To characterize posttreatment recurrence of Helicobacter pylori in Peru, 192 adults with H. pylori-positive gastric biopsy specimens were monitored by (14)C-Urea breath test, after eradication of H. pylori by use of amoxicillin, clarithromycin, and omeprazole. The cumulative risk of recurrence at 18 months was 30.3% (95% confidence interval, 21.4%-39.3%). Randomly amplified polymorphic DNA patterns and DNA sequence data established that, among 28 pairs of H. pylori isolates from pretreatment and recurrent infections, 6 (21%) were genetically similar, suggesting recrudescence of the previous infection, and 22 (79%) were different, suggesting reinfection with a new strain that differed from that involved in the initial infection. Eating mainly outside of the home was a risk factor for infection with a new strain (adjusted relative risk [RR], 5.07), whereas older age was a protective factor (adjusted RR, 0.20). Although an increase in the anti-H. pylori IgG antibody titer corresponded to recurrence, pretreatment and recurrent infections were similar with respect to quantitative culture colony counts and histologic characteristics, suggesting that neither prior eradication nor the memory immune response measurably alters the risk or burden of recurrent infection. Although eradication with antibiotics was successful, the high rate of reinfection suggests that treatment is unlikely to have a lasting public health effect in this setting.