An update on anti-Helicobacter pylori treatment in children

1Community Oriented Primary Care, Parkland Health and Hospital Systems, Dallas; 2University of North Texas, School of Public Health, Fort Worth, Texas, USA; 3Surrey GI Clinic/Research, Guelph; and Division of Gastroenterology, McMaster University, Hamilton, Ontario; 4Division of Gastroenterology, Dalhousie University, Halifax, Nova Scotia Correspondence and reprints: Dr R Khurana, Community Oriented Primary Care, Parkland Health and Hospital Systems, 4444 Hampton Road, Dallas, Texas 75232, USA. Telephone 214-266-1467, fax 214-266-1455, e-mail rkhura@parknet.pmh.org R Khurana, L Fischbach, N Chiba, S Veldhuyzen van Zanten. An update on anti-Helicobacter pylori treatment in children. Can J Gastroenterol 2005;19(7):441-445.

H elicobacter pylori infection acquired during childhood causes clinical disease in only a small proportion of infected individuals (1).Nevertheless, H pylori infection is a leading cause of peptic ulcer disease, and its eradication results in a long-term cure for patients (1)(2)(3)(4).In 1999 and 2000, the Canadian Helicobacter Study Group (5) and the North American Society for Pediatric Gastroenterology and Nutrition (6) recommended proton pump inhibitor (PPI)based triple therapies with clarithromycin (C) and either amoxicillin (PPI-CA) or metronidazole (PPI-CM) for one to two weeks as first-line treatment of H pylori infection in children.The North American Society for Pediatric Gastroenterology and Nutrition also recommended PPIbased triple therapy with amoxicillin and metronidazole (6).These previous guidelines were based on level I evidence from studies in adult populations and adult recommendations because there were insufficient pediatric clinical trials (7,8).
The following review provides an updated summary of evidence from pediatric and adult studies regarding the efficacy, safety, drug resistance and reinfection rates of anti-H pylori therapies.

METHODS
To summarize the efficacy of anti-H pylori treatment and adverse events in children, evidence was used from the authors' recent meta-analysis of pediatric studies (unpublished data).Additionally, data were used from the authors' existing databases and Medline searches using the key words 'Helicobacter pylori' or 'Campylobacter pylori', and 'therapy' or 'treatment', 'reinfection', 'drug resistance', and 'rescue' or 'second-line therapy'.

Efficacy
The meta-analysis included 99 treatment arms and 67 studies from 25 countries across four continents (unpublished data).It provided data up to May 2004 on 23 treatment regimens used to eradicate H pylori in 3175 children.The vast majority of studies were nonrandomized with fewer than 50 subjects.
Treatment efficacy varied according to the number and type of drugs used in the regimen, treatment duration, geographical location and the methods used to assess eradication after treatment.Only 5% or less of the infections were eradicated using a placebo, H 2 -receptor antagonist, sucralfate or omeprazole alone in five treatment arms (9)(10)(11)(12)(13).Dual therapy with amoxicillin and a nitroimidazole (tinidazole or metronidazole) (AN) was more efficacious in children than in adults (14).Overall, the mean eradication rate for this dual therapy was 83.9% (365 eradicated of 435 treated) in nine treatment arms when given from 10 days to six weeks (95% CI 80.2% to 87.2%); however, this dual therapy's efficacy was greatly reduced in Kuwait (15) and when given for only one week (10).Other dual therapies eradicated less than 80% of infections in most studies (unpublished data).
Triple therapies were more efficacious compared with dual therapies given for the same duration (Figure 1) (10,(16)(17)(18)(19). Tinidazole and amoxicillin dual therapy (TinA), which performed well when given for two weeks, tended to perform better when a PPI was added (PPI-TinA) (eradication 85% versus 100%; P=0.17) (10).However, eradication of H pylori with PPI-TinA was significantly better than with TinA when the treatment duration was one week (P=0.004)(10).PPI-CA, the most commonly tested regimen, cured 58% to 100% of H pylori infections in 29 treatment arms with 1014 children.More than 80% of H pylori infections in trials from Asia, the Middle East and European countries outside of Italy and Spain were eradicated using PPI-CA, irrespective of treatment duration.However, in the United States, only 68% of infections were eradicated using two-week PPI-CA treatments (20).Two-week therapy with a bismuth compound, amoxicillin and metronidazole (BAM), given in developed countries outside of Spain, eradicated H pylori in more than 80% of children.A PPI, macrolide and a nitroimidazole regimen (PPI-MacN) typically eradicated less than 80% of H pylori infections, but in one small nonrandomized study in Canada, 93.3% erradication was achieved with two-week therapy (21).Only three quadruple-treatment regimens were tested, each for one week in a small number of children.Treatment with a nitroimidazole, amoxicillin and furazolidone, and either ranitidine or omeprazole resulted in eradication of 70% (seven of 10) or 80% (four of five) of infections in Brazilian children, respectively (22).Furazolidone is an inexpensive antibiotic, commonly used in South America, but not available in North America.One-week treatment with metronidazole, clarithromycin, amoxicillin and omeprazole eradicated 94% (31 of 33) of H pylori infections in children in Hong Kong (23).

Adverse events
Only mild adverse events have been observed in H pylori treatment trials in children, but the manner in which adverse events were recorded and the occurrence of adverse events varied greatly between studies (unpublished data).Adverse events were observed in 0% to 34% of subjects given a dual therapy over one to six weeks of treatment, and in 0% to 80% of subjects given a triple therapy for one to three weeks (unpublished data).In head-to-head studies, triple therapies did not result in more adverse events compared with dual therapies.In one study (16), there were more adverse events with dual therapy using clarithromycin and amoxicillin compared with PPI-CA (34% versus 14%; P<0.0001).Although there has been some concern over the use of bismuth salts in children and its association with encephalopathy (3), bismuth levels were found to be well below toxic levels in children (24,25).
Resistance of H pylori to a nitroimidazole, clarithromycin and other drugs plays a major role in treatment failure in adults (14).H pylori eradication in adults decreased by 0.5% for every 1% increase in the prevalence of metronidazole-resistant strains (14).While data were sparse for pediatric studies, treatment regimens in head-to-head studies were consistently more efficacious when the strain of H pylori was sensitive to the drugs in the regimen (Figure 4).In five studies (29,30,46,51,52) that reported treatment efficacy by clarithromycin resistance, 83% to 100% of clarithromycin-sensitive H pylori infections, versus 0% to 56% resistant infections, were cured after treatment with PPI-CA.In Spain, 88% of the metronidazole-sensitive strains were eradicated with BAM, compared with only 50% of the metronidazole-resistant strains (31).

Reinfection
When reported, some reinfection was observed in children, especially in the first six to 12 months after treatment.Reinfection occurred in 0% to 7% of children per month after eradication (4,9,12,(56)(57)(58)(59).Lower reinfection proportions were reported when the follow-up times were longer; for example, Oderda et al (60), Huang et al (12) and Kato et al (61) reported reinfection in 0.13% to 1% per month after 18 to 28 months of follow-up.Other studies (62) have observed reinfection to be associated with treatment efficacy, family size, age, socioeconomic status, exposure to subsequent endoscopies and geographical location.Because treatment efficacy is a predictor of reinfection, it seems likely that a proportion of cases are due to recrudescence rather than true reinfection.

Rescue therapies
Only two studies were identified that reported eradication proportions with second-line treatment.After the initial treatment failure using amoxicillin and cimetidine or cimetidine alone, or bismuth and metronidazole in combination, only 12.5% and 41.2% of H pylori infections were eradicated with rescue therapies containing a PPI with clarithromycin and a nitroimidazole for one week (63), or amoxicillin alone for two weeks (9), respectively.

DISCUSSION
The ideal therapy to eradicate H pylori should be safe and effective in both the short and long term for populations requiring treatment.Studies of H pylori eradication in children (6) were smaller and generally methodologically weak compared with the better designed randomized controlled trials in adults.The treatment regimens that were more than 80% efficacious in children included two-week dual therapy with AN in Europe; two-week BAM treatment in developed countries; one-to two-week PPI-CA treatment in Asia, Northern Europe and the Middle East; and two-week PPI-MacN treatment in Canada.PPI-AN, although used in only five treatment arms in France and Italy, eradicated 74% to 100% of infections; further testing may show this therapy to be effective in some populations of children.
Previous consensus statements for North American children have recommended one to two weeks of treatment with PPI-CA or PPI-CM (5,6), which reflect recommendations for adults (7,8).Although one-to two-week treatment with PPI-CA eradicated 84% of H pylori infections in trials in Northern Europe, Asia and the Middle East, and appears to be safe and readily available in most populations of children, it eradicated less than 75% of infections in children resistant to clarithromycin, and in populations in Latin America and the United States (20,29,30,46,64,65), where both metronidazole and clarithromycin resistance rates in children are high (48,49).Trials using PPI-CA in Canadian children have not been reported, but in one nonrandomized trial, the efficacy of PPI-MacN was 93% in Canada (21).In many populations outside of Canada, however, the efficacy of PPI-MacN was less than 80%.Furthermore, regimens containing metronidazole

US United States
Figure 4) Proportion of Helicobacter pylori eradication in clarithromycin (CLR)-, metronidazole (MET)-and amoxicllin (AMX)sensitive and resistant strains in head-to-head studies may be less desirable for children because it is not available in liquid form and has an unpleasant taste (5).Many regimens have yet to be tested in developing countries and neither AN nor PPI-AN has been evaluated in North American children.However, the Canadian Helicobacter Study Group has endorsed this PPI-based triple therapy as an alternative therapy in adults (8).
There are some newer regimens that appear very promising in the treatment of H pylori in adults, which may also be effective in children.Quinolone triple therapies, when used in combination with a PPI and either amoxicillin, tinidazole or clarithromycin, have intention-to-treat eradication rates of approximately 90% in adults (66)(67)(68); however, the safety of quinolones in pediatrics has not been established.Another regimen, a one-week quadruple regimen with a PPI, amoxicillin, clarithromycin and metronidazole, eradicated 94% of H pylori infections in children (23).In adults, this regimen had an efficacy of approximately 90% when given for only five days (69)(70)(71)(72).This would be a suitable regimen for children because shorter duration may help to improve compliance.
Although only minor adverse events have been reported with anti-H pylori therapies in children, tetracycline may damage the dental enamel, and this drug should not be used in children younger than 12 years of age (73,74).Thus, although the quadruple regimen of PPI, bismuth, metronidazole and tetracycline (PPI-BMT) is highly effective in adults, and has been recommended for both first-line and salvage therapy (8), it is recommended only for children older than 12 years of age.Acetylsalicylic acid and nonacetylsalicylic acid salicylates, such as bismuth subsalicylate, are associated with Reye's syndrome in children with febrile illness and should be used with caution (only if the child does not have a febrile illness [75]).The two studies that measured bismuth levels with H pylori therapy in children showed that levels were well below the neurotoxic range (24,25).
Because there were almost no data for treatment of H pylori eradication failures in pediatrics, recommendations stem largely from adult data.The present recommendation for rescue therapy is to administer a second course of alternate PPI triple therapy (PPI-CM if PPI-CA was used first or vice versa) or a quadruple therapy of PPI-BMT (8).While the PPI-BMT therapy is the most robust in adults, the use of tetracycline can be problematic as already discussed.Rifabutin-and levofloxacin-based therapies appear effective in adults, but rifabutin is expensive and there are limited safety data for either drug in children (76)(77)(78).Thus, at this time, these regimens would need further study in children before recommendations for their general use can be made.
Metronidazole and clarithromycin resistance are common in developing countries, Canada and the United States, and no therapy has yet been identified as safe and consistently effective to eradicate H pylori infection in children in these populations.Therefore, in addition to the need to identify safe and effective anti-H pylori rescue therapies in children throughout the world, there is also a need to identify safe and effective primary eradication therapies for H pylori-infected children in North America and developing countries, especially in populations where the prevalence of metronidazole and/or clarithromycin resistance is high.

Figure 1 )Figure 2 )
Figure 1) Efficacy of head-to-head dual versus triple therapies given for the same duration.CA Clarithromycin and amoxicillin; H 2 -TinA H 2 -receptor antagonist, tinidazole and amoxicillin; PPI-A Proton pump inhibitor and amoxicillin; PPI-CA PPI, clarithromycin and amoxicillin; PPI-TinA PPI, tinidazole and amoxicillin; PPI-TinC PPI, tinidazole and clarithromycin; RbcA Ranitidine, bismuth citrate and amoxicillin; TinA Tinidazole and amoxicillin; wk Week

Figure 3 )
Figure 3) Prevalence of primary clarithromycin resistance in children.US United States