To review the available clinical data on the recurrence of gastroduodenal Helicobacter pylori (H. pylori), specifically reporting the incidence of recurrence, the nature of recurrence (recrudescence or reinfection), and the factors associated with each of these including the efficacy of initial therapy.

MEDLINE was searched from 1986 to 1995. It was not stated whether any language restrictions were applied. Abstracts from 'major meetings' were also searched.

Study designs of evaluations included in the review

The study designs were not stated. The results used were from observational or cohort studies.

Specific interventions included in the review

Eradication. The specific drugs used for eradication were: colloidal bismuth subcitrate, other bismuth salts, omeprazole, cimetidine, ranitidine, plaunotol, furazolidone, metronidazole, tinidazole, amoxycillin, and tetracycline. Those studies that reported recurrence of H. pylori after successful eradication were eligible for inclusion.

Participants included in the review

H. pylori infection. There were at least 1,868 patients included in the review. The characteristics of these patients was not reported, except that they had been treated for H. pylori infection.

Outcomes assessed in the review

The outcomes assessed were:

the recurrence rates;

the nature of recurrence (same strain or new strain of organism);

factors associated with recrudescence (same strain), such as techniques for monitoring, sampling methods, and the efficacy (eradication rates) of the original treatment; and

reinfection (new strain), i.e. susceptibility of the host and re-exposure to H. pylori.

How were decisions on the relevance of primary studies made?

The authors do not state how the papers were selected for the review, or how many of the authors performed the selection.

The authors do not report the method used to assess validity, or how the quality assessment was performed.

The authors do not state how the data were extracted for the review, or how many of the authors performed the data extraction.

How were the studies combined?

The study outcomes were combined using a narrative method. Details on the weighting or selection of the studies were not given.

How were differences between studies investigated?

The authors do not state how differences between the studies were investigated.

A total of 32 studies (1,868 patients) assessing recurrence rates were reviewed. Four of these (17 patients) assessed genetic mapping of the organism after recurrence. Twenty-one studies (at least 2,543 patients) reported the success rate of the original treatment and recurrence rates. The numbers of studies and participants for the other specific objectives were unclear.

Recurrence rates.

Overall, the recurrence rates ranged from 0 to 40% at 6 months' post-treatment (14 studies), from 0 to 41.5% at 12 months' post-treatment (24 studies), from 4.3 to 21.4% at 18 months' post-treatment (4 studies), from 0 to 10.9% at 24 months' post-treatment (7 studies), and from 2.2 to 9.3% at 48 months (2 studies); the rates in single studies were 9.3% at 60 months and 8.6% at 84 months. These studies used a variety of methods of identifying recurrence. The recurrence rates appear to have been the highest in the first 12 months after treatment completion.

Nature of recurrence (same strain or new strain of organism).

Very few studies conducted genetic mapping before and after eradication. Four studies found that the majority of recurrences were with the identical organism. However, there were only 17 patients in these studies, and the one that studied the spouse or partner of the patient found that the couples had the same strain. This made it impossible to distinguish between recrudescence and reinfection in such cases.

Factors associated with recrudescence (same strain).

Techniques for monitoring: the urea breath testing methods may not be as useful for monitoring recurrence as they are for identifying untreated infection; biopsy-based testing in combination with urea breath testing may have more consistent results; PCR (polymerase chain reaction) testing is still controversial and has mixed results.

Sampling methods: sampling only in the gastric antrum may lead to inaccurate classification of eradication during or after treatment, whilst sampling in the corpus and the fundus is recommended. Increasing the number of samples taken to at least 4 did not decrease the recurrence rates significantly. Sampling from at least the antrum and fundus reduced the rates of recurrence at 12 months to 3.5% or below.

Efficacy (eradication rates) of original treatment.

Triple-drug therapy had the best eradication rates, ranging from 55 to 93% (p<0.001, compared with single and double therapy). Regression analysis showed that the recurrence rate was approximately 5% when the eradication rate was 90%. The eradication rate was highly related to the recurrence rate. Regimens with eradication rates of at least 90% had recurrence rates of 0 to 13%.

Factors associated with reinfection (new strain).

Some individuals have been found to be more susceptible to infection with H. pylori. Reinfection was uncommon in adults. Reinfection may occur more often in developing countries, among children living together, in families, and through contaminated endoscopy equipment.

Recurrence of H. pylori infection occurs after apparently successful eradication. The 12-month recurrence rate varied in different studies, from 0 to 41.5%. Limited data, obtained using molecular fingerprinting techniques, have shown that in most cases recurrence is due to recrudescence of the original strain; a few cases appear to be due to reinfection with a new strain or a genetically identical strain from the same original source. Recrudescence is most likely during the first 12 months after apparent eradication, while reinfection may account for recurrence after this period.

Despite the high sensitivity and specificity of the available individual tests for detecting H. pylori infection in untreated patients, these techniques are not sensitive enough alone to monitor eradication. Sensitivity and specificity may be increased when biopsies for detection of H. pylori are taken from both gastric antrum and corpus (and fundus if possible), and when more than one type of test is applied. The best treatments have the lowest recurrence rates, and recurrence is rare when the eradication rate is over 90%. Individual susceptibility and the re-exposure to H. pylori were thought to be the two major causes of reinfection.

The research questions were clearly identified, and the search strategy attempted to identify unpublished research. The search strategy, however, could have been more specific with respect to language restrictions, the total number of studies identified by the search, and the method used to search for unpublished reports.

The inclusion criteria and the methods for assessing the validity of the included studies were not discussed in the paper. The primary data were tabulated, with the exception of data relating to the study design used. The synthesis did not take the indicators of study quality into account, and without these it is difficult to assess the quality of the studies in order to weight the outcomes.

The authors' conclusions are relevant to this area of research and clinical practice, in so far as they are valid. However, the conclusions should be taken as uncertain given the caveats highlighted.

Systematic reviews of this topic, which adequately assess and weight the quality of the included studies, should be conducted to confirm or clarify these findings. Also, further assessment of the individual regimens is necessary if this research is to be useful for clinicians.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.