Helicobacter heilmannii Gastritis Caused by Cat to Child Transmission

Helicobacter pylori is one of the most common causes of chronic gastritis in adults and children. Recently, another microorganism within the Helicobacter species has been described causing gastritis:H. heilmannii, formerly known as Gastrospirillum hominis (1–5). H. heilmannii infection rarely affects children: an overall incidence of H. heilmannii gastritis in children of 0.3% to 0.4% has been described (6,7). However, in domestic cats and dogs the frequency of H. heilmannii infection is as high as 80% to 100% (8–10). Therefore, H. heilmannii infection has been postulated to be a zoonosis (1,11), especially since humans with this infection often live in close contact with animals. Proof of this hypothesis is lacking so far. Here we present a 5-year-old boy with H. heilmannii gastritis, and demonstrate by sequence analysis that the causative H. heilmannii strain was also present in both of his cats. Given the incidence of H. heilmannii infection in humans and cats, we propose that our patient was infected by one of his cats. CASE REPORT A 5-year-old boy presented with a history of failure to thrive (weight for height deviated from 0 SDS at 3 years of age to −2 SDS at presentation) and several years of abdominal pain. Because the pain had worsened in the past two months, he was unable to attend school. The pain was not related to eating or defecation. Defecation frequency and consistency had been irregular for several years. He lived in close contact with his two cats. On the presumptive diagnosis of celiac disease, the referring pediatrician obtained anti-endomysium and anti-gliadin antibodies, which were negative and borderline positive, respectively. Therefore the patient was referred to our hospital for gastroduodenoscopy and duodenal biopsies. Because of the chronic nature of his complaints, we performed some additional laboratory investigations which revealed no anemia or elevated ESR. On gastroduodenoscopy, the stomach showed a slight hyperemic mucosa without ulcers or micronodular pattern. The duodenum appeared normal. Gastric tissue samples from the antrum were obtained for rapid urease test (CLO-test™, Tri-Med Specialties Inc., Australia), histology, and H. pylori culture. Duodenal tissue was obtained for histology and evaluation of Giardia lamblia. The biopsy specimens were fixed in 10% neutral buffered formalin. The CLO-test™ was highly positive within 30 seconds. Light microscopic examination of antral mucosa showed chronic inflammation in the lamina propria with infiltration of poly-morphonuclear leukocytes. Routine staining revealed gram-negative long, thin, spiral-shaped microorganisms in the mucus, which we identified as H. heilmannii (Fig. 1). These microorganisms are larger and morphologically easily distinguishable from H. pylori, which are gram-negative, short, comma-shaped rods (Fig. 2). A bacterial culture from antral tissue, with conditions specific for H. pylori, was negative. On histology and cultures, Giardia lamblia trophozooites were absent from duodenal tissue.FIG. 1.: Helicobacter heilmannii bacteria presenting themselves in the described patient as gram-negative, long, thin, spiral-shaped microorganisms in a gastric pit.FIG. 2.: Helicobacter pylori bacteria in another patient: gram-negative, short, comma-shaped rods in gastric mucus.Gastroscopy was performed (E.J.H.) on the boy's two cats. No macroscopic abnormalities were found in the esophagus and stomach. Tissue samples were obtained from the gastric antrum, corpus, and fundus for CLO-test™, histology, culture, and for polymerase chain reactions (PCR). The CLO-test™ was positive in both cats. Light microscopy revealed the presence of H. heilmannii in gastric tissue of both cats, without any signs of inflammation. H. pylori culture was negative. DNA from the patient's gastric biopsy, and DNA from fresh feline gastric biopsies was isolated using the QIAGEN DNeasy Tissue Kit, according to the manufacturer's instructions. A nested PCR was performed with primers described by Cattoli et al. (12), using water as a negative control template. In brief, two primers that can amplify the ureAB intergenic region from several gastric Helicobacter species (BRUrABrev, 5´-CRA AYT TRA KAT CTT CNC CAT AA-3´ and BRUrABfor 5´-TSG GNT CAC AYT TCC AYT TCT-3´) were used for a first amplification. A second PCR was performed using primers HbizUrABrev 5´-TTT CGA GAG ATT TTT TTC ATC-3´ and HbizUrABfor 5´-TTC AAA TTG GCG GTA ACC-3´, which can amplify an approximately 175 bp product. Presence or absence of products was checked by agarose gel electrophoresis. A 175 bp product was obtained from the gastric samples of the patient and both cats, while the control remained negative. Sequence analysis of the ureAB intergenic region of the Helicobacter DNA from patient and cat biopsies was performed by direct cycle sequencing on the nested PCR product, using dye terminator chemistry. Analysis was performed on an automatic sequencer (model 373, Applied Biosystems Inc. (Foster City, CA)) according to the instructions supplied by the manufacturer. The sequences of the nested PCR products obtained from biopsies of patient and the two domestic cats were 100% identical. This suggests that patient and cats were infected with the same Helicobacter isolate. During and after treatment for ten days with triple therapy (amoxicillin, clarithromycin, and omeprazole) the patient's complaints of abdominal pain subsided and appetite increased, while defecation consistency and frequency normalized. Subsequently, he started to gain weight. However, three months after therapy his complaints of abdominal pain, poor appetite, and irregular defecation had returned. A gastroscopy was repeated and showed no macroscopic abnormalities. CLO-test™ was negative and H. heilmannii-like organisms were not detectable in antral tissue by light microscopy. Furthermore, IgG-anti-Helicobacter pylori antibodies were negative. Despite the boy's young age, the suspicion of irritable bowel syndrome arose. Appropriate therapy, consisting of increased fluid intake, increased dietary fibers, and physical exercise, was instigated successfully. Long-term follow-up showed no relapse. DISCUSSION H. heilmanni has been recognized for over a century as a common microorganism in the stomach of various mammals such as cats, which causes no symptoms or inflammation (8–11). The presence of H. heilmannii in humans was first described in 1906 (13), but its capability of causing gastritis (14) and its association with MALT-lymphomas (15) in humans was not recognized until recently. Infection with H.heilmannii has been postulated to be a zoonosis (1,11), especially since humans with this infection often live in close contact with animals. However, proof of this hypothesis is still lacking, although Dietrich et al. (1) found a high degree of similarity between human and feline isolates. We described a 100% identical H. heilmannii strain in a symptomatic boy and his pets. Although this finding does not automatically implicate that the route of transmission is from cat to human, we postulate that the cats infected our patient, considering the very low incidence of H. heilmannii infection in humans and the high incidence in animals. As for H. pylori, a fecal-oral route of transmission seems likely. Similar to our case, Dietrich et al. (1) failed in culturing H. heilmannii. This is a recurrent diagnostic problem, as only Andersen et al. (16) has ever succeeded in culturing an H. heilmannii strain. The 13C urea breath test and CLO-test™ detect urease and therefore detect the different Helicobacter species, but do not discriminate between them. The H. pylori stool antigen test (HpSA™), which is a microplate enzyme-immunoassay for the qualitative detection of H. pylori antigens in feces (17), has not yet been developed specifically for H. heilmannii. Therefore, at this moment, morphologic recognition in histologic slides is the most practical way of discriminating between H.pylori and H. heilmannii infection. We suggest that in any patient with Helicobacter infection and cats or dogs in the same household, histologic examination of the gastric mucosa should also be focused on H. heilmannii. Because cats have an infection rate of almost 100% (8–10), it is controversial whether endoscopy or antibiotic treatment should also be applied to the animals involved. The effect of diagnostic and therapeutic interventions in pets on the reinfection rate in humans must be established in future studies. However, in patients with recurrent H. heilmannii infection, treatment of both patient and pets can be considered.