis (GE) and acute diarrhoea in children have been studied in clinical trials. These interventions include improved sanitation and hand hygiene (1–7), passive immunisation (8–13), probiotics (14–23), and promotion of breast-feeding (15,24–34). In this article, we have restricted our discussion to the studies that have considered RV infection or disease as a separate outcome measure. METHODOLOGY An extensive search of PubMed and the Cochrane Controlled Trials Registry and Cochrane Database of Systematic Reviews was performed in February 2006. The initial search was limited to randomised controlled trials (RCTs), involving children ages 0 to 18 years, published in English. This identified 47 articles. Screening the abstracts for relevance yielded 34 articles to be included in the final analysis. Any additional sources were identified from the references of these articles or from the authors' own libraries. GENERAL MEASURES FOR THE PREVENTION OF RVGE Avoiding Contact With Infected Individuals Avoiding contact with infected individuals minimises the risk for developing RVGE in theory; however, there have been no RCTs designed to assess the efficacy of this method for the prevention of RVGE in infants and young children. This approach has limited practical application because RV infection is ubiquitous and most children are infected before they are 5 years old (35–38), because some infections are asymptomatic but still infectious, and because RV is often shed in the faeces for prolonged periods following both symptomatic and asymptomatic infection (39,40). RV is shed by 50% of infected children 1 day before symptoms, by 31% of children 2 days before symptoms, and by 9% of children 3 to 5 days before symptoms (38,41,42). Children are at increased risk for contracting RV in day care centres and during a hospital stay (43). Despite the increased risk for infections, day care centres are considered to be a valuable and popular child care option for working parents, and admission of a child to hospital is often unavoidable. Reducing the Number of Children Sharing Paediatric Hospital Wards Multiple-bed rooms in paediatric wards have been associated with an increased risk for nosocomial acute viral diarrhoea in young children (44), but there have been no RCTs to assess the efficacy of implementing a policy of 1 or 2 patients per hospital room for the prevention of nosocomial RVGE. Because the incidence of acute viral diarrhoea in small rooms is still high, a reduction in the number of children per ward seems unlikely to relieve this problem (44). IMPROVED SANITATION AND HAND HYGIENE IN DEVELOPED COUNTRIES Use of Disinfectants to Reduce Rotavirus Environmental Spread and Improve Hand Hygiene Rotavirus can survive on human hands for at least 4 hours, other parts of the human body for several days (45), and for up to 10 days on dry nonporous surfaces in a low-humidity environment (46). Environmental spread of RV can be minimised with the use of certain disinfectants, such as hypochlorite or sodium dichloroisocyanurate; 70% ethanol solution also inactivates the virus (47). Alcohol-based hand-hygiene products eliminate RV from skin; 70% isopropanol and 70% ethanol are more effective than medicated or nonmedicated soap in reducing RV titres on fingertips (48,49); and gel containing 60% ethanol reduces the infectivity titre of RV by greater than 3 log (50). However, the efficacy of these interventions for the prevention of RVGE has not been studied in RCTs. Sustained long-term compliance with rigorous hand hygiene can be problematic in practice, limiting its long-term effectiveness. This conclusion has been reported in several studies (average 40% compliance among health care personnel; range 5%–81%) (51). The efficacy of multifaceted interventional programmes to promote hand washing and hygienic practices for the prevention of diarrhoea in children attending day care centres has been examined in 6 cluster RCTs (1–6). These studies were heterogeneous according to interventions used, definitions of diarrhoea, and methods for measuring outcomes. None of these studies reported RVGE as a separate outcome measure. There are no published RCTs to address the efficacy of interventions to promote hand or environmental hygiene in hospitals, or comparing different hand-hygiene products for their effectiveness in reducing nosocomial diarrhoea and nosocomial RVGE. One study adopting a "before and after" design showed that extensive, long-term, repeated campaigns (5 intervention periods of 2–4 months each over a 2-year period) to promote hand hygiene on 2 wards of a single paediatric hospital resulted in an increase in overall hand-hygiene compliance from 62% to 81% at the end of the study (7). The rate of nosocomial RVGE reduced correspondingly from 5.9 to 2.2 episodes per 1000 discharged patients (7). Soap and water was the most common method for practicing hand hygiene in this study. The use of alcohol-based hand gel increased from 4% at the first observation period to 29% at the final observation period (7). It should be noted, however, that a before-and-after design is prone to bias and therefore the results (including the magnitude of effect) should be interpreted with caution. Although effective, these measures did not eliminate nosocomial RVGE. PASSIVE IMMUNISATION FOR THE PREVENTION OF RVGE Passive immunisation with orally administered antibodies against RV has been evaluated in infants and toddlers (1–36 months of age) in 6 relatively small randomised, double-blind, placebo-controlled trials with variable results (8–13). A study by Barnes et al (11) showed a significant effect of human gammaglobulin administered orally with each feed for the first week of life, compared with placebo in premature neonates (75 neonates, each 2–2.5 kg, in Australia). Gammaglobulin reduced the incidence of nosocomial RV diarrhoea in the first 2 weeks of life by 85% (95% confidence interval [CI] 9–97) compared with placebo (2.5% vs 16.7%, respectively) (11). There was no significant difference in the incidence of all-cause nosocomial diarrhoea during that period (38.5% vs 33.3%, respectively; relative risk [RR] 1.15, 95% CI 0.6–2.1) (11). Three studies evaluated orally administered hyperimmunised bovine colostrum containing antibodies against G1 to G4 types of RV in prevention of nosocomial RVGE (8–10). In 2 of these studies, one evaluating 130 children in Australia and the other 152 children in India and Hong Kong (8,9), colostrum compared with placebo provided complete protection (0% vs 14% and 0% vs 12%, respectively). In the third study (116 children in 3 hospitals in Poland) (10), the difference was not significant (5% vs 7%, RR 0.7, 95% CI 0.2–2.5). A small study involving 13 infants from an orphanage in Japan demonstrated that bovine colostrum against the G1 type of RV significantly reduced the risk for RVGE during an outbreak (17% vs 86%, RR 0.2, 95% CI 0.03–0.8) (12). A further study by Turner and Kelsey (13) evaluated the efficacy of infant formula supplemented with bovine antibody against G1, G3, and G4 types of RV in prevention of community-acquired RVGE in a field trial in 64 infants in South Carolina. The incidence of RV diarrhoea was reduced in the experimental group compared with placebo (3% vs 18%), but the difference was not statistically significant (RR 0.18, 95% CI 0.03–1.03). There was, however, a significant reduction in the number of days with RV-associated diarrhoea in the experimental group (1/1000 days vs 13/1000 days, respectively; P < 0.01) (13). In those trials demonstrating that passive immunisation with orally administered antibodies was effective, an effect was only visible during the treatment phase, with no evidence for any longer-term protection. Passive immunisation may be practical for limited target groups, such as prematurely born neonates in hospital. However, no commercial products containing anti-RV antibodies for oral administration are available. PROBIOTICS FOR THE PREVENTION OF RVGE Probiotics are live microorganisms that when administered in adequate amounts confer a beneficial effect on the health of the host (52). Several studies have shown that probiotic bacteria can shorten the clinical course of acute RVGE (14). The effects of probiotics also have been studied in RCTs for the prevention of nosocomial or community-acquired acute diarrhoea, including RVGE, in infants and young children (14–23). Four RCTs that evaluated the effect of probiotics for the prevention of nosocomial diarrhoea (mainly rotaviral) produced conflicting results (14–17). Two evaluated the use of Lactobacillus GG in young children hospitalised for a relatively short stay (14,15). The first double-blind RCT involved 81 children, ages 1 to 36 months and hospitalised in 2 paediatric hospitals in Poland (14). This showed that 6 × 109 colony-forming units (CFUs) of Lactobacillus GG given orally twice per day during a hospital stay reduced the risk for nosocomial diarrhoea by 80% (95% CI 40%–94%), and the risk for nosocomial RVGE by 87% (95% CI 20%–98%; 2.2% vs 16.7%, respectively), when compared with placebo (6.7% vs 33.3%, respectively). The second double-blind RCT involved 220 children, ages 1 to 18 months and hospitalised in 1 paediatric hospital in Italy, and evaluated Lactobacillus GG given orally once per day (1010 CFUs) (15). This study did not show a significant treatment effect on the incidence of nosocomial RVGE compared with placebo (13% vs 21%, respectively; RR 0.6; 95% CI 0.35–1.16). Two small double-blind RCTs addressed the efficacy of Bifidobacterium bifidum and Streptococcus thermophilus in the prevention of nosocomial diarrhoea (16,17). The first trial involved 55 infants ages 5 to 24 months admitted to a chronic care hospital in the United States for a relatively long stay (16). In this study, the feeding of a standard infant formula supplemented with B bifidum and S thermophilus reduced the incidence of nosocomial diarrhoea by 80% (95% CI 20%–94%) when compared with nonsupplemented formula (7% vs 31%, respectively), and reduced the incidence of RVGE by 70% (95% CI 20%–91%) (16). Furthermore, the rate of RV shedding was reduced in those infants receiving formula supplemented with B bifidum and S thermophilus. The second RCT was conducted in 90 healthy infants, younger than 8 months of age, living in residential nurseries or foster-care centres in France (17). In this study, milk formula supplemented with viable B lactis strain Bb-12 did not reduce the incidence of diarrhoea compared with administration of nonsupplemented formula (28.3% vs 38.7%, RR 0.7, 95% CI 0.4–1.3) (17). The efficacy of probiotics for the prevention of community-acquired diarrhoea in infants and young children has been evaluated in 6 RCTs with variable results (18–23). In a field trial involving 204 undernourished infants in a community with a high burden of diarrhoeal diseases (Peru), Oberhelman et al (18) showed significantly fewer episodes of diarrhoea in children who received Lactobacillus GG for 6 days per week for 15 months (3.7 × 1010 CFUs/day) compared with placebo (5.21 vs 6.02 episodes per child-year), particularly in non–breast-fed children ages 18 to 29 months (4.69 vs 5.86 episodes per child-year). There was, however, no significant difference in the incidence of RVGE (7.2% vs 5.1%, respectively). In another double-blind RCT performed in 571 children ages 1 to 6 years who attended day care centres in Finland, long-term consumption of probiotic milk containing Lactobacillus GG (2 × 108 CFUs/day) did not influence the incidence of diarrhoea or the number of days with diarrhoea, when compared with placebo during the 7-month study period (19). Thibault et al (20) evaluated the incidence of acute diarrhoea in a double-blind, placebo-controlled RCT in France. The trial involved 968 infants (4–6 months of age, "regularly" exposed to child care and/or living at home with at least 2 siblings) who were fed either a formula supplemented with B breve C50 and S thermophilus 065, or nonsupplemented formula (controls) for 5 months. Although the aetiology of diarrhoea was not evaluated, the study was conducted from October to February, which corresponded with the RV epidemic season and encompassed the winter peak in the incidence of gastroenteritis. The incidence of diarrhoea during the 5 months (56.7% vs 55.9%, respectively; RR 1.01, 95% CI 0.9–1.14) and risk for hospitalisation (0.9% vs 0.5%, respectively; RR 1.9, 95% CI 0.4–9.0) did not differ significantly between the 2 groups. However, there were fewer episodes of dehydration (2.5% vs 6.1%; P = 0.01), fewer medical consultations (46% vs 57%; P = 0.003), and fewer episodes leading to formula change (59% vs 75%; P = 0.0001) in infants fed probiotic-supplemented formula (20). Pedone et al (21) evaluated the efficacy of L casei strain DN-114 001 in a double-blind RCT involving 928 children (6–24 months old) attending day care centres in France. The study lasted 4 months during the RV epidemic season, though the aetiology of diarrhoea was not investigated. Daily consumption of milk fermented by yogurt cultures and L casei strain DN-114 001 reduced the risk for at least 1 episode of diarrhoea during the supplementation phase by 28% (95% CI 3%–46%) compared with children who were fed traditional yogurt (15.9% vs 22%, respectively). There was no significant difference in the duration of diarrhoeal episodes over the study period, or the risk for diarrhoea when infants were not fed the study products (21). Saavedra et al (22) evaluated the long-term consumption of formula supplemented with B lactis (strain Bb-12) and S thermophilus in children (3–24 months of age) attending day care centres in the United States in a double-blind RCT. In comparison with nonsupplemented formula (placebo), there was no significant difference in the incidence of diarrhoea over the almost 7-month period (22). In a recent double-blind RCT conducted in Israel, 210 healthy infants (4–10 months of age) attending child care centres were assigned randomly to receive formula supplemented with B lactis (Bb-12), L reuteri, or no probiotic for 3 months (23). In groups of infants fed formula supplemented with B lactis or L reuteri, the mean number of episodes of diarrhoea per child was lower than in infants fed nonsupplemented formula: 0.13 (95% CI 0.05–0.21) vs 0.02 (95% CI 0.01–0.05) vs 0.31 (95% CI 0.22–0.40) episodes, respectively. A similar difference in the mean number of days with diarrhoea per child was noted: 0.37 (95% CI 0.08–0.66) vs 0.15 (95% CI 0.12–0.18) vs 0.59 (95% CI 0.34–0.84) days, respectively. However, there was no significant difference among groups in the prevalence of RV-positive episodes (41%, 34%, and 38%, respectively) (23). There is not enough evidence to generally recommend continued use of probiotics for the prevention of acute diarrhoea or RVGE in infants and young children in day care centres or in the community. The evidence is stronger for short-term use in hospital, but the effect was noted only during the supplementation phase, with no evidence of any sustained protection thereafter. In the community, compliance may limit the potential effectiveness of probiotics, even though many parents in Europe give these products to their infants. BREAST-FEEDING FOR THE PREVENTION OF RVGE Multiple benefits of breast-feeding for a child's development have been clearly documented and recognised (24,25). Several respected organisations, including the World Health Organization and the American Academy of Pediatrics, recommend that infants be breast-fed exclusively for the first 5 to 6 months of life, and be breast-fed as a supplement to other foods for at least 12 months (53,54). Several observational studies have shown that exclusive breast-feeding protects against diarrhoeal diseases in infants in developed and developing countries (24–27), although specific protection against RVGE was variable (27–30,55,56). Observational studies examining the protective effect of breast-feeding against RVGE have shown variable results (27–30,55,56). Five studies were conducted in developing countries (28–30,55,56). Only 1 prospective cohort study carried out in the Caribbean islands (229 infants followed for 2 years) confirmed that breast-fed infants had a lower rate of RV infection than those who were bottle-fed to the exclusion of any breast-feeding (55). Another longitudinal cohort study conducted in Brazil (80 children followed up for their first 3 years) failed to show any protective effect of breast-feeding against RV infection and gastroenteritis (56). Glass et al (28) retrospectively compared breast-feeding rates according to age and enteric pathogens among 2276 children younger than 4 years of age with diarrhoea who attended a diarrhoea hospital in Bangladesh. Infants 0 to 5 months of age were less likely to have been breast-fed, compared with children ages 6 to 11 months of age. This finding seemed to suggest that early breast-feeding affords some protection against diarrhoea caused by all pathogenic agents. In every age group studied, breast-feeding was more common among children with RV than among children with non-RV diarrhoea. Moreover, despite the prolonged breast-feeding that was common in the Bangladesh study, the mean age of hospitalisation with RV was approximately the same as in countries where the duration of breast-feeding was short. The authors concluded that these 2 independent observations do not support a protective role for breast-feeding against severe RV diarrhoea after the first few months of life (28). A case-control study conducted in Brazil (500 infants ages ≤12 months with acute diarrhoea and 500 age-matched controls without diarrhoea) found that the rate of breast-feeding in infants younger than 6 months was associated with protection against acute diarrhoea (specifically enteropathogenic Escherichia coli infection), but not RV (29). Another case-control study conducted in Bangladesh (132 children <24 months of age with severe RVGE and 2587 controls) concluded that exclusive breast-feeding of infants was associated with protection against severe RV diarrhoea (RR 0.10, 95% CI 0.03–0.34) (30). However, during the second year of life, the risk for this outcome was higher in previously breast-fed than in non–breast-fed children (RR 2.85, 95% CI 0.37–21.7), and breast-feeding during the first 2 years of life was not associated with protection against diarrhoea or severe diarrhoea (RR 2.61, 95% CI 0.62–11) (30). Although exclusive breast-feeding appeared to protect infants against severe RV diarrhoea, breast-feeding per se conferred no overall protection during the first 2 years of life, suggesting that breast-feeding postponed, rather than prevented, this outcome. Five studies were performed in developed countries to examine the benefits of breast-feeding against the development of RVGE (15,31–34). Three longitudinal cohort studies prospectively followed up 405 neonates and infants for several months (from 1 RV epidemic season (31) until 16 months of age (33)) and did not confirm a protective role of breast-feeding against RVGE in the community (31–33), although in 1 study the severity of RVGE was milder in breast-fed infants (31). The most recent prospective cohort study in 220 hospitalised infants, ages 1 to 18 months, showed that 47 breast-fed children (at least 2 breast-feedings per day) were protected against nosocomial RVGE during a relatively short stay in hospital compared with 173 non–breast-fed children (0% vs 21.4%, respectively) (15). Another small case-control study involving 50 US infants younger than 1 year with RVGE and controls showed that there was no significant difference in the prevalence of infants that had been breast-fed in both groups (64% vs 70%), but in 50% of cases breast-feeding was stopped before RVGE (34). Although the results of observational studies are variable and many of them are subject to bias, it seems that exclusive breast-feeding may offer some degree of protection against RVGE during the first few months of life. However, there is no evidence of any longer-term protection once exclusive breast-feeding is stopped. CONCLUSIONS Despite efforts to prevent RV disease in infants and young children, this disease remains the leading cause of morbidity in children younger than 5 years of age. Improvements in sanitation and hand-washing were shown to have limited benefit for the prevention of RV diarrhoeal episodes. This is underlined by approximately equal incidence of RV disease in children living in industrialised and developing countries, despite better sanitation in developed countries (38,57–61). Passive immunisation has been used successfully for RVGE prevention for children spending short periods in hospital, but commercial preparations are not available. Probiotics have shown promise in prevention, but require further studies and a cost-effective analysis before any recommendation for community use can be made. Breast-feeding is recommended, and should be promoted, but its effects are limited to postponing rather than preventing RVGE. Conflicts of Interests of the Authors J.M. has received honoraria for consultant services and lectures from GlaxoSmithKline (GSK), MSD, Wyeth, Nutricia Poland, Nestlé Poland, Sanofi Pasteur Poland, and Pfizer, research grants from Nutricia and Wyeth, and financial support for scientific congresses from Nestlé Poland and GSK. T.V. has received honoraria for consultant services and lectures from Chiron, Merck, GSK, MedImmune, and Wyeth. He has been the principal investigator of clinical trials for RotaShield (Wyeth-Lederle Vaccines), RotaTeq (Merck), and Rotarix (GSK Biologicals). H.S. has received lecture fees and/or honoraria for consultant services from Nestlé, Nutricia Poland, Numico, Mead Johnson Nutritionals Poland, Mead Johnson International, Biocodex France, Danone, Crotex, Merck, Biomed Lublin, Biomed Kraków, and GSK. She has received research grants or donations from Dicofarm Italy, Nutricia Research Foundation, and Biomed Lublin, and sponsorship to attend meetings from Nestlé Poland, Danone, and GSK.