Serological Status Can Confirm Cure of Helicobacter Infection

abstract & commentary

Synopsis: In the large majority of asymptomatic patients with H. pylori infection, the eradication of the pathogen reduces, often to undetectable levels, the IgG antibody response and also ameliorates the inflammatory response in the stomach.

Source: Feldman M, et al. JAMA 1998;280:363-365.

There are only a few bacterial infections, such as syphilis, that we follow with serological responses. Loss of serological response—seroreversion—may be an indication of cure but studies of such a seroreversion are lacking. Serology has become a mainstay of diagnosis in Helicobacter infection. The clinical meaning of seroreversion in this infection, particularly after therapy, awaits definition.

In this study from Dallas and the veterans population served by the University of Texas Health Science Center, 23 adults (10 men and 13 women) were followed prospectively during 1993-1994. To qualify for the study, the patients needed to have IgG antibodies to H. pylori in serum, as well as organisms seen in a baseline biopsy of gastric mucosa. Notably, patients had no symptoms of peptic ulcer disease.

After meeting these criteria, patients were treated for two weeks with triple therapy consisting of tetracycline, bismuth subsalicylate, and metronidazole. Compliance was strengthened by a telephone call midway during therapy.

Patients underwent serological testing and biopsy again at one month, three months, and 18 months. The gastric biopsies were obtained by a novel method, through a nasogastric tube under floursocopy requiring no sedation. Gastric inflammation was graded on a 0-3 scale. Patients with seroreversion were considered cured when biopsies from both the gastric body and the antrum showed no organisms 18 months after therapy.

Patients in the study had a mean age of 48 years. Cure of H. pylori infection occurred in 65% (15) of subjects. All l8 failures had persistent infection at months one and three. There was a significant difference in gastritis scores between the cured and failure groups at both gastric body and gastric antrum sites when month three and month 18 scores were compared to baseline.

For those patients who had a decrease of antibodies to an undetectable level (< 15 U/mL) at 18 months, seroreversion had a sensitivity of 60% and a specificity of 100% for ascertaining cure of H. pylori infection.

Comment by Joseph F. John, MD

Billions of people worldwide are infected with H. pylori, resulting in immense morbidity and some mortality. This study by Feldman and associates shows that in the large majority of asymptomatic patients with H. pylori infection, the eradication of the pathogen reduces, often to undetectable levels, the IgG antibody response and also ameliorates the inflammatory response in the stomach.

How can we use these data? Recall that only 60% of the cured patients reduced their serology to undetectable levels. So perhaps at least 40% of appropriately treated patients will have measurable antibody and still be cured. For a chronic bacterial infection, that is not bad, i.e., we can extrapolate from the study and say that if after one year our treated patients no longer have IgG to H. pylori, they are likely cured. The remaining patients should undergo gastric biopsy or a urea breath test to pick up the additional 40% of cures.

Feldman et al rightfully point out that 80-90% of currently FDA-approved regimens effectively treat H. pylori infection (Gastroenterology 1997;113:S126-S130). So perhaps even more patients who are cured will serorevert. The study was done at a VA Medical Center, and concommitant diseases like COPD and alcoholism were not treated as variables.

Additional studies should look at non-veteran populations and those containing more women and children who undergo therapy for H. pylori infection. Recent data from a Japanese cohort of children and adults followed over eight years for acquisition vs. loss of H. pylori gives us some insight about seroreversion in untreated populations.1 They found that seroreversion rates slightly exceeded seroconversion rates. The seroconversion rates per year were 1.1% and 1.0% for children and adults respectively compared to seroreversion rates of 1.8% for children and 1.5% for adults. Therapy of infection as shown in the Feldman study clearly results in a rate of seroreversion far in excess of seroreversion rates in untreated patients like the ones in the Japanese study.

H. pylori serological screening, in my opinion, should be a part of the "health profile" for all adults. Studies like this one by Feldman et al show that a positive serology is highly associated in many patients with smoldering H. pylori infection. The effective erradication of the offending pathogen in time will result in a seroreversion or marked decrease of anti-H. pylori antibodies, both of which will relate to a high rate of cure. The mechanism of this relatively short-lived immunologic memory for H. pylori antigen awaits to be elucidated.


1. Kumagai T, et al. J Infect Dis 1998;178:717-721.