Helicobacter pylori infection in India from a western perspective

Helicobacter pylori is a common and important transmissible bacterial human pathogen. The prevalence of this infection varies world wide being as low as 10 per cent in developed western nations to higher than 80 per cent among the indigent populations of many developing countries. The infection primarily involves the upper gastrointestinal tract causing progressive acute and chronic gastro-duodenal inflammation. Typically these inflammatory changes are silent but clinical disease manifestations occur in approximately 20 per cent, generally after a long latent period1. The manifestations of H. pylori infection include gastritis, gastric atrophy, duodenal ulcer disease, gastric ulcer disease, primary gastric B-cell lymphoma, gastric adenocarcinoma, iron deficiency anaemia, and vitamin B12 deficiency2–5. There are often regional differences with regard to which clinical manifestation is predominant ranging from iron deficiency anaemia in childhood to gastric cancer in the elderly. The predominant manifestation can also evolve over time. For example, in the first half of the 20^th century there was a rapid and progressive decline in the incidence of gastric cancer in the west which coincided with a sharp rise in the incidence of duodenal ulcer.

Gastric cancer is one of the most significant out-comes of H. pylori infection and understandably attracts the most attention from the research community. However, in many areas, especially in tropical and semitropical countries (e.g., Africa, southern India), the infection is common but gastric cancer is rare with duodenal ulcer being the dominant clinical manifestation of the disease5–7. Less dramatic, but no less important, is an increased susceptibility to enteric infections due to H. pylori gastritis-related hypochlorhydria and iron deficiency anaemia both of which can have major deleterious effects on physical and intellectual growth of children especially in developing countries8.

H. pylori infection is typically acquired in childhood. The risk of infection is inversely related to the overall sanitary conditions and requires exposure to other infected humans. Contaminated water is often the primary mode of transmission in rural areas without reliable supplies of potable water910. However, in regions of higher socio-economic status the risk of infection best correlates with the level of household hygiene.

Outcome of H. pylori infections

The outcome of an H. pylori infection reflects a complex interplay of environmental, host and bacterial factors including the virulence of the infecting bacterial strain. There are no nonpathogenic strains of H. pylori as even the least virulent strains cause gastric inflammation and have been associated with peptic ulcer disease and gastric cancer. The virulence of H. pylori strains correlates with the intensity of the inflammatory response to the infection. Established H. pylori virulence factors include the cag pathogenicity island (cag PAI), the vacuolating cytotoxin (VacA) and the outer inflammatory protein, OipA. Host factors involved in disease pathogenesis include polymorphisms of genes that govern the host's inflammatory response (i.e., a genetic predisposition to an increased response is associated with a greater risk of a clinical outcome). A genetically determined increased parietal cell mass is thought to predispose to H. pylori-related duodenal ulcer disease. Environmental factors also contribute to the outcome of H. pylori infection. This is best seen in relation to the association of H. pylori-related gastric cancer with diets low in fresh fruits and vegetables but high in salt and the rapid decline in the incidence of gastric cancer that paralleled dietary changes in those populations. Replacing salt/smoke with refrigeration as the primary means of food preservation and an increased intake of fresh fruits and vegetables has universally been associated with a fall in the incidence of gastric cancer among high risk populations without a change in the prevalence or virulence of H. pylori infections11–14.

Should H. pylori be eradicated?

H. pylori is a significant human pathogen responsible for considerable morbidity and mortality and is the major cause of gastric cancer. The majority of investigators in the field believe that whenever the infection is detected, it should be eradicated15. However, that goal may be difficult to attain in some non-western populations. Some have hypothesized that H. pylori infection may even be beneficial and that eradication is not always the best option16. H. pylori is a human pathogen. Although H. pylori is not present in wild monkeys, its association with mankind can be traced back to the time when humans migrated out of Africa17. Such a long association is neither unique nor evidence of a positive benefit, at least to man. Over the centuries, man has been host to many bacterial, viral, protozoal, and helminthic pathogens. Gradually, as populations dispersed and sanitation improved, most pathogens have been lost. H. pylori is one of the last of those pathogens to undergo clearance associated with improvements in the standards of living. Despite its long-standing presence, H. pylori is not a commensal; it always causes disease (e.g., progressive gastric inflammation). Commensals are very difficult to eliminate, however, with H. pylori, the chain of transmission is easily broken in association with improvements in hygiene and sanitation practices.

Are there adverse consequences to H. pylori eradication?

In epidemiology, when one variable is associated with a reduction of another (e.g., a disease) it is said to be “protective”. For example, populations with limited sanitation, high rates of malnutrition and early mortality have a low incidence of age-related diseases such as diabetes, stroke, and cancer. One could conclude that from an epidemiologic standpoint poor sanitation “protects” against those diseases. This would clearly be a misuse of the medical concept of “protection”. However, the same approach has been used to show that H. pylori infection possibly “protects” against other diseases18. For example, atrophic gastritis, the precursor lesion for gastric cancer, leads to low levels of gastric acid secretion (i.e., it can be considered to be a biologic anti-secretory agent) and thus protects against gastroesophageal reflux disease and its dreaded complication, oesophageal adenocarcinoma19. It has been suggested that possibly one should not eradicate H. pylori because H. pylori infection may protect against development of oesophageal adenocarcinoma, a clear misunderstanding of the concept of protection19–21. Similarly, H. pylori infection has been suggested to protect against obesity and childhood asthma2223. Childhood asthma is often thought to be related to a failure of exposure of the immune system to important environmental antigens (the hygiene hypothesis). Does H. pylori infection contain some of those critical antigens? The available data suggest that the presence of H. pylori infection is not responsible but is instead a surrogate or marker for inadequate household hygiene. Experience in rural Malaysia allowed the H. pylori protective hypothesis to be tested. In rural Malaysia hygiene is often poor, and enteric diseases are common, but H. pylori is essentially absent24. One would therefore expect childhood asthma to be common, gastroesophageal reflux, oesophagitis, adenocarcinoma of the oesophagus, and obesity to be significant problems. However, childhood asthma, obesity, gastroesophageal reflux, and oesophageal adenocarcinoma are all rare19. These data are directly in contrast to what was predicted by those who espouse dire consequences associated with H. pylori eradication. In terms of childhood asthma, the experience in Malaysia is consistent with the hygiene hypothesis (i.e., the likelihood of environmental antigens in priming the immune system) but is direct evidence that the critical antigens are unlikely to be related to H. pylori infection and support the premise that the presence of H. pylori infections is a marker for poor household hygiene24.

In summary, H. pylori is an important human pathogen and should be eradicated whenever possible. However, eradication is more complex than the simple matter of writing a prescription for antibiotics. This paper will review H. pylori treatment strategies and the approach to identifying the best regimens for a particular population with a special emphasis on H. pylori eradication in India.

H. pylori as an Indian problem

India is a vast country known for its rich history, culture and food. It is also the prototypical developing country with a vast rural population living in poverty. The prevalence of H. pylori in the Indian subcontinent can be as high as 80 per cent or more in rural areas. The most commonly recognized manifestation of H. pylori infection in India is peptic ulcer disease, particularly duodenal ulcer disease, which outnumbers gastric ulcers between 8:1 and 30:125. Singh et al26 calculated the point prevalence of active peptic ulcer disease at 3% with a lifetime prevalence of 9 per cent. As in other regions, the actual risk of a particular outcome from an H. pylori infection is predicated on the pattern of gastritis7. Antral predominant gastritis leaves an intact gastric corpus, poorly controlled acid secretion and promotes duodenal ulcer formation. In contrast, with pangastritis acid secretion often falls below the level needed to produce and sustain duodenal ulcer disease (e.g., approximately 12 mmol/h), gastric ulcer becomes more common than duodenal ulcer and the incidence of gastric cancer rises. Finally, atrophic pangastritis is the main precursor lesion associated with gastric cancer27.

Environmental factors, especially diet, play a role in the pattern of gastritis and this likely underlies the differences in the prevalence of H. pylori diseases in India. Tropical diets with plentiful fruits and vegetables year-round promote antral predominant non-atrophic gastritis and duodenal ulcer disease57. Seasonal diets where fresh fruits and vegetables are not available during the winter or dry season and food is commonly preserved with salt promote pangastritis, a lower incidence of duodenal ulcer disease and more gastric ulcers. These environmental factors are likely responsible for the higher prevalence of duodenal ulcer in the south and for the wide range in the age-adjusted incidence rate of gastric cancer (range from 2 to 57 per 100,000)71128.

The population of India is approximately 1.2 billion people. If the H. pylori prevalence was 60 per cent, more than 726 million individual would be infected with H. pylori29. The estimated prevalence of duodenal ulcers in India is 3 per cent and means that at least 18 million people could need anti-H. pylori therapy (approximately 50,000 per day if treated over one year). The enormity of the task of treating H. pylori infection is daunting and might dissuade some physicians and the government from aggressively managing the infection. However, when one considers the problem one patient at a time, the issue becomes much simpler.

H. pylori diagnosis

The first step in managing H. pylori infections is to establish the diagnosis. Physicians in India, as in many developing countries, face issues of availability and cost to establish the diagnosis of H. pylori infection. Worldwide, non invasive tests for active infection are preferred (e.g., urea breath test or stool antigen test) over tests that require endoscopy. Western versions of these tests are relatively expensive but all have been used at one time or another in India. The¹⁴C-urea breath test (UBT) is probably the least expensive. In China, it is available clinically for approximately 26 per test (DY Graham, personal communication). Endoscopy is available at tertiary care centers but can cost up to 150 at a public hospital and between 450 and 3000 at private clinics30. The advantage of endoscopy is that symptomatic patients can be evaluated for mucosal disease. Endoscopy also allows one to take biopsy specimens that can be examined by histology, rapid urease testing, brush cytology, or even culture. Rapid urease tests can be made very cheaply by any hospital laboratory and are reasonably accurate3031. Saksena et al32 showed that the rapid urease test (RUT) and brush cytology had the highest degree of agreement whereas histology was the most specific diagnostic test. The current gold standard to diagnose H. pylori infection invasively is two positive tests, which in India could be the combination of the rapid urease test and brush cytology3132. Importantly, the endoscope, the biopsy forceps, and other parts of the system (e.g., water bottles, tubes, rubber ports, etc.) can easily become contaminated with H. pylori, therefore, it is critical that high-level disinfection be done of all endoscopic equipment and any reusable devices to prevent the iatrogenic spread of the infection or for iatrogenic reinfection post H. pylori eradication. The minimally invasive oral brush culture method would seem to be ideal for obtaining culture specimens for susceptibility testing in India33.

If endoscopy is not available, for symptomatic patients, barium contrast imaging can be used to detect peptic ulcer disease followed by a non-invasive test to confirm H. pylori infection. Serologic testing for H. pylori antibodies has greater than 80 per cent sensitivity and greater than 90 per cent specificity and is widely available but can cost between 200 to 30030. Serologic testing is not recommended as the single diagnostic test in asymptomatic patients as it can remain positive for decades after successful H. pylori therapy and gives a false positive test result. However, it can be one of the two tests to confirm H. pylori infection or combined with a high pretest probability condition such as a duodenal ulcer to confirm the diagnosis.

Choosing the treatment regimen – the effect of antibiotic resistance

Treatment logically follows diagnosis. The goal is to choose a regimen that will reliably produce a 90 per cent or preferably >95 per cent treatment success23435. Recommendations from consensus statements should be ignored unless the antibiotic combinations have been proven to be effective locally. Bacterial resistance is the most common cause of treatment failure with what would otherwise be an effective regimen. Resistance is an evolving process such that antibiotic susceptibility and local treatment success rates must be monitored to update treatment selections. Because of the rapid development of resistance, it is recommended that the outcome of every treatment be confirmed with a non invasive test to confirm cure and provide an early warning of changes in local patterns of resistance.

Different antimicrobial agents have been used to treat H. pylori infection. Most successful regimens consist of an anti-secretory agent, most often a proton pump inhibitor (PPI) and two or three antimicrobial agents235. Successful treatment requires consideration of a number of factors (Box 1) including drug, dose, duration, and frequency of drug administration, formulation, and whether the drugs are taken fasting or with meals37. These factors are all largely under control of the prescribing physician. If one knows the local resistance rates to the different antibiotics one can generally choose a regimen that will reliably provide at least 90 per cent treatment success. As far as treatment is concerned, it is important to stop considering H. pylori as a “gastroenterology problem” (i.e., in the same light as we might consider gastroesophageal reflux, constipation, or functional bowel disease) and instead consider it as a common bacterial infectious disease more akin to urinary tract infection or pneumonia38. Most gastrointestinal diseases are not curable and have a considerable placebo response. In contrast, most common bacterial infectious diseases can be cured and with H. pylori there is no placebo response. The lessons learned regarding evaluation of common infections are also applicable to H. pylori (i.e., one should expect to cure them all). In addition, the reason for treatment failure should never be a mystery but rather is typically related to resistance or a poor choice of drugs, durations, etc.

Close

Box 1

Some factors known to influence outcome in H. pylori treatment studies

Export to PPT

Easy access to antibiotics contributes to overuse. In India, antibiotics can be obtained without a prescription which leads to overuse or misuse. Worldwide, antibiotics are frequently misprescribed or dispensed incorrectly due to lack of appropriate knowledge, desire to meet patient demands, mistrust in or delayed laboratory results and for economic incentives39. Metronidazole is inexpensive, available over the counter and widely used as an anti-diarrhoeal agent and for functional bowel disease. As such, there is generally a high prevalence of metronidazole resistance in India and in other tropical and sub-tropical developing countries40. If one were to start with a blank slate, one would probably start by testing locally obtained strains for antimicrobial resistance and use these data to select initial regimens. There are however, data available regarding antimicrobial resistance in India (Tables I & II), and these data suggest that resistance is both common and widespread but with considerable regional variation. Are these data accurate and valuable? To be useful, one would like to know whether the patients tested represented treatment naive patients or H. pylori treatment failures, whether they had received the antibiotic in question in the past, their age and socio-economic status and presentation. Accuracy can be assessed in terms of the methods used, whether known susceptible and resistant controls were included and how the results correlated with the results of treatment. In general, the E-test is more difficult to master than agar dilution. The laboratories should present data regarding their methods especially when the results are unexpected (e.g., high levels of amoxicillin resistance) and provide confirmatory testing. The E-test tends to overestimate the prevalence of metronidazole resistance but in experienced laboratories provides accurate results for clarithromycin and amoxicillin resistance50–52.

Table I Regional H. pylori in vitro antibiotic resistance in India

Table II Regional H. pylori in vitro metronidazole resistance in India

Based on the available data regarding antimicrobial resistance, one would expect regimens using clarithromycin or metronidazole to provide unacceptably low eradication rates (Table III). As noted earlier, when examining the results of a clinical trial it is important to take note of the drugs, doses and durations. The report by Bhatia et al47 tested appropriate doses of proton pump inhibitor (PPI), amoxicillin, tinidazole or clarithromycin for 14 days in a sufficient sample of patients (i.e., approximately 70 each) and used three tests to confirm cure to provide reliable data. The treatment success with clarithromycin-containing triple therapy was 42 and 65 per cent for tinidazole-containing triple therapy thus confirming the hypothesis that in Mumbai, legacy triple therapy provides unacceptably low treatment results. By contrast, in New Delhi the same regimen provided 80 per cent success after 14 days and 91 per cent after 21 days55. However, in this study the sample sizes were too small to provide a meaningful comparison between the 14 and 21 day groups. Without having concomitant susceptibility data, it is difficulty to suggest the mechanisms responsible for this surprising result. Most of the Indian studies have used less than optimal dosing, duration, or both and thus are not very helpful in deciding what to use and in which situation. The fact that sequential therapy provided 87 per cent treatment success in Puducherry suggests that the local resistance rate for clarithromycin was less than 20 per cent53.

Table III H. pylori treatment regimens in India

Which regimens are likely to be effective in India?

The available data from India do not provide the information one requires to prospectively identify a successful treatment regimen. India is ripe for prospective studies of different anti-H. pylori therapies. The investigators should follow the recommendations for efficient identification of effective regimens that allows one to expose the minimal number of subjects to a new regimen but at the same time, starts with one's “best shot” in terms of dose, duration, frequency of drug administration, etc23563. Treatment failures are very difficult to interpret unless one has susceptibility data. Such data do not need to be acquired before the trial but a biopsy or two can be saved frozen at -70C in transport media for subsequent culture and susceptibility testing locally or following shipment to a central laboratory in India or elsewhere.

If it were decided to routinely treat all H. pylori infections diagnosed in India, what regimen or regimens would be most likely be successful? Ideally, for an infectious disease, one would have information from clinical trials regarding which regimens were most effective (e.g., reliably provide 90% or greater treatment success) in that population as well as the effect of resistance to the antibiotics used both individually and in combination. The regimens would also have been optimized in terms of dose, drugs, and duration. This has been accomplished, at least partly, for most commonly used regimens in the West (Table IV and Box 2). In general, the results from one location are transferrable to another region provided the patterns of resistance are the same. Clearly in India, the problem of antibiotic resistance differs greatly from that in developed western countries. In addition, the pattern of resistance appears to vary widely and data regarding resistance in most areas are scant and not necessarily reflective of that population. In addition, although several trials have been done in India, most were small, rarely evaluated what are now known to be the optimal formulations, and often used lower than optimal doses and durations. An India-wide systematic approach is needed to identify the best regimen in general and specifically for each area. These results would be enhanced by simultaneous collection of susceptibility data and effectiveness from different populations throughout India.

Table IV Antimicrobials used for H. pylori eradication therapy

Close

Box 2

Recommended anti-H. pylori therapy

Export to PPT

Until accurate treatment data combined with susceptibility testing data are available, one must rely on first principles, (e.g., cannot overcome clarithromycin resistance or fluoroquinolone resistance by increasing either the dose or duration of therapy and these antibiotics are best avoided in regions where resistance is known to be high either from direct assessment or from a history of poor treatment success). In contrast, metronidazole/tinidazole resistance can be partly or largely overcome by using full doses (1,500 to 1,600 mg) and a 14 days of therapy. The quadruple therapy consisting of tetracycline, metronidazole, bismuth, and a PPI (Box 2) has been tested directly in metronidazole resistance and has generally had a success rate of 90 per cent or greater provided that therapy is at least 10 and preferably 14 days66–69. Bismuth quadruple therapy would be one of our first choices for testing. The reports of tetracycline resistance in India are troubling but the intragastric doses are very high and the effect of different levels of tetracycline resistance on outcome has not been assessed directly. An alternate would be the same combination but substituting furazolidone (or possibly secnidazole or even clarithromycin) for the metronidazole. The optimal dose of furazolidone is unknown; we use 100 mg t.i.d. with meals. A few caveats are necessary when using furazolidone. The drug is an monoamine oxidase inhibitor and thus interacts with many other drugs and foods (e.g., we provide patients with an information sheet that cautions against aged cheese, sausage including bologna, salami and pepperoni, lima beans, lentils, snow peas, and soybeans, canned figs and raisins, beer, ale and wine, licorice, soy sauce and any food product that is made with soy sauce, monamine oxidase inhibitors, phenylpropanolamine, ephedrine, and phenylephrine). A widely read letter to the editor stated that furazolidone had been removed from the US market because it was a proven carcinogen70. That information is erroneous; the author mixed animal food supplement regulations with human medicine. Furazolidone was originally sold in the U.S. by Roberts Pharmaceuticals which was purchased by Shire Pharmaceuticals which because there was almost no commercial market for the drug, chose not to continue to market the drug and notified the Food and Drug Administration (FDA) that they would stop marketing it. The FDA responded that they would “publish a notice in the Federal Register stating that you have voluntarily requested withdrawal of approval of these applications because you have stopped marketing the drug products under the NDA”. That notice was subsequently published (Federal Register / Vol. 70, No. 42 / Friday, March 4, 2005 / Notices, pp 11652). The letter to the editor described furazolidone as a carcinogen although it is listed by the IARC as a Group 3 carcinogen which is defined as “unclassifiable as to carcinogenicity in humans”70.

Clarithromycin-containing triple therapies are very effective regimens provided the appropriate doses and duration are used and local resistance to clarithromycin is low. The success rate with legacy triple therapy consisting of a twice a day PPI plus amoxicillin and clarithromycin or clarithromycin and metronidazole falls below 90 per cent when the level of clarithromycin resistance is 7 to 10 per cent38. Sequential therapy is an effective alternative until clarithromycin resistance exceeds 20 per cent and metronidazole resistance exceeds 40 per cent and then treatment success falls to unacceptable levels71. As such triple therapy should be avoided unless it were proven to be locally effective; there are likely few places in India where sequential therapy would provide 90 per cent or greater success. If one were to try a 4 drug clarithromycin containing regimen we suggest trying hybrid therapy (Box 2) as it has the best chance of being successful72. Alternatively, one could use bismuth-clarithromycin quadruple therapy but this would also need to be optimized for India (Box 2).

Development of India-specific regimens

The old concept of trial and error that was the norm for early H. pylori therapies has passed as it has become recognized that the same approaches used for other common infectious disease is likely to provide better results36. We now recognize that treatment results are predictable provided we know the pattern of resistance in a community. This allows us to quickly identify and optimize, (or reject) potential regimens, exposing the fewest patients to ineffective therapy. Simply trying some combination in an arbitrary number of subjects is no longer considered rational or ethical. For example, secnidazole has been used with some success in India although attempts to optimize its use and identify its effectiveness in the presence of metronidazole resistance are lacking. India has been on the forefront of suggesting new antibiotic formulations that differ in their intragastric time7374. However, this idea has not been advanced further with in vivo testing.

Recurrence of H. pylori after treatment

A high rate of reinfection after eradication in developing countries would undermine attempts to eradicate the infection in individuals, families, villages, or larger groups of individuals. Recurrence of infection following what appears to have been successful therapy is classified as either recrudescence, a form of failed therapy in which the same H. pylori strain is still present, or as reinfection when a different strain is present75. These definitions are general as recrudescence can actually be reinfection from a common source and reinfection from outgrowth of a minor population of a mixed infection. Recurrence, either reinfection or recrudescence, is more common following low treatment success regimens suggesting that most cases represent recrudescence. Recurrence within 12 months of treatment in developed countries is generally due to recrudescence76. Since sanitation in developed countries is generally good there and the prevalence of H. pylori is low, particularly in children, re-exposure is less problematic. However, in developing countries sanitation is often suboptimal and the presence of the infection is almost universal thus the potential for reinfection is high. Whether this is an actual problem in India is unclear (Table V) and is an excellent area for research. Studies in India in which individuals or families are followed prospectively using non-invasive tests after successful eradication are needed. Molecular characterization of the infecting strains such as RFLP (restriction fragment length polymorphism) checked before treatment and following recurrence would be needed to help unravel this problem and discover whether recurrence following eradication is a problem that must be addressed or mostly a theoretically concern768182. Routine endoscopy for surveillance should probably be avoided to reduce the possibility of iatrogenic reinfection from confusing the picture. One would imagine that the population at highest risk would be one living in poverty in relatively unsanitary conditions without access to potable water. Data over a longer interval (e.g., 5 year) regarding reinfection from the Indian subcontinent are needed in populations that eradication was definitely confirmed using 2 or 3 tests or a single test (e.g. urea breath test) on at least 2 occasions.

Table V Recurrence of H. pylori infection studies from India and Bangladeshi

Recommendations for the future

India is an ideal place for investigation of new ideas in a systematic and organized way as the infrastructure is present, the pharmaceutical industry is among the most innovative, and the medical expertise highly regarded and up to date. Questions that might be addressed include what is the recurrence rate of H. pylori infection after successful eradication. If the recurrence rate is high, what factors are responsible and how can it be reduced (e.g., the source of reinfection and whether it can be managed) and whether eradication of the infection from entire families or villages would eliminate recurrence. The high level of antimicrobial resistance allows direct study of agents in the presence of different patterns of resistance. It is imperative that susceptibility be assessed accurately and, if required laboratory personnel or individual technicians can be trained in the details of susceptibility testing for H. pylori and in designing treatment trials to investigate current and innovative treatment regimens. India should be leading the way in the research of treatment of H. pylori infections. Ultimately, eradication of H. pylori from India and other developing countries may need to await development of an effective vaccine83.