This is the first report of a major foodborne outbreak of enterohaemorrhagic Escherichia coli (EHEC) in Sweden. It occurred among the nursing staff at a children's hospital with approximately 1600 employees. Contaminated lettuce was the most likely source of infection. Nine persons were culture-positive for Escherichia coli (E. coli) O157 and verocytotoxin-positive by PCR and a further two were verocytotoxin-positive by PCR only. All 11 EHEC-positive individuals had attended a party for approximately 250 staff members, which was held at the hospital. In a questionnaire 37 persons stated that they had symptoms consistent with EHEC infection during the weeks after the party. There was no evidence of secondary transmission from staff to patients. The value of PCR as a sensitive and fast method for diagnosis is discussed in this paper. Pulsed-field gel electrophoresis (PFGE) was used to ascertain that staff members were infected by the same clone, and that two patients with E. coli O157 infection were not.

BACKGROUND:The aim of our study was to test the feasibility of culturing Helicobacter pylori directly from biopsies aimed for rapid urease test in routine clinical practice. MATERIALS AND METHODS:In 260 consecutive patients referred for gastroscopy because of dyspepsia one antral biopsy was routinely used for our "in house" rapid urease test (RUT). Positive biopsies were placed in a transport medium and sent to the laboratory. The biopsies were cultured and incubated at 37 degrees C for 5-7 days. H. pylori was identified and routinely tested for antimicrobial resistance by using the E-test. RESULTS:In 118 out of 260 patients (45%) the urease test turned positive and the growth of H. pylori was sufficient to allow testing of antimicrobial resistance. CONCLUSION:H. pylori could be cultured from almost all positive RUT specimens. A liquid RUT is thus more suitable for culture, saving additional biopsies.

AIM:Primary resistance of H pylori strains to antimicrobials has direct therapeutical implications. For this reason we studied the prevalence of primary resistance of H pylori of Sweden, in a regional setting. MATERIAL AND METHODS:A total of 244 patients referred to our endoscopy unit have been evaluated prospectively for the occurrence of H pylori. Helicobacter pylori infection was established by a positive rapid urease test on an antral biopsy. If positive the biopsy specimen was transferred to a transportmedium for culture on a Skirrow medium. Antimicrobial resistance was evaluated by the E-test. RESULTS:In 109 patients H pylori infection was seen and in all of them the culture was positive. No resistance was seen for amoxycillin and tetracycline. The resistance against metronidazole was 40.3% and and against clarithromycin it was 2.8%. For metronidazole there was a significantly higher resistance seen in women compared to men (53.3% vs 31.2%; p < 0.001). CONCLUSIONS:Primary resistance to metronidazole was higher than expected and over average in Europe. The resistance for clarithromycin is still rare. Further monitoring is mandatory to detect changes in the community.

Bacterial growth in relation to suction drains and suction drain tubes following total hip replacements was studied in 105 consecutive patients. All the patients were operated on in a Charnley Howorth Mark II sterile enclosure without walls, but with a body exhaust system and routine antimicrobial prophylaxis. Specimens for culture were taken from the suction drain tip, from inside the suction tube and the incision site. Positive cultures were found in a7 patients; in 20 of them from either the drain tip or the drain track. In 15 patients, the culture revealed coa-gulase-negative staphylococci; in 12 of them from the drain tip or the drain track. Eight patients had a positive culture from the tip of the drain and 16 from the drain track. Four patients had signs of wound hematoma and serous discharge. No patient developed signs of deep infection during the first 12 months after surgery. This study confirms that coagulase-negative staphylococci are the most common single species obtained from cultures after total hip replacement. Routine antimicrobial prophylaxis against coagulase-negative staphylococci might be considered.

Bacterial vaginosis (BV) has been considered by many investigators to be a risk factor for preterm labor. We have followed vaginal pH and the persistence of clue cells in Papanicolaou stained smears in 119 pregnant women during the course of pregnancy. Of 19 patients with clue cells in their smears during the first trimester, 11 (58%) still had clue cells at the second visit during the third trimester. Of the 100 patients without clue cells during their first trimester, none exhibited clue cells during the third trimester. If the persistence of clue cells is truly a risk factor for adverse pregnancy outcome, screening in the first trimester would identify a risk group of 15%. This risk group diminishes to 9% at the time of the third trimester. Vaginal pH >4.5 had a recovery sensitivity of 76% and specificity of 83%. If clue cells can be considered as the identifying standard for bacterial vaginosis, the sensitivity and specificity of pH is 89% and 94%, respectively. The establishment of the diagnosis of BV during pregnancy and, in some cases, the treatment of the condition may be important as routine procedures in the antenatal center.

Bacterial ascending from the cervix may infect the membranes and amniotic fluid, contributing to their premature rupture. In the present in vitro study, isolated amniotic membranes were exposed to different concentrations of disintegrated bacterial cells. Collagenolysis was estimated by the ratio of acid-soluble/insoluble hydroxyproline. The amniotic membrane proteins were separated by electrophoresis after incubation with 14C proline and disintegrated bacterial cells. The proportion of soluble hydroxyproline increased with the bacterial concentration used, and the radioactivity decreased in proteins of the 30- to 80-kD range. The data support the notion that bacteria contain collagenolytic enzymes affecting human amniotic membranes.

OBJECTIVE:To find out if there were any differences in infection rates if acute traumatic soft tissue wounds were cleaned with tap water instead of sterile saline.DESIGN:Randomised study.SETTING:Emergency department at one city hospital.SUBJECTS:705 consecutive patient with soft tissue wounds less than six hours old that did not penetrate a viscus, cavity, or joint and could be treated by primary suture.INTERVENTIONS:Randomly allocated to have the wound cleaned with either sterile saline or tap water in addition to debridement.MAIN OUTCOME MEASURE:Rate of wound infection, the presence of which was indicated by pus in the wound and prolonged healing.RESULTS:The infection rate in wounds cleaned with sterile saline was 10.3% compared with 5.4% in wounds cleaned with tap water (p less than 0.05). Infected wounds were significantly larger than uninfected ones (p less than 0.05) and more likely to be located on a lower extremity (p less than 0.05). There were no microbiological differences between the two groups, and no bacterial species grown from tap water was subsequently grown from an infected wound.CONCLUSION:Sterile saline should be replaced by tap water for the cleaning of acute traumatic superficial soft tissue wounds.

Microorganisms associated with bacterial vaginosis are commonly recovered from the amniotic fluid and chorion-amnion of patients who deliver prematurely. Bacteria closely related to those causing bacterial vaginosis may play a role in the initiation of uterine contractions, ripening of the cervix and weakening of the fetal membranes by stimulating prostaglandin synthesis. In the present investigation, cervical mucus was collected by brush from early pregnant women with and without bacterial vaginosis. The concentrations of PGE2, PGF2 alpha and 6-keto-PGF1 alpha were determined in the mucus samples by methyl oximation and then radioimmunoassay, utilizing antibodies raised against oximated prostaglandins. It was found that the concentration of PGE2 and PGF2 alpha was significantly higher in the mucus of women with bacterial vaginosis compared with healthy women. The concentration of 6-keto-PGF1 alpha was similar in both study groups. All patients had been instructed to abstain from sexual intercourse for 24 hours before sampling. However, it may be that women with high concentrations in their mucus may have had intercourse anyway. However, it is fairly well possible that the significant differences in the PGE2 and PGF2 alpha values are causally related to the higher rate of preterm labor in women with the commonplace infection of bacterial vaginosis.

The concentration of secretory immunoglobulin A (IgA) in the vaginal fluid of 20 untreated postmenopausal women (aged 75.4 ± 1.4 years) was compared with the levels in a group of 20 postmenopausal women (aged 73.2 ± 1.5 years) treated with oral oestriol (E3) (2 mg/day) and a further group of 20 healthy, non-pregnant, fertile women (aged 28.2 ± 1.8 years). Secretory IgA was determined using a paper disc modification of the single radial immunodiffusion technique. The vaginal concentration of IgA in the untreated women was 41.5 ± 5.7 mg/l, which was higher than that recorded in those treated with E3 (20.8 ± 5.7 mg/l, P < 0.05) and in the fertile women (16.4 ± 3.7 mg/I, P < 0.01). Epithelial cells predominated in the wet smear preparations obtained from the women who received E3 and the fertile women, while a predominance of leucocytes was observed in the untreated women. Lactobacilli were found more frequently in the vaginal flora of the women in the E3 group (P < 0.001) and the fertile women (P < 0.001) than the untreated group. Faecal-type bacteria were seen more frequently in the vaginal flora of the untreated women than the E3 group (P < 0.01) or the fertile women (P < 0.001). The possible implications of the present findings with regard to the individual variation in the intensity of symptoms experienced by women suffering from the oestrogen deficiency syndrome are discussed.

10 pregnant women with bacterial vaginosis were treated vaginally with a low-pH lactate gel intermittently during 6 weeks. Reappearance of a lactobacillus-dominated vaginal flora was observed after a few days treatment in all patients. All women experienced a subjective disappearance of abnormal vaginal discharge and malodour. This local treatment restores the normal vaginal acidity and facilitates recolonization with lactobacilli. It may be preferable to oral antimicrobial therapy, especially during pregnancy.

10 pregnant women with bacterial vaginosis were treated vaginally with a low-pH lactate gel intermittently during 6 weeks. Reappearance of a lactobacillus-dominated vaginal flora was observed after a few days treatment in all patients. All women experienced a subjective disappearance of abnormal vaginal discharge and malodour. This local treatment restores the normal vaginal acidity and facilitates recolonization with lactobacilli. It may be preferable to oral antimicrobial therapy, especially during pregnancy.

Intermittent treatment with an acid lactate gel (Lactal, ACO, Sweden) reduced symptoms of bacterial vaginosis (BV) and promoted the reestablishment of the normal vaginal flora of lactobacilli. Forty-two women seriously affected by recurrent BV were initially given acid gel (lactate gel, pH 3.8, 5 ml) to be inserted into the vagina daily for 7 consecutive days. Thereafter they entered into a double blind clinical trial and were treated prophylactically 3 days monthly for 6 months with either lactate gel or a placebo gel. Women treated with the lactate gel were clinically improved, i.e. no signs of BV in 88% compared to 10% in the placebo group (p less than 0.001). The vaginal lactobacilli flora was reestablished in 83% of the lactate group and in 16% of the placebo group. Local intermittent application of lactate gel was found to be free of side effects and is a preferable alternative to repeated treatments with antibiotics in patients with recurrent BV.

Resistance of the fecal bacterial flora to doxycycline was studied before, during and after systemic prophylaxis with that agent in 24 patients undergoing surgery for complex intestinal disorders, 14 of whom had received antibiotics during the preceding 6 months. Fecal specimens for bacteriologic culture were obtained preoperatively, at operation and 3 days, 1 week and 4 weeks postoperatively. In nine patients (38%) there was dominance of doxycycline-resistant strains preoperatively. At 3 days and 1 week postoperatively the corresponding figures were around 55%. Dominance of doxycycline-resistant bacteria was found in altogether 20 patients at some time during the 4 postoperative weeks. At 4 weeks the fecal flora was normal in 22 of the 24 patients. Antibiotics other than doxycycline should be preferred for prophylaxis in patients undergoing complex, revisional or repeat bowel surgery, especially if they have received antibiotic treatment in the preceding 3-6 months.

In the circle absorber system, a decrease in fresh gas flow means a higher degree of rebreathing, and, consequently, a higher temperature and humidity within the system. With our present hygienic routines, the circle system tubings are changed and decontaminated once daily. Thus, the same circle system is used for several patients each day. In order to evaluate whether the risk of bacterial contamination increased with the introduction of low–flow anaesthesia, 122 patients anaesthetized with either a low–flow technique (< 1.5 1 fresh gas flow/min) or with medium fresh gas flows (3–6 1/min) were studied. Bacterial samples were taken preoperatively from the oropharynx and postoperatively from five locations in the circle system. The patients were studied postoperatively for signs of respiratory tract infection. There were few positive bacteria cultures from the tubings in the circle system, regardless of fresh gas flow. No pathogens were found in the inspiratory tubings and no cases of postoperative respiratory tract infection could be related to bacterial spread from the anaesthesia machine. There were no indications that the present hygienic management was insufficient for the medium– or the low–flow circle system techniques.

SummaryForty‐one female geriatric in‐patients with recurrent urogenital infections took part in a cross‐over intervention study with oral estriol as an alternative to antibiotics.After one month of treatment with a dosage of 3 mg/day the vaginal flora of all patients showed a dominance of Lactobacilli, whereas without estriol treatment the flora was of a fecal type in two‐third of the cases. In all women the atrophic vaginal mucosa became restored and well vascularised.Antibiotics were given 16 times more often during the period without estriol treatment.It is concluded that treatment with estriol• restores the normal premenopausal vaginal flora and mucosa• prevents urogenital infections and disorders in postmenopausal women.