Search

The Alexander Project, initiated in 1992 as an international, multicentre, longitudinal surveillance study of antimicrobial susceptibility among common respiratory pathogens, has been pivotal in defining the role of global surveillance. At the time, there were few studies in which data were collected in a way that allowed meaningful comparisons to be made between studies, locations or over time. The project instituted the use of a central laboratory and standardized methods for the collection of isolates and determination of susceptibility, and this was continued with the addition of two further reference laboratories. Data from the study have provided a resource for measuring trends in the susceptibility patterns of Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis at country, regional and global levels. Determination and publication of quantitative MICs enabled detailed assessment of changes in susceptibility distributions and assessment of microbiological and potential clinical susceptibility using different breakpoints, including those based on pharmacokinetic/pharmacodynamic parameters. Comparisons of antimicrobial usage patterns and resistance prevalences over time allowed hypotheses to be examined with respect to the role of quantity and type of antimicrobial use in the selection and spread of resistance. The resulting collection of isolates has provided a valuable resource for molecular studies into the evolution of resistance over time and location; a substantial proportion of this collection is now in the public domain. This paper reviews the 10 years of the Alexander Project and the benefits it has brought to an understanding of the evolution of antibacterial resistance in community respiratory bacteria.

Objectives: The Alexander Project is a continuing surveillance study, begun in 1992, examining the susceptibility of pathogens involved in adult community-acquired respiratory tract infections (CARTI) to a range of antimicrobial agents., Materials and Methods: This study tested the susceptibility of isolates of Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis collected between 1998 and 2000 to 23 antimicrobials. Minimum inhibitory concentrations of agents were determined using the broth microdilution method and interpreted according to NCCLS and pharmacokinetic/pharmacodynamic (PK/PD) breakpoints., Results: In total, 8882 isolates of S. pneumoniae, 8523 isolates of H. influenzae and 874 isolates of M. catarrhalis were collected during 1998-2000 from centres in 26 countries. The world-wide prevalence of penicillin resistance (penicillin MICs > or = 2 mg/l) in isolates of S. pneumoniae was 18.2% over the study period, and the prevalence of macrolide resistance (erythromycin MICs > or = 1 mg/l) in this pathogen was 24.6%. Over the study period, macrolide resistance exceeded penicillin resistance in 19 of the 26 countries included in the study. Of the non-fluoroquinolone agents, the only oral agents to which over 90% of S. pneumoniae isolates were susceptible at both NCCLS and PK/PD breakpoints were amoxicillin (95.1%) and co-amoxiclav (95.5-97.9%). The prevalence of fluoroquinolone-resistant S. pneumoniae (ofloxacin MICs > or = 8 mg/l) was 1.1%. Gemifloxacin was the most potent fluoroquinolone tested against S. pneumoniae (99.9% susceptible). In isolates of H. influenzae, beta-lactamase production was 16.9%, whereas the prevalence of beta-lactamase-negative, ampicillin-resistant strains was low (0.2%). beta-Lactamase production in M. catarrhalis world-wide remained high over the period studied (92.1%). Using PK/PD breakpoints, the most active non-fluoroquinolone agents against H. influenzae were ceftriaxone (100% susceptible), cefixime (99.8%) and co-amoxiclav (98.1-99.6%). Co-amoxiclav, cefdinir and cefixime (100%) were the most active beta-lactams against M. catarrhalis. Both H. influenzae and M. catarrhalis were highly susceptible to the fluoroquinolones., Conclusions: These data demonstrate the continued evolution of and geographical variation in bacterial resistance and highlight the need for appropriate prescribing of antimicrobials in CARTI, using agents with adequate activity, based on local susceptibility profiles and PK/PD parameters.

The Alexander Project was established in 1992 to examine antimicrobial susceptibilities of bacterial isolates from community-acquired infections of the lower respiratory tract. Testing of a range of compounds was undertaken in a central laboratory. From 1992 to 1995, isolates were collected from geographically separated areas in countries in the European Union and various states in the USA. In 1996, the study was extended to include centres in Mexico, Brazil, Saudi Arabia, South Africa, Hong Kong and other European countries not included previously. Data generated by the project during 1996-1997 confirm France and Spain as European centres with high rates of resistance to penicillin among isolates of Streptococcus pneumoniae. Both intermediate (MIC 0. 12-1 mg/L) and resistant (MIC 2 mg/L) phenotypes are present. Combined resistance rates (intermediate and resistant) were >/=50% in 1997. Combined resistance rates in excess of 20% were found in the Republic of Ireland, Portugal, the Slovak Republic and Hungary. Penicillin resistance continues to evolve in the USA, with combined resistance rates of 16.4% (1996) and 18.6% (1997). In the new, non-European centres, e.g. Mexico and, in particular, Hong Kong (where resistant strains accounted for 50% of all isolates of S. pneumoniae in 1996 and 55.5% in 1997), there are centres where rates of resistance are high. Macrolide resistance is increasing generally among both penicillin-resistant and penicillin-susceptible isolates of S. pneumoniae. There is variation between countries, and in four out of the 16 centres for which both 1996 and 1997 data are available, rates of macrolide resistance have fallen. Overall, the percentage of S. pneumoniae strains that is resistant to macrolides exceeds the percentage that is resistant to penicillin. In 1996, 16. 5% of all S. pneumoniae isolates were resistant to macrolides compared with 10.4% resistant to penicillin, and in 1997 respective rates were 21.9% and 14.1%. beta-Lactamase production was the principal mechanism of resistance observed among isolates of Haemophilus influenzae. However, considerable variation in the percentage of isolates producing beta-lactamase (0-37.1%) was observed within this species. Within Europe, in the Republic of Ireland, France and Belgium, more than 15% of isolates were beta-lactamase producers. In Spain rates were as high as 31.7%. Outside Europe and the USA high rates were described in Mexico (25%), Saudi Arabia (27.9%, 16.7%) and Hong Kong (37.1%, 28.9%). Of H. influenzae from the USA, 30.4% were beta-lactamase producers in 1996 and 23.3% in 1997. beta-Lactamase production among isolates of Moraxella catarrhalis was observed in >90% of the isolates tested in 1996 and 1997.

There are several sets of guidelines for the treatment of infective endocarditis, reflecting the need for differing treatment in various countries and times. This review considers the need for differing treatment modalities and in particular the utility of the glycopeptide antibiotics vancomycin and teicoplanin. Specific recommendations are offered as to when to consider the use of glycopeptides, appropriate dosage, length of treatment course and whether to use monotherapy or combined therapy. Used judiciously, the glycopeptides give results as good as can be achieved with other antimicrobial agents without exceptional toxicity. The potential of teicoplanin for use in the outpatient treatment of infective endocarditis is considered.

A number of published studies have shown that the MICs of amoxycillin and/or co-amoxyclav are lower than those of ampicillin and/or penicillin for Streptococcus pneumoniae. Other published studies have concluded that the activities of amoxycillin and co-amoxyclav are comparable with that of penicillin for S. pneumoniae. A collection of 5252 S. pneumoniae isolates obtained during a 5 year period (1992-1996) was analysed to determine differences between the MICs of penicillin, amoxycillin and co-amoxyclav. Among the isolates analysed, 3788 (72%) were penicillin-susceptible, 615 (12%) were penicillin-intermediate and 849 (16%) were penicillin-resistant. Differences between the agents were assessed by examination of MIC distribution functions and simultaneous 95% CIs. In addition, penicillin-intermediate and -resistant isolates were analysed to determine the number and percentage of isolates which had an amoxycillin and co-amoxyclav MIC less than, equal to, or greater than the penicillin MIC. Results showed that the amoxycillin and co-amoxyclav MIC90s were one two-fold dilution lower than those of penicillin for all isolates collected between 1992-1993 and 1994-1996. Simultaneous 95% CIs showed that the mean differences between MICs of amoxycillin and penicillin, and between MICs of co-amoxyclav and penicillin, were less than zero. The majority of the penicillin-intermediate and penicillin-resistant isolates had an amoxycillin and co-amoxyclav MIC less than the penicillin MIC. In conclusion, amoxycillin and co-amoxyclav MICs were shown to be lower than the penicillin MICs for the S. pneumoniae isolates analysed in this study.

Practical guidelines for the diagnosis and treatment of chronic prostatitis are presented. Chronic prostatitis is classified as chronic bacterial prostatitis (culture-positive) and chronic inflammatory prostatitis (culture-negative). If chronic bacterial prostatitis is suspected, based on relevant symptoms or recurrent UTIs, underlying urological conditions should be excluded by the following tests: rectal examination, midstream urine culture and residual urine. The diagnosis should be confirmed by the Meares and Stamey technique. Antibiotic therapy is recommended for acute exacerbations of chronic prostatitis, chronic bacterial prostatitis and chronic inflammatory prostatitis, if there is clinical, bacteriological or supporting immunological evidence of prostate infection. Unless a patient presents with fever, antibiotic treatment should not be initiated immediately except in cases of acute prostatitis or acute episodes in a patient with chronic bacterial prostatitis. The work-up, with the appropriate investigations should be done first, within a reasonable time period which, preferably, should not be longer than 1 week. During this period, nonspecific treatment, such as appropriate analgesia to relieve symptoms, should be given. The minimum duration of antibiotic treatment should be 2-4 weeks. If there is no improvement in symptoms, treatment should be stopped and reconsidered. However, if there is improvement, it should be continued for at least a further 2-4 weeks to achieve clinical cure and, hopefully, eradication of the causative pathogen. Antibiotic treatment should not be given for 6-8 weeks without an appraisal of its effectiveness. Currently used antibiotics are reviewed. Of these, the fluoroquinolones ofloxacin and ciprofloxacin are recommended because of their favourable antibacterial spectrum and pharmacokinetic profile. A number of clinical trials are recommended and a standard study design is proposed to help resolve some outstanding issues.

Seventy laboratories in nine European countries (Belgium, France, Germany, Italy, The Netherlands, Portugal, Spain, Switzerland and the UK) each collected 100 consecutive gram-positive bacterial pathogens during 1995. MICs were determined by a co-ordinating laboratory in each country using an agar incorporation method with Mueller Hinton medium (NCCLS). Quality control was ensured by distribution of five test strains to the co-ordinating laboratories. A total of 7078 isolates was collected: 2885 Staphylococcus aureus, 1706 enterococci, 1480 coagulase-negative staphylococci (CNS), 932 Streptococcus spp. (including 289 strains of S. pneumoniae) and 75 miscellaneous species. Of these, the country coordinators successfully re-tested 6824 isolates. Using NCCLS interpretive criteria, overall 39 isolates (including 28 strains of enterococci) were teicoplanin-resistant (0.57%) and 38 (mostly CNS; 0.56%) were intermediate, whilst 32 isolates (including 30 strains of enterococci) were resistant to vancomycin (0.47%) and 7 (all enterococci; 0.10%) were intermediate. The overall resistance rate was < or = 0.5%. The two glycopeptides were essentially active against the major pathogens encountered in the survey. The only real difference with clinical implications from previously reported susceptibility data is the emergence and spread of resistance in enterococci, particularly in E. faecium. Resistance was highest in SSTI, UTI, bloodstream and GI infections; no resistance was encountered in RTI, gynaecological infections or central nervous system infections. This resistance was also geographically diverse: Resistance to vancomycin in E. faecalis was present only in France, Germany, Italy, Portugal and Spain (Italy and Spain only for teicoplanin), whilst resistance to teicoplanin and vancomycin in E. faecium was present in all countries except Spain. Eight isolates (0.5% of all enterococci) were vancomycin-resistant but teicoplanin-susceptible, exhibiting the vanB phenotype. These were four strains of E. faecalis and four strains of E. faecium. Whilst isolates of S. haemolyticus had higher MIC of teicoplanin than other CNS, and were more susceptible to vancomycin, overall resistance to teicoplanin was low (3.3% in S. haemolyticus; 0.6% in CNS). S. haemolyticus was a relatively rare pathogen, accounting for 6.3% of all CNS isolates, and 1.4% of all gram-positives collected. The results of this survey show that, despite occasional nosocomial problems (e.g. with enterococci and S. haemolyticus), teicoplanin or vancomycin remain adequate therapy for infections caused by gram-positive pathogens in the 1990s.

The antimicrobial susceptibility of 1078 isolates of Haemophilus influenzae, 348 Streptococcus pneumoniae and 258 Moraxella catarrhalis was determined. Overall 15.1% of H. influenzae produced beta-lactamase; 98.8% were susceptible to co-amoxiclav, 85.8% to cefaclor, 96% to clarithromycin and 100% to ciprofloxacin. The majority (94.2%) of M. catarrhalis produced beta-lactamase. The overall prevalence of low-level penicillin resistance (MIC = 0.12-1 mg/L) amongst isolates of S. pneumoniae was 3.4% and that of high-level resistance (MIC > or = 2 mg/L) was 3.7%. Most (96.3%) of the isolates of S. pneumoniae were susceptible to amoxycillin (MIC < or = 0.5 mg/L), 96% to cefaclor (MIC < or = 8 mg/L), 90.7% to clarithromycin (MIC < or = 0.25 mg/L) and 89% to ciprofloxacin (MIC < or = 1 mg/L).

The usefulness of many anti-Gram-positive antibiotics is being compromised by the spread of antibiotic resistance in bacteria. The most reliable agents for serious infections are the glycopeptide agents vancomycin and teicoplanin. The appropriate maintenance dosage for teicoplanin in serious infections is 6 mg/kg/day, i.e. usually 400 mg/day. There are 3 exceptions for which the daily maintenance dosage should be 12 mg/kg/day and these are intravenous drug abusers, septic arthritis (but not osteomyelitis), and Staphylococcus aureus endocarditis treated with teicoplanin monotherapy. When teicoplanin is given at these doses, it achieves clinical and bacteriological results that are equivalent to those obtained with vancomycin, irrespective of pathogen or type of infection. The toxicity profile favours teicoplanin over vancomycin, especially when other, potentially toxic, drugs are coadministered. Teicoplanin also has an advantage in terms of ease and convenience of administration, which, together with its lack of need for routine blood level monitoring, facilitates its use outside hospital. New agents hold some promise for the future; however, oral agents, if developed, could present the risk of being overused, which might compromise their long term utility.

In 1992, an ongoing, international multicenter study was established to investigate the antimicrobial susceptibility of community-acquired lower respiratory tract bacterial pathogens: the Alexander Project. Isolates cultured from patients living in geographically separated areas, ten in the European Union (EU) and five in the United States (US), were collected and tested using standard methods in a central laboratory. A total of 4,155 isolates of Haemophilus influenzae was collected during the period 1992-1994. beta-lactamase production was the principal mechanism of resistance observed with overall rates in the US (1992 = 26.3%; 1993 = 28.2%; and 1994 = 30.1%) generally twice those seen in the EU (1992 = 12.3%; 1993 = 14.4%; and 1994 = 15.5%). Chloramphenicol resistance was generally low except in Spanish centers where rates ranging from 4.0 to 15.9% were observed during the study period. One thousand one hundred ninety-three isolates of Moraxella catarrhalis were tested. beta-lactamase production was the only mechanism of resistance of any importance detected, with the vast majority of isolates producing the enzyme. Two thousand eight hundred twenty-nine isolates of Streptococcus pneumoniae were tested. French and Spanish centers provided isolates with the highest rates of either low-level (intermediate) or high-level penicillin resistance, which in 1994 ranged from 10.2 to 31.4% and 30.4 to 40.1% for each resistance category, respectively. With the exception of the fluoroquinolones, rates of resistance to other antimicrobials including the macrolides, doxycycline, chloramphenicol, and trimethoprim/sulfamethoxazole were high, generally, in centers with a high prevalence of penicillin resistance. However, in some centers (Toulouse, France and Genoa, Italy) this association was not complete for the macrolides.

An international collaborative survey of susceptibility in community-acquired lower respiratory tract infection pathogens collected > 6000 strains from six countries during 1992 and 1993. The four major pathogens were Streptococcus pneumoniae, Haemophilus influenzae, Moraxella catarrhalis and Staphylococcus aureus. MICs of 15 antibiotics were determined and sensitivity interpretations applied using breakpoints based on those of the NCCLS. This analysis highlighted some anomalies, notably for beta-lactams against S. pneumoniae and macrolides against H. influenzae, where apparent sensitivity proportions did not accord with the distribution of MICs. Further analyses were undertaken in order to rank the antibiotics in order of potential usefulness for empirical treatment of LRTI: these included in-vitro potency (mode MIC and MIC90) and a pharmacodynamic comparison, using the ratio Cmax (free drug): MIC90. According to study breakpoints, the most active agents overall were, for S. pneumoniae, cefuroxime, clarithromycin, ofloxacin and chloramphenicol; for H. influenzae, azithromycin, amoxycillin/ clavulanate, cefixime, ceftriaxone, quinolones and doxycycline. However, other analyses suggested that the most active agents overall were, for S. pneumoniae, amoxycillin (+/- clavulanate) and ceftriaxone, and, for H. influenzae, quinolones, ceftriaxone, cefixime and amoxycillin/clavulanate. Overall, the antimicrobial agents with the greatest potential usefulness for empirical treatment were amoxycillin/ clavulanate, ceftriaxone, cefuroxime, ofloxacin and co-trimoxazole. The choice of empirical therapy depends upon local epidemiology and clinician choice, but the Project data may be of value in the decision-making process.

Antibiotics are used in 80% of patients in the ICU, encouraging nosocomial infections with resistant organisms. If the antibiotic susceptibilities of the pathogen are known, a narrow-spectrum antibiotic is preferable to preserve the patient's resistance to colonization. However, treatment is often empirical and broad-spectrum combinations are commonly used. Gram-positive bacteraemia is associated with invasive monitoring or intravascular catheters. If the device cannot be removed easily, the glycopeptides are the only agents likely to be active against most strains of the commonest pathogen, the coagulase-negative staphylococcus. Long-stay patients are susceptible to infection with enterococci and methicillin-resistant Staphylococcus aureus, which are often resistant to all the usual agents other than glycopeptides. Vancomycin is long established, but is nephrotoxic, requires serum monitoring, must be administered as an infusion and can cause red man syndrome. Teicoplanin can be given as a single daily bolus without similar side-effects or monitoring. In deep-seated staphylococcal infection, the usual dose of teicoplanin is adequate if given in combination with other agents, but it may need to be doubled if used as monotherapy. Monitoring of the levels in the serum is helpful to ensure an adequate dose in patients with renal failure or in drug abusers, but is not needed to prevent toxicity.

Despite potential nephrotoxicity and ototoxicity, aminoglycosides are widely used in the treatment of severe sepsis but monitoring of serum levels is necessary. The use and assay of aminoglycosides, together with microbiologists' and pharmacists' advice on dosage and potential toxicity were examined in a teaching hospital group during an eight week period. A total of 480 courses of aminoglycoside was recorded in 440 hospital in-patients and of these, 306 (64%) were for prophylaxis. For 79 (45%) of 174 therapeutic courses, bacteriological results were available at the start and the choice of antibiotic was appropriate in 69 cases. Serum aminoglycoside assays were performed in 86% (149/174) of therapeutic courses. Correct assay request forms were used in 79% of 473 assay requests, and the clinical diagnosis was given on 73% of forms. In so-called peak samples with a documented time of collection, this ranged from 5 min to over 9 h post dose. Potentially toxic concentrations were present in 56 (12%) serum samples from 50 courses, in ten of which there was a serum creatinine rise of > 40 microM. However, there were confounding factors in five cases. Probable nephrotoxicity was rare, although assays were not performed in 14% of therapeutic courses. There were no symptomatic cases of ototoxicity but audiometry and vestibular studies were not performed. Closer liaison between the microbiologists, pharmacists and clinical staff is essential to improve clinical practice. Computer notification of inadequately monitored courses would be helpful. The routine therapeutic use of aminoglycosides needs to be reviewed in the light of the rising costs of assay litigation and the increasing number of alternative antibiotics available.

Antibiotic assays are most often ordered for the purpose of toxicity monitoring, which usually involves determination of peak and trough antibiotic concentrations in the blood. The cost of monitoring antibiotic levels is probably higher than is commonly appreciated. Factors that contribute to the cost of this service include staffing the microbiology laboratory with appropriate personnel, who are responsible for determining the adequacy of sample collection and related patient information; analyzing the specimens in a timely manner; and taking action to modify drug dosage and dosage intervals in light of the test results. There are also costs related to the reagents, consumables, and equipment used in the assay, as well as to revenue and capital overheads. Additional clinical and laboratory costs can be incurred in the event of litigation pursuant to antibiotic-induced toxicity. With hospital and government policymakers devoting increasing attention to the escalating costs of health care, pressure to move away from the routine use of drugs having dose-related toxicity may increase, the objective being to save on assay costs.

Significant urinary isolates have been prospectively recorded since 1971. Enterococcus species, a common cause of nosocomial urinary tract infection, have been identified, and susceptibilities to a range of antibiotics have been determined. In addition, isolates in 1988 were tested for breakpoint susceptibility to vancomycin and teicoplanin. Despite changes in the hospitals covered, isolation of Enterococcus species rose steadily from 4% in 1971 to 12.6% in 1990 in hospital patients and from 2% to 5.6% in general-practice patients (P less than .01). All isolates of Enterococcus species were sensitive to ampicillin. Teicoplanin inhibited all 526 strains tested at a concentration of 2 micrograms/mL, but the same concentration of vancomycin inhibited only 370 (70%). The increased prevalence of enterococcal urinary tract infection is probably the result of increasing use of catheterization and broad-spectrum antibiotics. Glycopeptides reach high levels in the urine, and teicoplanin might be an alternative for the treatment of urinary tract infections due to enterococci.

Ramoplanin is a novel lipoglycopeptide antimicrobial complex, isolated from the fermentation products of a strain of Actinoplanes sp. (ATCC 33076), which comprises three closely related polypeptides, each containing chlorinated phenyl moieties and D-mannose. The in-vitro activity of ramoplanin was compared with those of vancomycin and teicoplanin. Ramoplanin was very active against Staphylococcus spp., irrespective of methicillin susceptibility, with all isolates inhibited by 1 mg/l or less. Concentrations of vancomycin and teicoplanin required to inhibit the same population of bacteria were 4 and 16 mg/l, respectively. Ramoplanin was also very active against Streptococcus spp. (alpha- and beta-haemolytic species, Str. pneumoniae and Enterococcus faecalis, Corynebacterium spp. (including Cory. jeikeium), Listeria monocytogenes, Gardnerella vaginalis, Propionibacterium acnes and Gram-positive anaerobic bacteria, with all isolates inhibited by 2 mg/l, or less. In general, the activity of ramoplanin against these species was either equal to or only slightly less than teicoplanin and equal to or somewhat greater than vancomycin. With the exception of Bacteroides melaninogenicus and Bact. bivius, ramoplanin was not active against Gram-negative bacteria.

Two prospective trials suggested that teicoplanin was less effective in preventing Gram-positive wound infections after cardiac surgery than a combination of flucloxacillin and an aminoglycoside. The initial dose of 400 mg was associated with a subinhibitory concentration of teicoplanin in fat at the end of operation. In this study, the behaviour of a higher initial dose (12 mg/kg) was investigated in ten patients during coronary artery surgery. As in the earlier trials, a second dose (400 mg) was given after 24 h. The mean concentration at the end of cardiopulmonary bypass was 15 mg/l in serum, 6 mg/kg in fat and 9 mg/kg in skin, all in excess of the expected break-point for staphylococci (4 mg/l). The higher dose regimen might be more successful in surgical prophylaxis.

A controlled trial of antibiotic prophylaxis in cardiac surgery compared a two-dose regimen of teicoplanin with a longer conventional course of flucloxacillin and tobramycin. In 12 patients the susceptibility of the bacterial skin flora of four different sites to each of the three antibiotics was determined and the results are reported here. Less than 1% of the Gram-positive colonies showed reduced sensitivity to teicoplanin (MIC greater than or equal to 4 mg l-1). Before operation, 99% inhibition of Gram-positive growth was achieved at 26 (54%) of 48 sites by 1 mg l-1 of flucloxacillin and 13 (27%) sites by 2 mg l-1 tobramycin. By the 7th day after operation there was a significant reduction in the number of sites showing similar sensitivity to flucloxacillin [16 (33%) sites, P less than 0.05]. The use of teicoplanin was not associated with the emergence of Gram-negative skin flora but tobramycin promoted acquisition of aminoglycoside-resistant strains.

Ramoplanin (A-16686/MDL 62,198) is a novel lipoglycopeptide antimicrobial, comprised of three closely related polypeptides containing chlorinated phenyl moieties and D-mannose, isolated from the fermentation products of Actinoplanes sp. ATCC 33076. The antimicrobial activity of ramoplanin is limited to Gram-positive bacteria and its reportedly unacceptable administration side-effects suggest that any potential clinical role will be limited to the topical therapy of superficial skin infections and the eradication of bacteria, representing a possible nosocomial cross-infection source, from carriage sites. In this study the MICs of ramoplanin have been determined for methicillin-susceptible and methicillin-resistant isolates of Staphylococcus aureus, S. epidermidis and S. haemolyticus and compared with those of two glycopeptide antimicrobials, vancomycin and teicoplanin. MICs were determined using an agar incorporation technique in Mueller-Hinton medium with an inoculum of 10(5) cfu. Ramoplanin was 2-8 times more active than either vancomycin or teicoplanin against methicillin-susceptible and methicillin-resistant isolates of S. aureus and methicillin-susceptible isolates of S. epidermidis. Isolates of methicillin-resistant S. epidermidis and both methicillin-susceptible and -resistant isolates of S. haemolyticus were generally less susceptible to teicoplanin than to vancomycin. Ramoplanin was significantly more active than either vancomycin or teicoplanin against these isolates. These results suggest that the clinical evaluation of ramoplanin as a topical antibacterial agent for the control of superficial infections caused by Staphylococcus spp. and for the eradication of methicillin-resistant S. aureus from carriage sites, is justified.

The minimal inhibitory concentrations (MICs) of twelve 4-quinolone antimicrobials were determined for Salmonella typhi (25), Salmonella spp. (50), Shigella spp. (50), Campylobacter jejuni (100), Vibrio cholerae (10), Vibrio parahaemolyticus (10), Yersinia enterocolitica (25), Aeromonas hydrophila (25) and Plesiomonas shigelloides (10). MICs were determined using an agar dilution technique in Mueller-Hinton agar (Oxoid, England) supplemented with 10% lysed horse blood. Antibiotic containing plates were inoculated with approximately 10(4) colony forming units of each organism, contained in 10 microliters of Mueller-Hinton broth (Oxoid, England), using a multipoint inoculator. Following inoculation plates were incubated aerobically for 18 hours at 37 degrees C, except for plates inoculated with Campylobacter jejuni which were incubated microaerophilically for 48 hours at 37 degrees C. The MICs of each antimicrobial for each isolate examined, together with the minimum concentrations of each antimicrobial required to inhibit 50% (MIC50) and 90% (MIC90) of the isolates examined, were also determined. The more recently synthesized 4-quinolones showed very good activity against all of the enteric pathogens examined with ciprofloxacin being the most active (MIC90: Salmonella typhi 0.015 microgram/ml, Salmonella spp. 0.015 microgram/ml, Shigella spp. 0.015 microgram/ml, Campylobacter jejuni 0.12 microgram/ml, Vibrio cholerae 0.008 microgram/ml, Vibrio parahaemolyticus 0.06 microgram/ml, Yersinia enterocolitica 0.015 microgram/ml, Aeromonas hydrophila 0.015 microgram/ml and Plesiomonas shigelloides 0.015 microgram/ml. Where considered clinically appropriate these compounds may have a useful role in the treatment and prevention of diarrhoeal disease caused by these enteric pathogens.

All urinary pathogens from general practice and from hospital have been tested for sensitivity to a range of antimicrobial drugs for the last 12 years. During that period there have been marked changes. In general practice there has been a marked increase in the proportion of staphylococcal infections, from 5.1% to 14.8%, and a noticeable decrease in the proportion caused by Proteus mirabilis, from 9.2% to 4.1%. Similar, but smaller, changes have been seen in the proportions of hospital UTI caused by those organisms, while the proportion of hospital infections due to Klebsiella-Enterobacter spp. has fallen from 16.8% to 8.3%. These, and other, changes have been reflected in changing antibiotic sensitivity patterns. In particular, sensitivity of urinary pathogens to ampicillin/amoxycillin and to cephaloridine has continued to fall both in general practice and in hospital. In general practice UTI nalidixic acid-resistance is becoming more important as the proportion of Gram-positive urinary pathogens increases. There has been little change in sensitivity to trimethoprim or co-trimoxazole over the last 12 years.

Antibiotic-resistant coagulase-negative staphylococci are a frequent cause of infection in indwelling central venous (Hickman) catheters. Teicoplanin has been evaluated in the treatment of 19 immunosuppressed patients with staphylococcal Hickman catheter infections, nine of whom were septicaemic. All infections were eradicated, with minimal side effects. In 16 cases, the catheter was retained until no longer required. Two recurrent infections were eradicated by a second course of teicoplanin. We conclude that teicoplanin is an effective and well-tolerated antibiotic in the treatment of Hickman catheter infections in immunosuppressed patients.

The in vitro activity of aztreonam, cefuroxime and ceftazidime was determined against 2,372 Gram-negative rods (including Pseudomonas spp.) isolated from hospital patients with urinary tract infections during 1985. Minimum inhibitory concentrations (MICs) were determined using an agar incorporation technique in Mueller-Hinton agar. The inoculum used was approximately 10(5) colony forming units (cfu) contained in 10 microliter Mueller-Hinton broth, which was applied to the surface of the agar plates using a multipoint inoculator. Following inoculation plates were incubated aerobically at 37 degrees C for 18 h. The MIC of each antimicrobial for each organism examined was determined as the lowest concentration of the antimicrobial which completely inhibited growth of the inoculum. The minimum concentration required to inhibit the growth of 90% (MIC90) of the bacterial isolates in each genus or species examined was also determined. In general the antibacterial spectrum of aztreonam was comparable to that of ceftazidime and superior to that of cefuroxime. Against Escherichia coli, which accounted for 72% of the isolates examined, aztreonam (MIC90 less than or equal to 0.25 microgram/ml) was slightly more active than ceftazidime (MIC90 0.5 microgram/ml) and considerably more active than cefuroxime (MIC90 8 micrograms/ml). Aztreonam was active against Pseudomonas spp. (MIC90 16 micrograms/ml), although somewhat less so than ceftazidime (MIC90 4 micrograms/ml). Cefuroxime showed low activity against this genus (MIC90 greater than 128 micrograms/ml).

To compare antibiotic regimens for their effectiveness in preventing or treating wound sepsis, well-defined criteria for outcome are needed. A method of assessing wound healing has been devised that defines carefully the characteristics to be considered and how they are to be awarded points. Objective criteria are also included in the assessment. Points are given for the need for Additional treatment, the presence of Serous discharge, Erythema, Purulent exudate, and Separation of the deep tissues, the Isolation of bacteria, and the duration of inpatient Stay (ASEPSIS).

A prospective randomized clinical trial is in progress to compare the efficacy of teicoplanin with flucloxacillin and tobramycin in the prevention of endocarditis and wound infection following cardiac surgery. To date, 198 patients have completed the trial, of whom 95 have received teicoplanin and 103 flucloxacillin and tobramycin. One patient developed prosthetic valve endocarditis 3 months after surgery covered by flucloxacillin and tobramycin. There was no significant difference in the incidence of sternal wound infection (P = 0.15). Severe sternal sepsis occurred in four patients in the teicoplanin group and two in the flucloxacillin/tobramycin group. There were more postoperative urinary tract infections among those given teicoplanin (15 of 95 patients compared to six of 103 patients P less than 0.05). The trial continues.

In this study, the pharmacokinetics of teicoplanin have been studied in serum, fat and bone during and after cardiac surgery with two dose (400 and 200 mg) and three dose (400 mg each) regimens in a total of 49 patients. For comparison, 20 other patients, who had received a regimen of flucloxacillin (500 mg qds for five days) and tobramycin (1.5 mg/kg initially then 80 mg tds for three days), were similarly investigated. The lowest mean serum level of teicoplanin during operation was 6.9 mg/l in the two dose regimen and 9.7 mg/l with the three dose regimen. Mean serum levels of flucloxacillin and tobramycin fell to 7 and 1.4 mg/l respectively by the end of operation. At the end of bypass, fat washings contained a mean of 1.2 mg/l of teicoplanin and bone washings a mean of 6.3 mg/l. Mean tobramycin levels were 0.4 and 0.9 mg/l respectively and flucloxacillin less than 1 mg/l in washings from both tissues. The low levels of teicoplanin in fat may explain an excess of Gram-positive sternal infections in the accompanying clinical trial.

Ciprofloxacin is a new fluorinated 4-quinolone with a broad spectrum of antimicrobial activity which includes both Gram-negative and Gram-positive bacteria. In this study the in vitro activity of ciprofloxacin has been determined against bacteria associated with respiratory tract infections and compared with that of other antimicrobial agents used in the therapy of such infections. Ciprofloxacin (MIC90 0.008 mg/l) was highly active against Haemophilus influenzae, including isolates producing beta-lactamase which were resistant to amoxycillin. Ciprofloxacin (MIC90 0.06 mg/l) was also highly active against Branhamella catarrhalis, again including those isolates resistant to amoxycillin as a result of beta-lactamase production. Isolates of Streptococcus pneumoniae were less susceptible to ciprofloxacin (MIC90 2 mg/l) but were highly susceptible to amoxycillin (MIC90 less than 0.12 mg/l) and erythromycin (MIC90 0.25 mg/l). Isolates of Klebsiella aerogenes were highly susceptible to ciprofloxacin (MIC90 0.06 mg/l) but much less so to amoxycillin, sulfamethoxazole, trimethoprim, oxytetracycline and erythromycin. Ciprofloxacin (MIC90 0.5 mg/l) was very active against Staphylococcus aureus, including those isolates resistant to amoxycillin and flucloxacillin, and against Mycoplasma pneumoniae. Together with rifampicin and erythromycin, ciprofloxacin was highly active against Legionella pneumophila (MIC90 0.015 mg/l). These results suggest that clinical evaluation of ciprofloxacin in the treatment of respiratory tract infections is justified.

FCE 22101 (sodium [5R, 6S, 8R]-6-alpha-hydroxyethyl-2-carbamoyloxymethyl-2- penem-3-carboxylate) is a new penem antimicrobial with a broad spectrum of activity. In order to investigate the pharmacokinetics of FCE 22101 in human volunteers, a high performance liquid chromatographic assay has been developed for determining concentrations of FCE 22101 in serum. Serum samples were deproteinised with an equal volume of acetonitrile followed by centrifugation. The supernatant was then diluted with an equal volume of 0.05 M phosphate buffer (pH 7.0) and 50 microliters was applied to the separation column. Chromatographic separation of FCE 22101 was achieved using an SP8700 solvent delivery system (Spectra-Physics), an SP8780XR autosampler (Spectra-Physics) and a reverse-phase C18 mu-BondaPak column (8 X 100 mm) radially compressed in a Z module (Millipore, UK). The mobile phase consisted of a water, methanol and orthophosphoric acid mixture (74:25:1), which was filtered through a 0.22 micron filter and de-gassed under vacuum prior to use. The flow rate was 1.4 ml/min and FCE 22101 was detected in the column eluant using an LC871 UV-Vis detector set at 318 nm. The data were analysed using an SP4270 integrator (Spectra-Physics). The lower limit of sensitivity of the assay method was 0.8 mg/l and recovery from deproteinised serum was of the order of 70%. Intra-assay variation was less than 5% and inter-assay variation was less than 10%.

Teicoplanin, a new glycopeptide antibiotic, has a serum half-life of 47 h and excellent activity against Gram-positive bacteria, including methicillin resistant staphylococci, making it a potentially useful drug for cardiac surgical prophylaxis. In two prospective randomized studies, we have compared it with a broad spectrum regimen of an aminoglycoside and flucloxacillin. In the first trial, teicoplanin (400 mg on induction of anaesthesia and 200 mg 24 h later), was compared with tobramycin (80 mg tds for three days) and flucloxacillin (500 mg qds for 5 days) in 314 patients. Teicoplanin prophylaxis resulted in a significantly greater number of sternal wound infections (P less than 0.01), due to Gram-positive bacteria. Furthermore, Gram-negative bacteria were responsible for more respiratory and urinary infections after teicoplanin prophylaxis. In the second trial, comprising 203 patients, the teicoplanin dose regimen was changed to three doses of 400 mg but this did not improve the rates of infection.

Despite antibiotic prophylaxis in cardiac surgery, gram-positive bacteria can be isolated in up to 10% of intraoperative blood cultures. During a prospective randomized trial, blood was collected from the oxygenator at the end of bypass in 58 patients given teicoplanin and in 60 others given flucloxacillin and tobramycin. Coagulase-negative staphylococci were cultured from 16 patients given teicoplanin but in only four cases after flucloxacillin and tobramycin (Fisher's exact test, P = 0.005). In contrast, Propionibacterium spp. or coryneforms were isolated from 22 patients given flucloxacillin and tobramycin and from only one patient in the teicoplanin group. There were no cases of prosthetic valve endocarditis. After 3 h exposure to 4 x MIC of teicoplanin there was only a 10-60 fold reduction in cfus of Staphylococcus epidermidis, which may partly explain the excess of these organisms.

Teicoplanin was used to treat 94 hospital in-patients of confirmed or presumed Gram-positive infections over a period of 12 months. Eighty-five patients were subsequently found to be evaluable; 31 had soft tissue infections, 10 endocarditis, 8 urinary tract infections, 12 septicaemias, 2 chest infections, 7 osteomyelitis or septic arthritis, and 15 were immunosuppressed patients with infected Hickman line site infections. The cure rate of the 85 evaluable episodes was 90% (76 cured). Teicoplanin was well tolerated intravenously and intramuscularly. Adverse reactions occurred in five patients. One patient suffered high tone hearing loss, two patients suffered transient rash, one developed a drug fever and one patient who had concomitant gentamicin developed vestibular damage. It is concluded that teicoplanin is a relatively safe and effacacious treatment for Gram-positive infection.

An accurate and sensitive bioassay for determining concentrations of teicoplanin in serum has been developed using modifications of procedures described for assaying vancomycin. A linear relationship (r = 0.9983) was obtained between the diameter of the zone of inhibition and log10 teicoplanin concentration over the range 0.25-32 micrograms/ml. The medium used was Antibiotic Medium No. 1 (Oxoid, UK) adjusted to pH 5.5-5.7 by the addition of hydrochloric acid and containing sodium chloride to a final concentration of 3%. The indicator organism used was a Bacillus subtilis ATCC 6633 (NCTC 10400). Teicoplanin was assayed: (i) in the presence of beta-lactams and cephalosporins, including ceftazidime, by prior treatment of serum with broad-spectrum beta-lactamase mixtures (Genzyme Laboratories, UK); (ii) in the presence of aminoglycosides by prior treatment of serum with cellu-ion phosphate (100 mg/ml) followed by centrifugation; (iii) in the presence of sulphamethoxazole and/or trimethoprim by the addition of p-aminobenzoic acid and/or thymidine to the assay medium. Teicoplanin was assayed in the presence of either rifampicin or erythromycin by using a rifampicin-resistant, erythromycin-resistant clinical isolate of Staphylococcus aureus as indicator organism. The lower limit of sensitivity of this assay was 1 microgram/ml. The presence of undeclared, broad-spectrum antimicrobials was detected by screening serum samples for antimicrobial activity using an assay plate seeded with Escherichia coli NCTC 10418.

The minimal inhibitory concentrations (MICs) of twelve 4-quinolone antimicrobials were determined for the Bacteroides fragilis group (50), Bacteroides melaninogenicus (20), Bacteroides bivius (10), Fusobacterium spp. (10), anaerobic Gram-positive cocci (50) and Clostridium spp. (20). MICs were determined using an agar dilution technique in Mueller-Hinton agar supplemented with 10% lysed horse blood. The inoculum used was approximately 10(4) colony-forming units, contained in 10 microliter of Mueller-Hinton broth, which was applied to the agar plates using a multipoint inoculator. Following inoculation, plates were incubated at 37 degrees C for 48 h in an anaerobic atmosphere. The MIC of each antimicrobial for each isolate examined was determined as the lowest concentration of the antimicrobial which completely inhibited growth of the inoculum. The minimum concentrations required to inhibit the growth of 50% (MIC50) and 90% (MIC90) of the organism examined were also determined. All of the more recently synthesised 4-quinolones showed increased activity against the anaerobic bacteria used in this study. Ciprofloxacin and ofloxacin were the most active compounds examined (Bacteroides fragilis group MIC90 ciprofloxacin 4 micrograms/ml; ofloxacin 4 microgram/ml; Bacteroides melaninogenicus MIC90 ciprofloxacin 2 micrograms/ml, ofloxacin 2 micrograms/ml; Bacteroides bivius MIC90 ciprofloxacin 16 micrograms/ml, ofloxacin 32 micrograms/ml; Fusobacterium spp. MIC90 ciprofloxacin 2 micrograms/ml, ofloxacin 4 micrograms/ml; Clostridium spp. MIC90 ciprofloxacin 1 microgram/ml, ofloxacin 1 microgram/ml and anaerobic Gram-positive cocci MIC90 ciprofloxacin 4 micrograms/ml, ofloxacin 4 micrograms/ml).

The sensitivites of urinary pathogens from general practice and from hospital to a range of antimicrobial drugs have been recorded for the period 1971-8. There have been changes in the proportions of the different bacterial species and in their sensitivites to antibiotics. In particular, the position of ampicillin/amoxycillin and cephalosporins has deteriorated, partly because more resistant species have somewhat increased in prevalence and partly because the usually sensitive species, such as Escherichia coli, have become more resistant. Over the period 1971-8 the sensitivity of urinary pathogens, whether in general practice or in hospital, to co-trimoxazole and to trimethoprim has been maintained at a high level.

An in vitro technique is described for testing the proposition that combining two antibiotics prevents the emergence of bacterial resistance to either. The model was applied to testing the combination of sulphamethoxazole and trimethoprim (co-trimoxazole). Under the test conditions when a doubly sensitive organism was exposed to sulphamethoxazole or trimethoprim alone, resistance to sulphamethoxazole or trimethoprim readily emerged. When exposed to both drugs together, the doubly sensitive organism was eliminated from the cultures without emergence of bacterial resistance to either drug. A sulphamethoxazole-resistant organism exposed to trimethoprim alone or to sulphamethoxazole plus trimethoprim developed trimethoprim resistance with equal facility. The experiments show that the use of sulphamethoxazole-trimethoprim prevents the emergence of bacterial resistance among doubly-sensitive strains.

Teicoplanin, a new glycopeptide antibiotic, has been used to treat twelve patients with bacterial endocarditis due to Gram-positive organisms. Teicoplanin has activity against Gram-positive bacteria similar to vancomycin but therapeutic levels are maintained by a single daily dose, given as an intravenous bolus. Of six patients with native valve infections, two cases, due to viridans streptococci, were successfully treated with teicoplanin alone and two others, caused by Streptococcus faecalis, were cured by combinations including teicoplanin. One of these patients sustained high tone hearing loss during treatment. The remaining two patients were drug addicts with endocarditis due to Staphylococcus aureus which recurred despite repeated multiple therapy. Of six prosthetic valve infections, antibiotic combinations including teicoplanin cured three cases, caused by streptococci. Infection persisted or treatment was curtailed in three cases of Staphylococcus epidermidis endocarditis. In this small open study, teicoplanin appeared as effective as vancomycin in the treatment of endocarditis but had the considerable advantage of ease of administration.

A survey was conducted of the current practice of antibiotic prophylaxis in cardiac surgery throughout the United Kingdom. Most surgeons (84%) use a regimen covering a broad spectrum of bacterial species that is continued for two to three days after the operation. The most used regimens are a combination of beta lactamase resistant penicillin with an aminoglycoside (44%) or a single broad spectrum cephalosporin (30%). Just 16% of surgeons preferred a narrow spectrum regimen effective against only the Gram positive organisms commonly responsible for postoperative infection in these patients. Antibiotic prophylaxis has been adopted by all cardiac surgeons in the United Kingdom but is sometimes continued longer than is indicated by the clinical or experimental evidence.

The results of testing all urinary pathogens from general practice and hospital practice for their sensitivity to a range of antimicrobial agents over the four years 1971-74 are presented. The changes observed in each situation with the passage of time are discussed. Resistance has become more frequent to those drugs to which resistance is commonly R factor-mediated. Less change has been observed in resistance to other antimicrobial agents.

The occurrence of coliform bacilli carrying resistance-transfer factors (R factors) in children was studied. The frequency of R+ coliform bacilli as causes of urinary-tract infection acquired outside hospital was found to be similar to that in adults from the same geographical area and in the same years. The frequency of R+ coliform bacilli in the faeces in our children was also similar to that in the adult population, and oral chemotherapy produced similar changes in the faecal flora.

R factors conferring a high level of resistance to trimethoprim have been identified in one strain of Escherichia coli and one of Klebsiella aerogenes, both isolated from infected urines.