Objective: To systematically review studies examining the relationship between hyperemesis gravidarum and Helicobacter pylori (H pylori) infection.
Data sources: A 1966 to January 2007 search using MEDLINE/PubMed, EMBASE, and Web of Science included MeSH terms: "Helicobacter pylori," "Helicobacter infections," "hyperemesis gravidarum," and the text words "nausea," "vomit," "pregnancy," and "Helicobacter." References of selected papers were examined for additional relevant studies.
Methods of study selection: We evaluated studies investigating a relationship between hyperemesis gravidarum and H pylori infection. Studies were included in which the diagnosis of hyperemesis gravidarum was made at or before entry into the study, and H pylori diagnosis was made by serum antibody sample, gastric biopsy, saliva test, or stool sample. The search produced 169 titles; 22 were reviewed in further detail.
Tabulation, integration, and results: Fourteen case-control studies met established criteria, involving 1,732 participants and controls tested for H pylori infection. Studies were evaluated according to patient demographics and study methodology (case definition, exclusion criteria, H pylori testing). An estimate of the odds ratios with 95% confidence intervals was calculated by using a random effects model for dichotomous variables with review article software. Ten studies showed a significant association between hyperemesis gravidarum and H pylori infection. Odds ratios varied from 0.55 to 109.33; three results were less than 1.0. Tests for heterogeneity applied to several subgroups were considerable with values above 75% for all groups.
Conclusion: An association between hyperemesis gravidarum and H pylori infection is suggested by this systematic review. However, the considerable heterogeneity among studies highlights study limitations.