Penicillin tolerance and treatment failure in group A streptococcal pharyngotonsillitis.

Failure of treatment of group A streptococcal pharyngitis and tonsillitis is well documented. One of the possible explanations for treatment failure is penicillin tolerance in group A streptococci. Reports on the prevalence of penicillin tolerance among group A streptococci (0-100%) and the presumed relationship with therapeutic failure vary considerably. Therefore, it appears worthwhile to review pharyngotonsillitis studies, devoting special attention to the variables of MIC-MBC laboratory determinations such as inoculum preparation, composition and volume of test medium, and the criteria used to define penicillin tolerance. Alternative methods (gradient-replica plate method, beta-lactamase disk test, time-kill assay, and cell-lysis assay) are discussed. It is concluded that technical factors and the definitions used influenced the reported rates of penicillin tolerance. The epidemiological data suggest that tolerance is not limited to a single streptococcal serotype. Furthermore, there is not sufficient data to support a correlation between in vitro penicillin tolerance of group A streptococci and treatment failure, either in clinical cases or in animal studies. On the other hand, evidence to exclude penicillin tolerance as a cause of treatment failure is also not available. Therefore, at present penicillin tolerance cannot be ruled out as a cause of penicillin treatment failures.

Susceptibility of Streptococcus pyogenes to azithromycin, clarithromycin, erythromycin and roxithromycin in vitro.

The susceptibility of 180 clinical isolates of Streptococcus pyrogenes from six regions of The Netherlands to the macrolide antibiotics azithromycin, clarithromycin, erythromycin and roxithromycin was analysed. The results of a microbroth MIC method, the E-test method and a disk diffusion assay were compared, and the MBC determined. In addition, the susceptibility to erythromycin of 436 clinical isolates of S. pyogenes from the Leiden region was determined. The microbroth MIC90s of azithromycin, clarithromycin, erythromycin and roxithromycin for group A streptococci were < or = 0.5 mg/L. Erythromycin had the lowest MIC90 (0.09 mg/L). The MIC data obtained with the E-test method suggested that clarithromycin and erythromycin had slightly higher anti-streptococcal activity than azithromycin and roxithromycin in vitro. MICs obtained with the E-test were lower than those found with the microbroth method. Only minor discrepancies were observed among the three methods. The MBC50 for both clarithromycin and erythromycin was 0.75 mg/L and 5.0 mg/L for azithromycin and roxithromycin. None of the 180 strains and two of the collection of 436 strains (0.5%) were resistant to erythromycin and the other macrolides tested; MICs ranged from 1 to 16 mg/L. The erythromycin-resistant strains showed an inducible type of macrolide-lincosamide-streptogramin B (MLS) resistance.

A comparison of pefloxacin/metronidazole and doxycycline/metronidazole in the treatment of laparoscopically confirmed acute pelvic inflammatory disease.

A double-blind, randomised study was conducted to compare the efficacy and safety of a combination of pefloxacin and metronidazole versus doxycycline and metronidazole in patients with pelvic inflammatory disease (PID). The clinical diagnosis had to be confirmed by laparoscopy before patients were included. Of the 74 patients who fulfilled the clinical criteria for PID, laparoscopy confirmed the diagnosis in only 40 patients (54%). The microorganism most frequently found as causative pathogen was Chlamydia trachomatis. Both treatment groups showed a good response to the study-medication. At discharge 9 patients in the pefloxacin group (45%) were cured and 10 patients (50%) had improved. In the doxycycline group 7 patients (35%) were cured and 10 patients (50%) had improved. Obviously pefloxacin/metronidazole and doxycycline/metronidazole are equally effective in the treatment of PID.

Detection of penicillin tolerance in Streptococcus pyogenes.

Three traditional assays were used to determine the minimal bactericidal concentration (MBC) and minimal inhibitory concentration (MIC) for Streptococcus pyogenes (group A streptococci) in two phases of growth and the time taken to kill the organisms. Three other methods were used for the determination of penicillin tolerance: a cell-lysis assay, the beta-lactamase disk method and the replication method. Twenty strains, comprising penicillin-tolerant clinical isolates and two laboratory mutants, were used to evaluate the six tests. Results indicated that two groups of S. pyogenes can be distinguished--four highly tolerant and three moderately tolerant strains. The moderately tolerant strains were not recognised when rapidly growing instead of stationary cultures were used for the MBC and MIC determinations. The MBC/MIC ratio for tolerant strains was > 100. Tolerance percentage ranged from 0.30 to 1.07 and 0.29 to 3.96 for cultures in the mid-logarithmic and stationary phases of growth, respectively. The cell-lysis assay, the beta-lactamase disk method and the replication method may be used to screen for tolerance. Detection of high or moderate tolerance by determining the MBC/MIC ratio for mid-logarithmic or stationary cultures is recommended.

In vitro susceptibility of quinolone-resistant Campylobacter jejuni to new macrolide antibiotics.

The MICs of erythromycin and three new macrolide antibiotics were determined for 36 quinolone-susceptible and 106 quinolone-resistant Campylobacter jejuni. The MIC90 values of azithromycin, clarithromycin, roxithromycin and erythromycin were 0.5, 4, 16 and 4 mg/l respectively. No difference was found between macrolide activity against the quinolone-susceptible and the quinolone-resistant strains. Clarithromycin and especially azithromycin might eventually replace erythromycin for the treatment of Campylobacter jejuni infections in view of their pharmacological properties.

High-level aminoglycoside resistance among enterococci and group A streptococci.

The occurrence of high-level aminoglycoside resistance (HLAmR) was determined for 73 enterococci and 54 group A streptococci by the high-load disc method, tube macrodilution and the polymerase chain reaction (PCR). The PCR method revealed the presence of genes coding for aminoglycoside-3'-O-phosphoryltransferase-III (APH(3')-III), aminoglycoside-6'-N-acetyltransferase/2''-O-phosphoryltransferase (AAC(6')/APH(2'')), or both, in 20.6%, 9.6% and 4.1% of the enterococci, respectively. The prevalence of HLAmR to at least one aminoglycoside among local enterococci was 37% (27/73). Only one of 54 Streptococcus pyogenes isolates produced APH(3')-III and exhibited high-level resistance to kanamycin and streptomycin. In general, the three methods yielded comparable results, with only three discrepancies among the 127 isolates examined. High-load disc screening and tube macrodilution proved to be practical, reliable and reproducible, and thus suitable for routine screening. Of 20 Enterococcus faecalis strains tested, all were penicillin-tolerant. Only one of seven penicillin-tolerant S. pyogenes strains was HLAmR. No association between the two forms of resistance was found.

Bacterial contamination and the effect of filters in anaesthetic circuits in a simulated patient model.

In order to investigate bacterial contamination of anaesthetic breathing circuits and means of prevention of this, six different laboratory experiments were performed. These experiments involved the bacterial contamination of Dräger Narkose Spiromat 650 and Dräger AV-1 circle system circuits and of an isolated soda lime carbon dioxide absorber. The effects of anaesthetic gas, gas flow rate and the incorporation of a hydrophobic membrane heat and moisture exchanging bacterial/viral filter (HMEF) at the patient end of these circuits were investigated. It was found that without a HMEF the whole interior of the anaesthetic circuits became contaminated with bacteria. Components closest to the simulated patient showed the highest levels of contamination. Higher gas flows were associated with decreased levels of circuit contamination, presumably because more bacteria were expelled from the system. Halothane (1 volume %) and soda lime were not found to have any demonstrable bactericidal action. The presence of a HMEF between the simulated patient and the Y-piece prevented any detectable contamination from reaching the circuit. Consequently, the presence of a HMEF provides protection of the anaesthetic circuit as well as other patients, healthcare workers and the environment.

[Clinical, bacteriological and cost-saving effects of antibiotic prophylaxis in craniotomy]. / Klinische, bacteriologische en kostenbesparende effecten van antibiotische profylaxe bij craniotomie.

Objective of the study. To investigate the effect of antibiotic prophylaxis on the incidence of infections, on the bacterial flora of wounds and on the health-care costs. A retrospective study disclosed an incidence of infection of 8.1% in patients who underwent craniotomy. Methods. Double-blind, placebo-controlled study of the effects of cloxacillin (4 x 1 gr intravenously during 24 h, perioperatively) in 310 patients who had to undergo a craniotomy. Results. In the cloxacillin group significantly fewer infections occurred than in the placebo group, 6 infections in 156 and 20 infections in 154, respectively. In the cloxacillin group significantly fewer samples contained micro-organisms than in the placebo group. Cloxacillin prophylaxis reduces the cost of patient care by about 20%. Conclusion. Cloxacillin prophylaxis in craniotomy cases reduces the percentage of infections, the percentage of positive cultures of the wound, and the costs of patient care.

Distribution of beta lactamases in Enterobacteriaceae: indoor versus outdoor strains.

The distribution of beta-lactamases of indoor and outdoor origin was studied over a 9-month period. We consecutively selected 37 indoor and 48 outdoor strains of the following genera of the Enterobacteriaceae family: Escherichia (species coli), Enterobacter, Proteus, and Klebsiella. All isolates were resistant to amoxicillin and/or cephalothin and/or cefamandole. All strains showed beta-lactamase activity. We characterized the enzymes by an isoelectric focusing method and by a disc diffusion test. For both indoor and outdoor isolates we found that plasmid-mediated beta-lactamases were encountered mostly in the E. coli and Klebsiella species, whereas chromosomally mediated enzymes predominated in the Proteus and Enterobacter species. No significant difference in distribution of beta-lactamases could be found comparing both groups, but it was noted that chromosomally mediated beta-lactamases predominated in the Department of Urology, while plasmid-mediated beta-lactamases prevailed in other departments (p less than 0.05).

Prevalence of resistance of N. gonorrhoeae to penicillin and three other antibiotics in a rural area in Kenya.

Of 90 isolates of Neisseria gonorrhoeae in a rural area in Kenya, 44 (48.9%) produced beta-lactamase (penicillinase). Testing for susceptibility of 35 penicillinase producing N. gonorrhoeae (PPNG) strains to four antibiotics yielded the following results: 16 (45.7%) showed a decreased susceptibility to tetracycline; six (17.1%) showed resistance, probably plasmid mediated; 10 (28.6%) had intermediate susceptibility to gentamicin; one (2.9%) was resistant; and two (5.7%) isolates were resistant to cefotaxime. 16 (57.1%) of 28 non PPNG strains showed a decreased susceptibility to penicillin; 10 (35.7%) were resistant. Nine (32.1%) of 28 non-PPNG isolates showed intermediate susceptibility to tetracycline; one (3.6%) was resistant. Eight of non PPNG isolates (28.6%) showed decreased susceptibility to gentamicin. These results imply that penicillin and tetracycline should be abandoned as primary therapy. For the time being, thiamphenicol and spectinomycin seem to be good alternatives. The observation of cefotaxime resistance in N. gonorrhoeae in a rural area implies a warning concerning future possibilities for use of third generation cephalosporins.

Hygiene, skin infections and types of water supply in Venda, South Africa.

The impact of the installation of a system to supply chlorinated drinking water in Venda, South Africa, on water quality, water use and health status was evaluated by means of questionnaires, examination for skin infections, and microbiological analysis of water samples. Although the water collection journey became shorter in comparison with use of traditional water supplies such as boreholes and unprotected springs, water use per caput showed no increase. The improved water supply showed no contamination with coliforms even after storage. Borehole water exhibited low coliform counts at the source, but after storage a 10- to 15-fold increase took place. Water samples from unprotected springs exhibited high coliform counts, which declined during storage. The prevalence of infectious skin diseases (27.5%) and diarrhoea (3.7%) among pre-schoolchildren showed no correlation with the quality of drinking water or the use of water per caput. Although the prevalence of infectious skin diseases did exhibit a negative correlation with the frequency of washing, no significant health benefit of the improved water supply could be demonstrated in this limited study.

Comparison of six media, including a semisolid agar, for the isolation of various Campylobacter species from stool specimens.

A recently described semisolid blood-free selective motility medium (SSM) (J. Goossens, L. Vlaes, I. Galand, C. Van den Borre, and J. P. Butzler, J. Clin. Microbiol. 27:1077-1080, 1989) was compared with two charcoal-based selective media (charcoal-based selective medium [CSM] and modified charcoal cefoperazone deoxycholate agar [CCDA]), two blood-based media (Skirrow medium [SKM] and CampyBAP), and a passive, 0.65-microns-pore-size cellulose acetate membrane filter technique for the recovery of campylobacters from stools of patients with diarrhea. A total of 1,980 specimens were tested, 161 of which were found to be positive for campylobacters. Campylobacter jejuni was isolated in 148 specimens (91.9%), C. coli was isolated in 27 (7.5%), and "C. upsaliensis" was isolated in 1 (0.6%). After 72 h of incubation with a single medium, the cumulative percentages of Campylobacter-positive specimens isolated on CSM, CCDA, SKM, and SSM were 87, 83, 80, and 72%, respectively. The filter method alone enabled us to recover 61% of all campylobacters. The "C. upsaliensis" strain was isolated by this method only. The highest isolation rates were observed when two media, including CSM, were combined. The combination of CSM and SSM yielded the highest rates (96%), but these were not statistically different from the rates observed with combinations of CSM and SKM (94%) or of CSM and the filter method (91%). Extending the incubation time from 48 to 72 h led to an increase in the isolation rate regardless of the medium used (P less than 0.001). CSM and CCDA were the most selective media. SKM and CampyBAP appeared to be the most inhibitory media for the isolation of C. coli.

Quinolone resistance in campylobacter isolated from man and poultry following the introduction of fluoroquinolones in veterinary medicine.

Eight hundred and eighty-three strains of Campylobacter spp. isolated between 1982 and 1989 from human stools and poultry products were screened for quinolone resistance. In this period the prevalence of resistant strains isolated from poultry products increased from 0% to 14%. During the same period the prevalence in man increased from 0% to 11%. The emergence of quinolone resistance has implications for the identification of campylobacter up to species level: the susceptibility for nalidixic acid can no longer be used as a criterion for identification in the laboratory. The rapid emergence of resistant campylobacter may also have important implications for the treatment and prophylaxis of diarrhoeal disease. The increase of quinolone resistance coincides with the increasing use of fluoroquinolones in human and veterinary medicine. Extensive use of enrofloxacin in poultry and the almost exclusive transmission route of campylobacter from chicken to man, in The Netherlands, suggests that the resistance observed is mainly due to the use of enrofloxacin in the poultry industry.

Efficacy of cloxacillin prophylaxis in craniotomy: a one year follow-up study.

A previous prospective double-blind placebo-controlled study showed that cloxacillin prophylaxis significantly reduces the rate of infection after craniotomy. The purpose of the present follow-up study was to find out whether the data on routine cloxacillin prophylaxis would confirm the decreased infection rate after craniotomy. During this 1-year study, 8 infections occurred after 201 operations (4.0%) in 175 patients. No prophylaxis was given during 17 operations in 15 patients; 4 infections occurred in this group. The most frequently isolated microorganisms were Staphylococcus aureus, S. epidermidis, and Propionibacterium acnes, all of which were sensitive to cloxacillin. We conclude that cloxacillin is beneficial as prophylaxis in patients undergoing craniotomy.

Variation in induction of chromosomal beta-lactamase expression in strains of Enterobacter cloacae.

Eight strains of Enterobacter cloacae with varying patterns of susceptibility to beta-lactam antibiotics were studied to compare the inducibility and expression of their beta-lactamase genes. These strains included two isolates from one patient, which differed in their susceptibility to cefamandole. A resistant strain constitutively produced large amounts of beta-lactamase; a sensitive strain produced an inducible beta-lactamase. Study of induction and growth characteristics of all strains revealed that the induction and the expression of inducible beta-lactamase genes may vary considerably among strains of E. cloacae.

Correlations between consumption of antibiotics and methicillin resistance in coagulase negative staphylococci.

The correlation between antibiotic consumption, expressed in defined daily doses (DDD), and antibiotic resistance rates was studied, using 976 isolates of coagulase negative staphylococci (CNS) from human pathological material. Data from four hospitals, including 14 participating departments, were analysed for this purpose. Susceptibility tests were performed according to Dutch national standards, except for methicillin, which for the majority of isolates was tested according to adapted NCCLS standards. Resistance to methicillin was most frequent in Staphylococcus epidermidis (29%) and S. haemolyticus (16%). Among the departments, thoracic surgery (29-47%), surgical intensive care (68%) and neonatology (32%) scored highest. Significant correlations were found between percentages of methicillin resistance in CNS and consumption (DDD/month/bed) of (flu)cloxacillin (P0.008), of cephalosporins (P0.01) and of gentamicin (P0.005). (Flu)cloxacillin was used mainly prophylactically, cephalosporins and gentamicin therapeutically. Results were similar for S. epidermidis (n = 639) alone. There was no significant correlation between consumption and resistance to trimethoprim, erythromycin (P0.08) or gentamicin (P0.09). Analysis of data from individual patients showed significant differences in proportions of methicillin resistance rates in CNS, between use and non-use of penicillinase resistant beta-lactams or gentamicin. It is concluded that clinical use of both (flu)cloxacillin and cephalosporins selects for methicillin resistant CNS.

[Inventory and comparison of guidelines for antibiotic utilization in Dutch hospitals]. / Inventarisatie en vergelijking van richtlijnen voor antibioticagebruik in Nederlandse ziekenhuizen.

In 1976 the Dutch Health Council advised hospitals to formulate guidelines for use of antibiotics. These guidelines and their use should improve the quality of care in terms of medical effectiveness and cost-effectiveness. In 1988 the Peer Review Council held a survey among all (140) Dutch hospitals to collect data about these guidelines. Thirty-seven sets of guidelines used in 71 hospitals were obtained. We analysed these sets of guidelines as to the following general aspects: the status of the guidelines in the hospitals, the problem-oriented approach, the topics dealt with in the guidelines and the authors of the guidelines. Four specific aspects related to rational use of antibiotics were analysed as well: the use of cephalosporin, the use of different antibiotics for prophylaxis and therapy and the cost-effectiveness of antibiotic treatment in relation to urosepsis. Results of these analyses show that guidelines are too often formulated in a noncommittal way and that there is a need for a more functional registration system to link information about the clinical working diagnoses, the bacteria isolated and the sensitivity to the antibiotics used. These linkages are essential for feedback to clinicians, microbiologists and pharmacists and a prerequisite for management of the quality of care with respect to use of antibiotics.