24. Jones RN, Erwin ME, Bale M: New insights into the activity of third-generation cephalosporins against pneumonia-causing bacteria. Diagnostic Microbiology.
23. Sessegolo JF, Levin AS, Levy CE, Asensi M, Facklam FIR, Teixeria LM: Distribution of serotypes and antimicrobial resistance of Streptococcuspneumoniae strains isolated in Brazil from 1988 to 1992. Journal of Clinical Microbiology 1994, 32: 906-911. 24. Jones RN, Erwin ME, Bale M: New insights into the activity of third-generation cephalosporins against pneumonia-causing bacteria. Diagnostic Microbiology and Infectious Diseases 1992, 15: 73-80.
25. BradleyJS, ConnorJD: Ceftriaxone failure in meningitis caused by Streptococcuspneumoniaewith reduced susceptibility to beta-lactam antibiotics. Pediatric Infectious Diseases 1991, 10: 871-873. 26. Kalager T, Andersen BM, Bergan T, Brubakk O, Bruun JN, D~skeland B, Helium KB, Hopen G, vonder Lippe E, Rahm V, Ritland S, Schreiner A: Ciprofloxacin versus tobramycin/cefuroxime combination in the treat-
ment of serious systemic infections. A prospective, randomized and controlled study of efficacy and safety. Scandinavian Journal of Infectious Diseases 1992, 24: 637-646.
Presence of Helicobacter pylori in the Oral Cavity, Oesophagus, Stomach and Faeces of Patients with Gastritis E N a m a v a r l*, R. R o o s e n d a a l 2, E . J . K u i p e r s 3, P. d e G r o o t 4, M.W. v a n d e r BijI 1, A.S. P e f i a 3, J. d e G r a a f f I
The presence of Helicobacter pylori in the oral cavity (6 sites), oesophagus, stomach and bowel of 20 dyspeptic patients was investigated. Samples were cultured on three selective media and analyzed by 16S rDNA polymerase chain reaction (PCR) and southern hybridization. Helicobacterpylori DNA was detected by PCR from oral-cavity samples of three (20 %) and from faeces samples of only one (7 %) of the patients whose stomach biopsies were positive for Helicobacter pylori. When culture was used, the I Department of Medical Microbiology, Free University, van der Boechorststraat 7, 1081 BT Amsterdam, The Netherlands. 2 Department of Clinical Microbiology and 3Department of Gastroenterology, Free University Hospital, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands.
4Department of Oral Microbiology,AcademicCentre for Dentistry, Amsterdam, van der Boechorststraat 7, 1081 BT Amsterdam, The Netherlands.
Eur. J. Clin. Microbiol. Infect. Dis.
microorganism's rate of recovery from the oral cavity and faeces was 13 % and 7%, respectively. One patient had a Helicobacter pylori-like organism in samples collected from the tongue and palate. Both strains were urease, catalase and oxidase positive and grew microaerophilically but were negative on PCR analysis. This demonstrates the possibility of false identification of Helicobacter pylori by use of routine enzyme reactions. Interestingly, specimens collected from the cheeks of three patients were positive for Helicobacterpylori by PCR analysis. This is the first instance of detection of this microorganism in the cheek.
Helicobacter pylori is now accepted as an important cause of chronic gastritis and as an aetiological agent in the pathogenesis of most peptic ulcers, particularly duodenal ulcer (1, 2). There are reports of the sporadic recovery of Helicobacterpylori from the oral cavity and faeces. In most studies, only dental plaque, saliva or both have been investigated, and the microorganism was detected using either culture on selective media or amplification by polymerase chain reaction (PCR). The purpose of this study was to investigate the presence of Helicobacter pylori in the oral cavity, oesophagus, stomach and bowel of dyspeptic patients using culture and PCR analysis in parallel. Materials and Methods. Twenty consecutive patients referred for upper gastrointestinal endoscopy and for dyspeptic complaints were included in this study. Fourteen men and six women, with a mean age of 46 years (range 32 to 72 years), participated in the study. After informed consent was obtained, all patients were first referred to the dental unit of the hospital. Subjects who had used antibiotics, proton pump inhibitors or bismuth in the preceding three months were excluded as were those who had any contraindication for biopsy. Fully edentulous patients were also excluded. The general condition of the oral cavity was examined by a dentist for decayed, missing and filled teeth, pocket depth, bleeding on probing, full or partial denture, inflammation, aphthae, mucositis, abscesses and fistulas. Samples were obtained from the subgingival and supragingival plaque with a curette and placed in 1 ml reduced transport fluid (RTF) (3). Palate, tongue and cheek samples were collected from patients by rubbing the mucosa with a swab which was suspended in 1 ml RTF. Unstimu-
Vol. 14, 1995
lated saliva was collected in a sterile plastic tube. Specimens were transported to the laboratory and processed within 1 h. After specimen collection fyom the oral cavity, patients underwent oesophago-gastroduodenoscopy. Biopsy specimens Were taken from antrum, corpus and oesophagus for microbiological and histological examination. Patients were asked to deliver their faeces sample as soon as possible. The mean delay between defaecation and arrival of the specimens at the laboratory was 1 h. In a separate study with laboratory strains, no loss of viability was found within this time interval for samples collected from both the oral cavity and from the faeces. Oral cavity specimens were homogenized, and 50 gl was used for PCR analysis. The homogenares were diluted with phosphate buffered saline (PBS), and serial dilutions were cultured on four freshly prepared media: non-selective blood agar ase no. 2 supplemented with 5 gg/ml haemin, 2 gg/ml menadione and 7 % horse blood lysed With saponin; selective Belo Horizonte medium, Skirrow's medium and Dent's medium, all supplemented with 40 mg/1 2,3,5-triphenyltetrazolium chloride (4-6). Faecal samples were emulsified to produce a dense suspension (5 g faeces/20 ml PBS). This SUspension was divided into two portions. One Portion was centrifuged at 100 x g for 5 rain, serial dilutions of the supernatant were cultured on selective media and 2.5 ml of this supernatant was used for PCR. The other portion was sieved through a 250 gm pore filter, the filtrate was centrifuged and the pellet was washed once at 20,000 x g for 30 min (7). Samples of the bacterial layer were plated on selective media. All plates Were incubated microaerophilically in an Anoxomat-system WS 8000 (Mart, The Netherlands) and were read on days 3, 5 and 7. Biopsies from oesophagus, antrum and corpus Were cultured on non-selective blood agar and selective Belo Horizonte agar by rubbing the Specimen several times over the agar surface. The remaining part of the specimen was placed in lysis buffer (10 mM Tris-HC1, pH 8.0, 0.5 % Tween 20) for PCR (8). Suspected Helicobacter pylori colonies from all cultures were identified by morPhology, urease production, catalase and oxidase positivity and PCR analysis followed by Southern hybridization. The specimens for histopathologieat examination were processed by standard methods; haematoxylin-eosin and modified Giemsa stained slides were examined for the preSence of gastritis and Helicobacter pylori.
Twice concentrated lysis buffer was added to 50 gl of the oral specimens to give a final concentration of 10 mM Tris-HCl, pH 8.0 and 0.5 % Tween 20, and samples were frozen at -20°C..For pretreatment of samples, oral, oesophageal and gastric tissue specimens were thawed and proteinase K was added to achieve a final concentration of 0.5 mg/ml. After incubation for 60 min at 55°C and heating for 10 rain at 100°C, 10 lxl was used for the PCR. To 2.5 ml of the faecal suspension, 2.5 ml of 20 mM Tris-HCl (pH 8.0) were added, and the samples were frozen. After thawing, sodium dodecylsulfate and proteinase K were added to 400 gl of faecal suspension to achieve a final concentration of 1 % and 1 mg/ml, respectively. Samples were incubated at 55°C for 60 min, and DNA was isolated by a final extraction according to a standard procedure (9). Ten gl of the serial tenfold dilutions of the isolated DNA was tested by PCR. PCR protocol was carried out as described elsewhere (9). The sensitivity of the PCR was high because approximately 10 to 100 cfu per reaction could be detected in oral and faecal samples.
Results and Discussion. Endoscopic examination revealed that nine of the 20 subjects had no mucosal abnormalities, and 11 of them showed signs of gastritis or duodenal ulcer. Three of the nine subjects had normal histology, and all were PCR and culture negative. Histological signs of chronic active gastritis were found in 17 of the 20 subjects, and this is in agreement with other studies that found no good correlation between endoscopy and histological features (10). Fifteen of the 20 patients had Helicobacterpylori in the gastric mucosa as indicated by a positive PCR result for the antrum, corpus or both. Of these 15 patients, 13 Results of endoscopy, histological investigation, PCR and culture for stomach specimens of 20 dyspeptic patients.
Endoscopy result Histology*
No abnormalities (n=9)
(n=5) Duodenal ulcer (n=6)
*Signs of chronic active gastritis.
Eur. J. Clin. Microbiol. Infect. Dis.
Table 2: Distribution of Helicobacterpyloriin six oral-cavity sites and in the faeces of seven positive patients. Patient no. Site 3 Saliva Tongue Cheeks Supragingiva Subgingiva Palate Stomach Faeces
. . +b +~,b
. . . _ --
. . .
+n +a . . . . . . . . +b,c _ -+a +a,b + a , b +~,b +.,b
a Positive by PCR. b Positive by culture. Strain was positive for urease, catalase and oxidase but negative by PCR analysis with primers directed at three "Helicobacter pylori specific targets. a Positive by PCR analysis but negative by Southern hybridization.
(87 %) were positive by culture which indicates that P C R is slightly more sensitive than culture for gastric biopsies (Table 1). We looked for the presence of Helicobacterpylori at six sites in the oral cavity and in the faeces of the study subjects (Table 2). Helicobacter pylori was detected in oral cavity of four subjects and in the faeces of one subject by PCR and Southern hybridization. When culture and identification by enzyme activity, microaerophilic growth and Gram stain were used, two patients (no. 3 and no. 5) were positive for Helicobacter pylori, patient no. 3 both in the oral cavity and the faeces and patient no. 5 only in the oral cavity. Oral (subgingival plaque) and faecal specimens from patient no. 3 were positive for Helicobacter pylori by culture but were negative by PCR analysis. Although PCR analysis is more sensitive than culture, the possibility of a false-negative result with PCR still exists. The strain isolated from the subgingival plaque of patient no. 3 was positive by PCR analysis and Southern hybridization, whereas the strain isolated from faeces was only positive by PCR analysis but was negative by hybridization with a targetspecific oligonucleotide probe. HeIicobacter pylori was detected in a specimen from the tongue of patient no. 4 by PCR. This patient had no gastritis, and histological examination showed no signs of inflammation. All the gastric specimens, faeces and oral cavity were negative for Helicobacterpylori by PCR. Water controls were used to exclude the possibility of PCR contamination. This may suggest the presence of a specific receptor for this microorganism in the mouth but not in the stomach of this patient. Patient no. 5 had a positive Helicobacter pylori culture in specimens
collected from tongue and palate. These strains were urease, catalase and oxidase positive and grew microaerophilically but were negative on PCR analysis directed at 16S rDNA. These strains were also negative using two other targetspecific primers (urease gene and a Helicobacterspecific cloned fragment). This highlights the possibility of false identification of Helicobacter pylori using routine identification methods.
Helicobacter pylori was detected in the faecal specimens of only two patients, one by PCR analysis and the other by culture; this is in close agreement with the finding of Van Zwet et al. (11) who found no Helicobacter pylori target D N A in the stools of 24 Helicobacter pylori positive patients. In the study of Shimada et al. (12), Helicobacterpylori D N A was detected in faecal specimens of only one of 79 patients. However, these results are not in agreement with another study that found Helicobacterpytori D N A in the faecal specimens from 90 % of patients suffering from gastritis (13). A saliva specimen from one patient was positive for Helicobacter pylori by PCR analysis but was negative by culture. Interestingly, three of the seven patients had Helicobacter pylori D N A in samples taken from their cheeks. This is the first detection of this microorganism from this site. There was no relationship between the patient's dental condition and the presence of this microorganism. In conclusion, this study shows that Helicobacter pylori may contaminate the mouth and occasionally be found in the faeces. Therefore, the use of oral and faecal screening tests for the detection of Helicobacter pylori do not seem to be adequate for diagnostic purposes. However, these findings do suggest the possibility of the oral-oral as well as faecal-oral route of transmission.
References 1. Marshall BJ, Warren JR: Unidentified curved bacilli in the stomach of patients with gastric and peptic ulceration. Lancet 1984, i: 1311-1315. 2, Rauws EAJ, Tytgat GJN: Cure of duodenal ulcer associated with eradication of Helicobacter pylod. Lancet 1990, 335: 1233-1235. 3. Syed SA, Loesche WJ: Survival of human dental plaque flora in various transport media. Applied Microbiology 1972, 24: 638-644. 4. Skirrow MB: Campylobacter enteritis: a new disease. British Medical Journal 1977, 2: 9-11.
Vol. 14, 1995
5. Dent JC, McNulty CAM: Evaluation of a new selective medium for Campylobacter pylori. European Journal of Clinical Microbiology & Infectious Diseases 1988, 7: 555-568. 6. Queiroz DMM, Mendes EN, Ftocha GA: Indicator medium for isolation of Campylobacter pylorL Journal of Clinical Microbiology 1987, 25: 2378-2379. 7. Thomas JE, Gibson GR, Darboe MK, Dale A, Weaver LT: Isolation of Helicobacterpylorifrom human faeces. Lancet 1992, 340: 1194-1195. 8. Roosendaal R, Kuipers EJ, Van Den Brule AJC, Pefia AS, Uyterlinde AM, Walboomers JMM, Meuwissen SGM, De Graaff J: Importance of the fiberoptic endoscope cleaning procedure for detection of He/iCobacter pylori in gastric biopsy specimens by PCR. Journal of Clinical Microbiology 1994, 32: 1123-1126. 9. Maniatis 1, Fritsch TEF, Sambrook J: Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1982. 10. Khakoo SI, Lobo AJ, Shepherd NA, Wilkinson SP: Histological assessment of the Sydney classification of endoscopic gastritis. Gut 1994, 35: 1172-1175. 11. Van Zwet AA, Thijs JC, Kooistra-Smid AMD, Snijder JAM: Use of PCR with feces for detection of Helk CObacterpylori infection in patients. Journal of Clinical Microbiology 1994, 32: 1346-1348. 12. Shimada B, Ogura K, Ota S, Yerano A, Takahashi M, Hamada E, Omata M, Sumino S, Sassa R: Identification of Helicobacter pylori in gastric specimens, gastric juice, saliva and faeces of Japanese patients. ancet 1994, 343: 1636-1637. 13. MapstoneNP, Lynch DAF,Lewis FA, Axon ATR, Tompkins DS, Dixon MF, Quirke P: PCR identification of Helicobacter pylori in faeces from gastritis patients. Lancet 1993, 341: 447.
Aeromonas hydrophila Myonecrosis Accompanying Mucormycosis Five Years After Bone Marrow Transplantation A.E. Moses1,, M, Leibergal 2, G. Rahav 1, M, Perouansky 3, R, Or4, M. Shapiro I
36"year-old man, five years after bone marrow ansplantation for aplastic anemia, was adrnitted with myonecrosis of the forearm after he had immersed his hand in sewage water several days prior to his admission. Blood cultures and
The WilliamCollinsDepartment of ClinicalMicrobiology and Infectious Diseases, 2Department of Orthopedics, ~epartment of Anesthesiologyand Intensive Care and e.partmentof Bone MarrowTransplantation, Hadassah mversity Medical Center, Jerusalem 91120, Israel.
specimens taken from the necrotic tissue of the arm all grew Aeromonas hydrophila. Following extension of the infection, the patient underwent amputation of the arm but ultimately died of cerebral mucormycosis. The epidemiology of Aeromonas infections is discussed and the literature of Aeromonas myonecrosis is reviewed.
The genus Aeromonas includes gram-negative, facultative anaerobic bacilli composed of several species which belong to the family Vibrionaceae. There are two major groups: the motile mesophilic aeromonads and the non-motile psychrophilic aeromonads. Whereas, the latter are of no known pathogenic significance in humans, the mesophilic aeromonads, also known as the Aeromonas hydrophila complex, produce a broad spectrum of disease in humans and are commonly isolated from the environment, mostly from water reservoirs. Aeromonas hydrophila has been implicated as the causative organism of soft tissue infection including cellulitis and myonecrosis (13). Contact with water has been recognized as a risk factor for infection (4). We describe a patient who five years after allogeneic bone marrow transplantation developed infection with Aeromonas hydrophila that produced myonecrosis of the upper limb which necessitated amputation and was associated with the patient's death. Case Report. The patient was a 36-year-old male who had T-cell depleted, allogeneic bone marrow transplantation five years earlier for severe idiopathic aplastic anemia. There was no evidence of acute or chronic graft versus host disease. Three years later he developed splenomegaly, hypoalbuminemia and mild hepatocellutar enzyme abnormalities, accompanied by moderate pancytopenia secondary to hypersplenism. Two days before admission fever, rigors and pain in the right arm developed. One week earlier, while repairing a neighbor's nonfunctioning sewer pipe, he had immersed both upper extremities in sewage but had sustained no obvious injury. On examination the patient appeared ill. The temperature was 39°C. Blood pressure was 110/80 mmHg. The right arm was swollen, red and tender with multiple bullae on the forearm (Figure 1). There was swelling of the elbow suggestive of right oleeranon bursitis. Peripheral pulses in the arm were absent, and compartment syndrome was diagnosed. Gram stain of fluid aspirated from the olecranon bursa disclosed numerous gramnegative rods. Radiograph of the arm showed air