Invasive group A streptococcal infections in Ontario, Canada. Ontario Group A Streptococcal Study Group

BACKGROUND

Several reports suggest that the incidence of invasive group A streptococcal infections, including streptococcal toxic shock syndrome and necrotizing fasciitis, is increasing.

METHODS

During 1992 and 1993 we conducted prospective, population-based surveillance of invasive group A streptococcal disease in Ontario, Canada. We reviewed clinical and laboratory records, searched for secondary cases of invasive disease, and cultured specimens from household contacts.

RESULTS

We identified 323 patients with invasive group A streptococcal infections, for an annual incidence of 1.5 cases per 100,000 population. The rates were highest in young children and the elderly. Fifty-six percent of the patients had underlying chronic illness. Risk factors for disease included infection with the human immunodeficiency virus, cancer, diabetes, alcohol abuse, and chickenpox. The most common clinical presentations were soft-tissue infection (48 percent), bacteremia with no septic focus (14 percent), and pneumonia (11 percent). Necrotizing fasciitis occurred in 6 percent of patients, and toxic shock in 13 percent. The mortality rate was 15 percent overall, but it was 29 percent among those over 64 years of age (P<0.001) and 81 percent among those with toxic shock (P<0.001). Fourteen percent of the cases were nosocomial, and 4 percent occurred in nursing home residents, often in association with disease outbreaks. Invasive disease occurred in 2 household contacts of patients with infection, for an estimated risk of 3.2 per 1000 household contacts (95 percent confidence interval, 0.39 to 12 per 1000).

CONCLUSIONS

The elderly and those with underlying medical conditions are at greatest risk for invasive group A streptococcal disease, toxic shock, and necrotizing fasciitis. Invasive steptococcal infection is associated with a substantial risk of transmission in households and health care institutions.

Evaluation of nutritive capacities of 10 broth media

Ten broth media were evaluated for their ability to support the growth of low inocula of selected fastidious aerobic, microaerophilic, and anaerobic bacteria. Thioglycolate medium USP (Oxoid, Basingstoke, England), cooked meat broth (Oxoid), and fastidious anaerobe broth (Quelab Laboratories Inc., Montreal, Quebec, Canada) were best able to support the growth of the greatest variety of organisms at the lowest inoculum tested (10(1) CFU). In most cases, incubation of these broths resulted in visible turbidity within 3 days. These three broth media were found to be equally suitable for use as an enrichment broth in the clinical laboratory.

An outbreak of fatal nosocomial infections due to group A streptococcus on a medical ward

Group A streptococcus is an uncommon but important cause of nosocomial infections. Outbreaks of infection most often have occurred in surgical or obstetrical patients. We describe an outbreak of severe group A streptococcal infections that occurred on a medical unit of a community hospital. Within an 8-day period, three patients developed fatal nosocomial skin and soft-tissue infection due to group A streptococcus. Three nurses who had provided care to one or more of these patients subsequently developed streptococcal pharyngitis, and three other nurses were treated with antibiotics for pharyngitis (cultures not obtained). Patient isolates were serotype M-nontypeable, T-11, opacity factor-positive, and shared identical DNA profiles when typed by pulsed-field gel electrophoresis; staff isolates were not available for typing. To prevent further spread of infection, the ward was closed to new admissions, and symptomatic staff were treated with antibiotics and relieved of patient-care duties. This outbreak demonstrates the ability of group A streptococcus to spread rapidly in a hospital setting and to cause severe life threatening disease in hospitalized patients.

Evaluation of enzyme immunoassay for detection of salivary antibody to Helicobacter pylori

The Helisal test is a quantitative enzyme immunoassay for the measurement of Helicobacter pylori-specific immunoglobulin G antibodies in saliva. This test was evaluated in comparison with culture and histopathologic examination of gastric biopsy specimens obtained from 195 patients who underwent 200 endoscopic procedures for the investigation of gastrointestinal symptoms. Forty-one (21%) patients were found to have peptic ulcer disease, and one other patient had a gastric carcinoma. H. pylori was detected in gastric biopsy specimens obtained from 98 (49%) of the procedures. The sensitivity, specificity, and positive and negative predictive values of the Helisal test were 81, 75, 76, and 80%, respectively. The test was negative for 16 (38%) of the 42 patients with peptic ulcer disease or a gastric malignancy diagnosed at endoscopy. These results suggest that the Helisal assay is only moderately accurate for the detection of H. pylori infection in symptomatic patients.

Tuberculosis infection among staff at a Canadian community hospital

OBJECTIVE

To determine risks for tuberculin skin-test conversion among employees of a community hospital in Ontario, Canada.

DESIGN

Cohort morbidity study.

SETTING

Of 14 metropolitan Toronto area hospitals surveyed for data on tuberculin skin-test conversions, only one provided tuberculosis (TB) test data on all employees. Between 1991 and June 1994, 24 patients were treated at this hospital for pulmonary TB.

POPULATION STUDIED

The population at risk included those on staff from January 1991 through December 1993 who previously were skin-test negative; they were followed until the end of June 1994. Exposure was estimated (a) based on ranking departments according to an estimate of the number of hours of direct patient contact during a typical day, and (b) based on location of sputum-positive patients.

OUTCOME MEASURE

Risks of skin-test conversion among hospital employees with documented prior negative skin tests.

MAIN RESULTS

A total of 809 skin-test negative employees were followed for 2,084 person-years; 18 employees with skin-test conversions were identified. The overall conversion rate was 0.9% per year (0.86 per 100 person-years). After excluding two conversions attributed to contact with coworkers, the relative risk of conversion was 4.5 (5.5 after adjusting for age and gender) among those in the highest exposure category (> or = 4 hours per day), compared to those in departments ranked as having the lowest exposure (< 2 hours per day). Among those working in wards in which sputum-positive patients were treated, 2.4% converted; the risk of conversion was over six times greater than among those working on wards with no TB patients or in departments with no patient contact, of whom 0.4% converted. Among the emergency room staff, the department in which the greatest number of sputum-positive patients were treated, at least 5% of staff converted. In those instances in which conversions were associated with exposure to a specific TB patient, the involved patients had been in the hospital for at least 4 days prior to being isolated.

CONCLUSIONS

These results indicate that even in a hospital with few admissions due to tuberculosis, skin-test conversions associated with occupational exposure may occur (Infect Control Hosp Epidemiol 1996; 17:29-35).

Comparison of E test and agar dilution for antimicrobial susceptibility testing of Stenotrophomonas (Xanthomonas) maltophilia

Currently recommended dilution test methods for the determination of antimicrobial susceptibility of Stenotrophomonas (Xanthomonas) maltophilia are labor-intensive and often impractical in many clinical laboratories. We compared the E test with the agar dilution method for susceptibility testing of 176 clinical isolates of S. maltophilia against 16 antimicrobial agents. The MICs obtained by E test correlated well with those determined by the agar dilution method, with an overall agreement of 94%. Very major and major errors occurred infrequently (0.6 to 2.9%) when testing beta-lactam agents, tobramycin, trimethoprim-sulfamethoxazole, and fluoroquinolones. The E test was found to be accurate and easy to perform. For most antimicrobial agents tested against S. maltophilia, the E test is an acceptable alternative susceptibility test method.

Influenza and pneumococcal vaccination and tuberculin skin testing programs in long-term care facilities: where do we stand?

OBJECTIVE

1) To compare policies and procedures for distribution of influenza and pneumococcal vaccines to long-term care facilities for the elderly in Canada, 2) to determine vaccination rates of residents and staff, and 3) to describe vaccination and tuberculin skin testing programs in these facilities.

DESIGN

A cross-sectional survey consisting of telephone interviews and a mailed questionnaire was conducted in the spring of 1991. Telephone interviews were conducted with provincial/territorial epidemiologists. The questionnaire was sent to all (N = 1.520) Canadian long-term care facilities for the elderly with > or = 25 beds.

RESULTS

There were 1,270 responding facilities (84%). The mean overall influenza vaccination rate for residents was 78.5%. The mean vaccination rate was higher in those provinces in which the vaccine was paid for by the government (79% versus 71%; P = 0.002). Only 19% of facilities reported staff vaccination rates > 25%; rates again were higher in those provinces in which vaccine for staff was provided by the government. Pneumococcal vaccine was offered to residents in 12% of the facilities. The proportions of facilities with > 10% and > 75% of residents vaccinated were significantly higher in the provinces where the pneumococcal vaccine was recommended and paid for as compared with those where it was not (P < 0.001 for both). Tuberculin skin testing programs for residents existed in 360 long-term care facilities (28%) across the country.

CONCLUSION

In 1990, the number of residents living in Canadian long-term care facilities who were vaccinated against influenza and Streptococcus pneumoniae was suboptimal. Staff influenza vaccination rates were very low across the country. Most facilities did not have a baseline tuberculin skin test status for their residents. Vaccination rates are higher in jurisdictions in which governments provide the vaccine without charge.

Use of the polymerase chain reaction for the detection of Chlamydia trachomatis from endocervical and urine specimens in an asymptomatic low-prevalence population of women

The Amplicor Chlamydia trachomatis test is a polymerase chain reaction (PCR)-based methodology used for the detection of a cryptic plasmid found in C. trachomatis. It was evaluated in comparison with cell culture and the Microtrak II Chlamydia enzyme immunoassay (EIA) for the detection of C. trachomatis in urogenital specimens from women. Endocervical swabs were collected from 993 women attending the women's unit at the Mount Sinai Hospital in Toronto. In addition, concomitant first void urine specimens were collected from 394 of these women for PCR testing only. As compared with culture of the endocervical specimens, PCR and EIA had a sensitivity, specificity, positive predictive value and negative predictive value of 84.6%, 99.2%, 57.9%, and 99.8% and 61.5%, 99.7%, 72.7%, and 99.5%, respectively. As compared with the secondary gold standard of a positive culture and/or a positive PCR using a primer to the major outer membrane protein the sensitivity, specificity, positive, and negative predictive values for culture were 72.2%, 100%, 100%, and 99.5%, respectively. For the Amplicor PCR and EIA the results were 88.9%, 99.7%, 84.2%, and 99.9% and 61.1%, 99.9%, 91.7%, and 99.6%, respectively. When the urine PCR was compared with the same standard, the test had a sensitivity of 91.7% and a specificity of 99.5%. Based on this study the Amplicor C. trachomatis test was found to be sensitive and specific for the detection of C. trachomatis in both endocervical and urine specimens.

Successful treatment of systemic and local infections due to Exophiala species

We report the successful treatment of three cases of infection due to Exophiala species. These organisms belong to the heterogeneous group of dematiaceous (darkly pigmented) fungi. Two cases of infection occurred in organ transplant recipients who were receiving immunosuppressive medications. Both of these infections remained localized to the subcutaneous tissues and were successfully treated with surgical excision of the lesions. We also describe what is to our knowledge the first reported case of prosthetic valve endocarditis due to Exophiala castellanii that was managed with a combination of medical and surgical therapies. Exophiala species remain an uncommon cause of infection. However, as the population of immunocompromised patients continues to grow and further improvements in the microbiological techniques for identification of these fungi occur, these organisms will be recognized with increasing frequency as a cause of human disease.

Evaluation of three methods for the rapid identification of Staphylococcus aureus in blood cultures

A total of 445 blood cultures containing Gram-positive cocci in clusters were tested for the presence of Staphylococcus aureus with the Accuprobe, heat-stable thermonuclease, and latex agglutination using Staphaurex. The results show that the Accuprobe, thermonuclease, and Staphaurex correctly identified 95, 96, and 62 of the 100 specimens containing S. aureus. The corresponding specificity for the methods was 99.1%, 100%, and 98.5%, respectively.

Identification and characterization of multiple species of vancomycin-resistant enterococci, including an evaluation of Vitek software version 7.1

A total of 374 clinical isolates of Enterococcus spp. were characterized to determine the species distribution and vancomycin resistance. The ability of the Vitek system (bioMerieux Inc, Hazelwood, St. Louis, Mo.) to identify enterococci to the species level and to recognize vancomycin resistance by using computer software version 7.1 was evaluated. Conventional methods were used for identification and agar dilution was used for susceptibility testing, the results of which were as follows (presented as number of vancomycin-resistant isolates/number of members of that species identified): 219/234 E. faecium, 9/112 E. faecalis, 2/3 E. mundtii, 0/1 E. durans, 0/1 E. hirae, 0/1 E. raffinosis, and 0/1 E. avium. Ten enterococci were in the vancomycin-intermediate category (six E. gallinarum, two E. casseliflavus, and one each of E. faecium and E. faecalis). The Vitek GPI card correctly identified 98% of E. faecium isolates, 99% of E. faecalis isolates, and only two isolates of other enterococcal species. The GPS-TA card was 98% sensitive and 95% specific for the detection of vancomycin resistance, generating a total of four very major (1.6%) and five minor errors (1.3%).

Infection due to Clostridium difficile among elderly residents of a long-term-care facility

In a study of the epidemiology of infection due to Clostridium difficile at long-term-care facilities, we conducted point-prevalence surveys and obtained stool samples from residents receiving antibiotics and from those developing diarrhea during 1 year at a 350-bed nursing home and an adjoining 280-bed chronic-care hospital. C. difficile and/or its cytotoxin was detected in 236 specimens from 94 residents. Only 16 (17%) of these 94 individuals had diarrhea at the time C. difficile was detected. The prevalence of C. difficile infection ranged from 2.1% to 8.1% in the nursing home and from 7.1% to 14.7% in the hospital. The organism was recovered from six (8.8%) of 68 residents receiving antibiotics, and four of the six developed antibiotic-associated diarrhea. The receipt of antibiotic treatment within the previous 8 weeks (odds ratio [OR], 7.9), the presence of a nasogastric or gastrostomy feeding tube (OR, 6.5), urinary and fecal incontinence (OR, 2.5), and the presence of more than three underlying diseases (OR, 2.0) were statistically significant independent variables associated with C. difficile infection. Typing of isolates by restriction-endonuclease analysis indicated that most C. difficile infections at this long-term-care facility were associated with endogenous enteric carriage of the organism, with little evidence of cross-infection.

Utilization review of the use of BACTEC PLUS high-volume blood culture bottles

The BACTEC PLUS 26 (NR26) (Becton Dickinson, Towson, Md.) high-volume blood culture bottle replaced the less expensive smaller-volume NR6A bottle in our hospital. An audit carried out several months after their introduction revealed that only 17.5% of the NR26 bottles received the required blood volume. Several audits and educational programs were required in order to achieve a compliance rate of > 60%.

Prosol-chlorhexidine irrigation reduces the incidence of bacteremia during ultrasonic scaling with the Cavi-Med: a pilot investigation

The purpose of this pilot investigation was to determine whether the incidence of bacteremia following subgingival ultrasonic scaling and root planing could be reduced by the use, pre- and intraoperatively, of an irrigant containing 0.12 per cent chlorhexidine (CHX); Prosol. Individuals having evidence of significant periodontal disease (minimum of seven sites per quadrant 4.0 mm and bleeding on probing) were entered into this study. By use of a random number table, patients were assigned to either the experimental or control groups. The procedures, as described below, were carried out in a double blind fashion so that neither the investigator nor the patient was aware of whether Prosol or placebo was being used. The placebo solution was flavored to make it indistinguishable from Prosol. Patients were first anesthetized. Their gingival crevices were then irrigated using the Cavi-Med ultrasonic scaler. At this point, the ultrasonic action was not activated. Ten minutes later, ultrasonic scaling and root planing with the Cavi-Med unit were begun with a continuous flow of either the placebo or control solutions. Blood samples were taken preoperatively, while postoperative samples were taken one minute after completing the scaling of each quadrant and then 10 minutes after scaling the second quadrant. Routine aerobic and anaerobic bacterial culture methods were used to identify viable blood-borne bacteria. The results show that there was no difference in the distribution or presentation of periodontal disease between the experimental and control quadrants.(ABSTRACT TRUNCATED AT 250 WORDS)

Rapid detection of infectious mononucleosis-associated heterophile antibodies by a novel immunochromatographic assay and a latex agglutination test

A novel immunochromatographic assay, the CARDS O.S. MONO test (Pacific Biotech, San Diego, Calif.), and a latex agglutination test, the Infectious Mononucleosis Kit (Unipath Ltd., Hampshire, United Kingdom) were compared with the Paul-Bunnell-Davidsohn test. Of the 957 serum specimens studied, 78 were positive and 879 were negative by the Paul-Bunnell-Davidsohn test. After discrepancies were resolved by determining Epstein-Barr virus serology, the sensitivities of the CARDS O.S. MONO test and the Infectious Mononucleosis Kit were 91.0 and 96.2%, respectively, and both tests had a specificity and a positive predictive value of 100% and a negative predictive value and overall agreement of greater than 99%. The results show that both tests can accurately detect infectious mononucleosis-associated heterophile antibodies.

Severe invasive group A streptococcal infections in Ontario, Canada: 1987-1991

During the past few years, there has been an apparent increase in serious infections due to group A streptococci (GAS) worldwide. We describe our experience with severe invasive GAS infections in Ontario, Canada, during the past 5 years (February 1987 through December 1991). A case was defined as the isolation of GAS from blood or normally sterile tissue in association with hypotension (systolic blood pressure, < 90 mm Hg). Fifty cases were identified in patients ranging in age from 4 to 100 years (median age, 47 years); 29 (58%) of the patients died. A primary focus of infection was identified in 38 cases (76%), with soft tissue being the site involved most frequently (68%). No focus of infection was found in 12 patients, and 36 patients (72%) were bacteremic. Complications included acute respiratory distress syndrome (21 of 50), acute renal failure (20 of 50), hypocalcemia (19 of 24), elevated creatinine kinase values (21 of 27), coagulation abnormalities (15 of 21), and hepatitis (15 of 24). Eleven cases (22%) were nosocomial; one of these was secondary to another nosocomial case. Thirty-three isolates were available for M and T typing and for determination of the presence of the genes for streptococcal pyrogenic exotoxin (SPE). The most frequent types were M1T1 (10) and M12/T12 (8). Twelve isolates possessed the speA gene, and 16 isolates had the speC gene. Only three isolates possessed both speA and speC. All isolates possessed the speB gene.

A cluster of surgical wound infections due to unrelated strains of group A streptococci

Group A streptococci account for less than 1% of all surgical wound infections but are an important cause of nosocomial outbreaks. We report here a cluster of four group A streptococcal infections that occurred within an 11-day period on a single surgical service. The index case presented with toxic shock-like syndrome. Epidemiologic investigation did not identify any relationship between infections. Restriction endonuclease analysis and M and T typing found the four isolates to be unrelated. Restriction endonuclease analysis is a useful tool for determining relatedness of nosocomial isolates of group A streptococci.

A study of the epidemiology of an endemic strain of staphylococcus haemolyticus (TOR-35) in a neonatal intensive care unit

Coagulase-negative staphylococci (CNS) are among the most prevalent microorganisms that colonize and cause sepsis in neonatal intensive care units (NICU). We had previously identified a strain of CNS, Staphylococcus haemolyticus (TOR-35), in the NICU at Mount Sinai Hospital, that had been repeatedly isolated from blood cultures from neonates. We therefore carried out a prospective study to determine the frequency and time of colonization and the frequency of bacteremia in neonates over a 3.5 month period. This was accomplished by obtaining surface swabs within 1 h of birth and on days 3, 5, and 7 and by characterizing all blood culture isolates of CNS. We also determined what percentage of neonatal CNS bacteremias were due to this strain, between January 1, 1987 and December 31, 1990, by retrieving and typing all stock cultures of CNS from that period. All isolates were typed by species identification and antimicrobial susceptibility profile code. There were 76 (38%) neonates that became colonized with the TOR-35 strain at some time during their NICU stay. Lower birth weight was associated with colonization (p < 0.001), as was lower gestational age (p < 0.001). Only 1 neonate had a positive blood culture isolate for the TOR-35 strain during the prospective study. Of the 4 years of neonatal bacteremias that were studied retrospectively, there were 252 episodes of CNS bacteremia, of which 27 (11%) were due to the TOR-35 strain. The TOR-35 strain has become endemic in our NICU and appears to selectively colonize premature, low birth weight newborn infants, but only infrequently causes bacteremia.

Broth microdilution testing of Haemophilus influenzae with haemophilus test medium versus lysed horse blood broth. Canadian Haemophilus Study Group

Broth microdilution testing of 702 community-acquired isolates of Haemophilus influenzae from across Canada was performed with both Mueller-Hinton broth supplemented with 3% lysed horse blood broth (LHB) (BBL Microbiology Systems, Cockeysville, Md.) and haemophilus test medium (HTM). The prevalence of beta-lactamase production was found to be 26% with no regional variation. MICs determined with LHB tended to be higher than those with HTM, but interpretive errors due to these differences were observed only rarely with trimethoprim-sulfamethoxazole (n = 5), cefaclor (n = 8), and cefamandole (n = 3). The interobserver variability in MIC determinations was found to be greater when LHB was used than when HTM was used. There was no difference in intraobserver variability between the two medium formulations. beta-Lactamase-positive isolates developed false resistance to amoxicillin-clavulanate 2 weeks after microdilution panels of both types of medium were stored at -20 degrees C but not when panels were stored at -70 degrees C. In conclusion, this study supports the use of HTM rather than LHB for sensitivity testing of H. influenzae because of its lower rate of interobserver variability and its ability to support the growth of these organisms, which is comparable to that of LHB.

Evaluation of commercial and standard methodology for determination of oxacillin susceptibility in Staphylococcus aureus

Agar dilution with and without 4% NaCl, broth microdilution with 2% NaCl, the dried MicroScan Rapid Positive MIC 1 panel (Baxter Health Care Corp., West Sacramento, Calif.), the Vitek GPS-SA card (Vitek Systems, Hazelwood, Mo.), and the oxacillin agar screen plate were compared with a DNA probe encoding the mec gene for their abilities to detect oxacillin resistance in 506 clinical isolates of Staphylococcus aureus. The results of testing for the mec gene showed that there were 254 oxacillin-resistant and 252 oxacillin-susceptible isolates of S. aureus. There were 14.2% very major errors with Vitek (a resistant isolate was interpreted as susceptible) and 6.7% very major errors with MicroScan. Fewer major errors were seen: 0.8% with MicroScan (a susceptible isolate was interpreted as resistant) and 0.4% with Vitek. No very major errors but 2.4% major errors occurred by agar dilution with 4% NaCl supplementation, whereas there were 0.8% very major and 0.4% major errors without 4% NaCl supplementation. By broth microdilution there were 2.0% very major and 0.8% major errors. The results of the oxacillin agar screen plate method were 100% concordant with those of the mec gene probe method.

Detection of vancomycin resistance in Enterococcus species

Enterococcus faecalis and Enterococcus faecium isolates that are resistant to vancomycin have recently been identified in North America and Europe. Of 155 clinical isolates of enterococci (113 E. faecium and 42 E. faecalis), we found that 98 were resistant, 52 were moderately susceptible, and 5 had intermediate susceptibilities to vancomycin by using broth microdilution susceptibility testing according to the National Committee for Clinical Laboratory Standards (NCCLS) (Approved Standard M7-A2). Using NCCLS disk diffusion methodology (Approved Standard M2-A4), we evaluated the NCCLS supplemental M100-S3 revisions for zone diameter interpretive standards and incubation conditions and found 5.8% minor errors. A total of 234 isolates, which included an additional 79 E. faecium isolates that were moderately susceptible to vancomycin, were used to evaluate the Vitek GPS-TA card (bioMerieux, Inc., Hazelwood, Mo.) and the Pos MIC type 6 panel (MicroScan; Baxter Health Care Corp., West Sacramento, Calif.) for the detection of vancomycin resistance. The Vitek card was 100% specific and 72% sensitive, whereas the MicroScan panel with the Walk/Away system was 98% specific, with a sensitivity of 93% which increased to 99% when readings were performed manually. An agar screen plate method was evaluated with vancomycin concentrations of 6, 8, 10, or 12 micrograms/ml; plates were inoculated so as to obtain a final concentration of 10(5) CFU per spot. This method was found to be 100% sensitive and specific at all concentrations.