Search

Objectives: Performance evaluation of routine laboratory methods to determine the susceptibility of Enterobacterales urinary isolates to fosfomycin (oral administration) and mecillinam., Methods: We collected 347 Enterobacterales isolates from monomicrobial midstream urine samples from women with significant bacteriuria and leukocyturia. Mostly non-Escherichia coli isolates (i.e. Klebsiella spp., Citrobacter koseri, Enterobacter cloacae complex and Proteus mirabilis) were included (n = 298). Performance of VITEK®2, ETEST®, and disc diffusion to determine fosfomycin and mecillinam susceptibility was evaluated following International Organization for Standardization (ISO) 20776-2:2021 (or 20776-2:2007 for disc diffusion) in comparison with the agar dilution reference method., Results: For fosfomycin testing, VITEK®2 and ETEST® were close to reaching ISO requirements (essential agreement  ≥ 90%; bias  ±30%) for C. koseri, E. coli and P. mirabilis. Categorical agreement (CA) and major error rates were acceptable for disc diffusion. Fosfomycin displayed lower activity against E. cloacae complex and Klebsiella spp., with MIC50 (minimum inhibitory concentration required to inhibit the growth of 50% of tested isolates) equal to the E. coli EUCAST breakpoint (8 mg/L). For these species, the three alternative techniques overestimated MICs and resistance, and did not meet performance criteria. For mecillinam testing of Enterobacterales isolates, apart from P. mirabilis, ETEST® nearly fulfilled ISO requirements, and CA rates were acceptable for disc diffusion. ISO criteria were reached for C. koseri and E. coli testing with VITEK®2, apart from too high rates of very major errors. For P. mirabilis, performances were unacceptable, whatever the routine method used., Conclusions: Commercially available tests may serve as alternatives to agar dilution to assess fosfomycin (oral) and mecillinam susceptibility of Enterobacterales urinary isolates, with important interspecies variabilities. Additional studies comprising more fosfomycin- and mecillinam-resistant isolates are needed to strengthen our conclusions., (© The Author(s) 2024. Published by Oxford University Press on behalf of British Society for Antimicrobial Chemotherapy.)

Bacteriuria in pregnancy, with or without clinical symptoms, is frequent. If left untreated, it can in 20-30% of cases lead to acute pyelonephritis, which is a serious threat to the mother and fetus, increasing the risk of preterm labour and low birthweight infants. This paper is a review of the literature concerning antibacterial treatment of bacteriuria in pregnancy. It is crucial to ensure that drugs to be used in pregnancy are safe and effective. Established first-line drugs such as ampicillin (pivampicillin) and amoxycillin, and other commonly used treatments such as trimethoprim-sulphamethoxazole, are associated with a high degree of resistance in Escherichia coli, the most common pathogen in the urinary tract. A recent survey of physicians in Denmark, Finland, Norway and Sweden confirms that beta-lactam antibiotics (particularly pivmecillinam) and nitrofurantoin are the drugs of first choice in the treatment of bacteriuria in pregnancy in the Nordic countries. No teratogenic effects have been associated with these agents. In contrast to nitrofurantoin, pivmecillinam is also efficient against pyelonephritis. In spite of resistance in E. coli and possible adverse effects on the fetus, many physicians still prescribe sulphonamides during the first two trimesters of pregnancy.

Case records from 219 female patients between 1975 and 1992 who were given long-term prophylaxis (1 year) with nitrofurantoin for the prevention of recurrent urinary infections have been reviewed. Patients' age ranged from 9 to 89 years (median 31-35 years, mode 26-30 years); most (61%) were < 40 years old. The median number of symptomatic episodes in the 12 months immediately before prophylaxis was six (mode 4, mean 6.9). 14.4% of the patients were allergic to an antibiotic, and 23.6% had an imaging abnormality. Three regimens were used: group A (43 patients), 50 mg microcrystalline nitrofurantoin, bd; group B (110 patients), 100 mg macrocrystalline nitrofurantoin (Macrodantin), od; group C (66 patients), 50 mg Macrodantin, od. There were no obvious differences in efficacy between the patient groups (173 assessable patients). The mean incidence of symptomatic episodes decreased 5.4-fold during prophylaxis. Four-fifths of the 43 breakthrough infections (mostly due to Escherichia coli), were caused by nitrofurantoin-sensitive strains. An important finding was that patients with an imaging abnormality responded as well as those with no such abnormalities. In 16% of patients, prophylaxis was not helpful, objectively or subjectively, for no obvious reasons. In most patients where prophylaxis was successful, clinical improvement was maintained for at least 6 months after the end of prophylaxis. Nausea was more common in group A (P < 0.001), as were 'all adverse events'. Of those in group A 25.6% stopped prematurely as a result of an adverse event of any type, compared with 13% of those taking Macrodantin (P < 0.01). Older patients (> 65 years) did not report more adverse events than younger patients. No adverse event was life-threatening. Faecal flora analysis showed neither overgrowth by nitrofurantoin-resistant bacteria nor elimination of sensitive coliforms. Thus, macrocrystalline nitrofurantoin 50 mg at bedtime is appropriate for use in the long-term (12 months) prophylaxis of recurrent urinary infections, in view of its efficacy and favourable safety and tolerability profile. Patients can be managed by their family doctor.

Recent experimental evidence implies that polyunsaturated fatty acids (PUFAs) possess anti-infective activity which is unrelated to any alteration of eicosanoid biosynthesis or cytokine production provoked by PUFAs and it seems necessary to establish their possible influence on Gram-negative bacteria. Forty-two Escherichia coli strains were cultured in vitro in the presence of gamma-linolenic acid (GLA) at concentrations of 50 mg/L and 300 mg/L and a total of 77 killing curves were performed with GLA concentrations of 100, 200 and 300 mg/L GLA. At 50 and 300 mg/L, GLA inhibited 9.5% and 33.3% of strains respectively; GLA killing curves demonstrated a > or = 1 log GLA inhibitory effect in 0%, 18.2% and 72.7% of strains after the 5 h sequential exposures at 100, 200 and 300 mg/L GLA but this was not observed after the 24 h GLA exposure. Following removal of the E. coli strains from the culture medium with GLA, > or = four-fold increases in MICs and/or MBCs of various antimicrobials, were observed in 42.9% and 60% of strains after exposure to 50 and 300 mg/L GLA respectively; most of these increases involved aminoglycosides. The reproducibility of GLA inhibitory effects and increase in MICs and/or MBCs for E. coli in the two different experimental procedures used, was 82% and 73% respectively. Further studies are necessary to clarify the mechanism of GLA action on E. coli and assess the clinical relevance of these findings.

The in-vitro susceptibilities of aerobic bacteria isolated from 1804 blood and 4529 urine specimens collected at nine hospitals in the UK were examined. An agar dilution method was used to determine the MICs of each isolate to three cephalosporins, cefotaxime, cefuroxime and ceftazidime, and to two fluoroquinolones, ofloxacin and ciprofloxacin. Sensitivities were then calculated using British Society for Antimicrobial Chemotherapy recommended breakpoints. Of the cephalosporins tested cefotaxime was the most active against the Enterobacteriaceae. All the systemic staphylococcus isolates collected were sensitive to both cefotaxime and cefuroxime. As expected, ceftazidime was the only cephalosporin active against the Pseudomonas isolates. Both quinolones were highly active against the Enterobacteriaceae and Pseudomonas spp. They also demonstrated good Gram-positive activity, particularly against Staphylococcus aureus and Enterococcus spp.

The evaluation of new regimens for treatment of urinary tract infections (UTIs) has often been unsatisfactory. A major reason has been the lack of internationally accepted guidelines for design and conduct of UTI trials. The introduction of new guidelines has largely overcome this problem. However, some problems remain, and this review especially addresses the problem of defining outcomes of treatment of UTI. New, somewhat controversial definitions of bacteriuria are well documented for various types of UTI, but have not been properly evaluated as end-points for defining bacteriological responses to treatment.

DNA probes for specific integrase genes were used to study 122 R plasmids encoding the predominant trimethoprim-insusceptible type Ia dihydrofolate reductase (DHFR) found in clinical isolates of Escherichia coli. The predominance of the type Ia DHFR was thought to result from the location of its gene on transposon Tn7, but of trimethoprim R plasmids carrying this gene that were collected between 1978 and 1983, between 1987 and 1988, and during 1992, only 49/60 (81.6%), 30/43 (69.8%) and 9/19 (47.4%) respectively hybridized with a probe for the Tn7 integrase gene. It has been suggested that novel genetic elements termed 'integrons' may play an important role in the dissemination of antibiotic resistance genes. Known integrons encode an integrase similar to that encoded by transposon Tn21, and 28 Tn7-negative plasmids (10/60 from 1978-83, 10/43 from 1987-8 and 8/19 from 1992) showed homology with a probe specific for the Tn21 integrase gene. Six plasmids were negative with both probes. It is concluded that Tn7 has played an important role in the dissemination of the gene encoding the type Ia DHFR amongst clinical isolates of E. coli in the Nottingham region of the UK, but that other genetic structures, some of which seem to have an integrase function similar to that of known integrons, may be playing an increasingly significant role.

Trimethoprim resistance in 64 Escherichia coli urinary isolates from five hospitals in Greece was studied. Of the 40 isolates exhibiting transferable high-level resistance (MIC > 1024 mg/L), 21 hybridized with a specific probe for dihydrofolate reductase (DHFR) I, 13 with a probe for DHFR II, and one with a probe for DHFR V. Eleven isolates hybridized with a probe for transposon Tn7. Among the 17 isolates with non-transferable high-level resistance, seven hybridized with the probe for DHFR I, three with the probe for DHFR II, and eight were Tn7-positive. None of the seven isolates with low-level resistance (MIC 4-1024 mg/L) reacted with the probes used. Of the 28 isolates positive for DHFR I, 12 (43%) failed to hybridize with the Tn7 probe. Conversely, three isolates hybridized with the Tn7 probe, but not with the probe for DHFR I. Colony hybridization experiments showed that all but three transconjugants reacted similarly to their respective parent strains. The plasmids coding for trimethoprim-resistant DHFRs were found to differ on the basis of restriction enzyme analysis. These findings suggest that trimethoprim resistance among E. coli urinary isolates in Greece is mediated predominantly by heterogeneous transferable plasmids encoding either DHFR I or DHFR II. The dissociation between DHFR I and Tn7, together with the high incidence of trimethoprim-resistant isolates which did not hybridize with the probes for the common DHFR I or II types, indicates the continued evolution of trimethoprim resistance determinants.

Analysis with a double-disc synergy test (DDST) of clinical isolates of Klebsiella pneumoniae and Escherichia coli during the period October 1988-September 1989 revealed that 24% of the former and 4% of the latter, mainly isolated from urine, possessed an extended-spectrum beta-lactamase. During this period no DDST was positive for isolates of other enterobacterial species. Transfer of ceftazidime resistance was demonstrated from six K. pneumoniae and two E. coli isolates resistant to third generation cephalosporins and aztreonam. Apart from one strain of K. pneumoniae, these strains harboured self-transferable multiresistance plasmids (c. 91 kb) with closely related EcoRI, HindIII, AvaII and PstI restriction patterns. These plasmids encoded an extended-spectrum beta-lactamase that conferred an unusually high level of resistance to ceftazidime and aztreonam (MIC greater than or equal to 64 mg/l). This enzyme had a pI of 8.2 and a substrate profile similar to that of the SHV-5 enzyme isolated initially in Chile, and later in France (CAZ-4). The remaining K. pneumoniae isolate harboured a transmissible multiresistance plasmid (c. 182 kb) that encoded the widely distributed SHV-2 enzyme. The resistance to cefoxitin that was observed in some of these strains was associated with outer membrane protein alterations.

A three day oral antibiotic course was given to 71 elderly bacteriuric subjects with no or only moderate mobility problems. Seven of 17 men (41%) and 34 of 54 women (63%) had strongly positive antibody coated bacteria (ACB) in the urine. Following sensitivity tests and randomization one of the following agents was given: cefadroxil 1 g tid (13 subjects): co-trimoxazole 160/800 mg bd (23 subjects); or norfloxacin 400 mg bd (35 subjects). One week after therapy urines were negative in 13 men (76.5%) and 37 women (68.5%). Patients who were fully mobile and/or were ACB(-) responded better than those with moderate mobility problems or who were ACB(+). At six months, urines were negative in six (40%) of 15 men and 15 (33.3%) of 45 women. Two men and six women of these 21 subjects had a positive urine at one month. Of the three agents tested cefadroxil was less effective in women. The study indicates that a three day course will clear bacteriuria in about 70% of patients at one week, but only about 25% will remain free of infection at six months; these are usually patients with adequate mobility and normal renal function.

Pivmecillinam was given to 44 women with bacteriuria in pregnancy. Treatment was successful in 33 (87%) out of the 38 patients assessed. Thirty women subsequently received at random either a low-dose of pivmecillinam for up to three months or acted as a control group. Further bacteriuric episodes during pregnancy were recorded only in three patients in the control group. Thirty-nine out of the 41 women followed to term delivered healthy babies. One infant was stillborn and another child had a cleft palate. Neither was considered to be related to treatment with pivmecillinam.

Two hundred and eighty four strains of Enterobacteriaceae, responsible for significant bacteriuria, were isolated, over a three month period, in Vellore, India. Sixty-four per cent of these strains were resistant to 10 mg/l of trimethoprim. Moreover, this population was dominated by high level resistance (minimum inhibitory concentration greater than 1000 mg/l) and these accounted for 57.3% of all strains studied. Over half of the resistant strains were able to transfer trimethoprim resistance to standard Escherichia coli strains. However, the high incidence of transferable resistance did not result from the spread of one plasmid type as 58 different plasmid types were identified. These results are in marked contrast to recent findings in Europe where the incidence of high level transferable trimethoprim resistance is falling.

Multi-resistant Enterobacteriaceae isolated mainly from urine specimens from patients at the Department of Urology, Kramáre Hospital, Bratislava, were characterized for resistance phenotype. Seventeen gentamicin-resistant isolates were further studied for the presence of aminoglycoside-modifying enzymes. Five enzymes were detected: AAC(2'), AAC(3)-II, AAC non-characterized, ANT(2") and APH(3')-I. The substrate range of these enzymes was found to correlate with the resistance phenotype in most isolates. In our collection the AAC(3)-II enzyme that inactivates gentamicin, sisomicin, tobramycin and netilmicin was predominant. Predominance of this type of modifying enzyme has been observed also in resistant Gram-negative strains in Belgium, The Netherlands and Chile, in contrast to the United States, Federal Republic of Germany, Switzerland, Greece and Turkey, where ANT(2") has been the most common enzyme.

Norfloxacin produced a reliably bactericidal effect at concentrations from 3 to 90 mg/l against urine pathogens suspended in human urine. These included Enterobacteriaceae, Pseudomonas aeruginosa, Streptococcus faecalis and Staphylococci. Resistant mutants of Staphylococcus aureus were isolated on two occasions (out of 33 experiments). At high concentrations (c. 90 mg/l) the activity was less but this was probably due to the need for a low pH to dissolve the antibiotic. The pH for optimum activity of norfloxacin in urine was 7.5 to 8.0. Human blood had little effect on the bactericidal activity. Compared to other antibiotics, the rate of killing of cultures in urine was second only to gentamicin.

Ninety-nine patients with simple symptomatic urinary tract infections from three general practices were treated with norfloxacin 400 mg bd for 3 days and were followed up for 2 to 4 weeks. Forty of the 99 patients had bacteriuria. Bacteriological eradication of initial pathogen was achieved by 5 to 7 days in 100%. At 2 to 4 weeks three patients had become reinfected but with a different organism. Ninety per cent of patients were symptomatically cured or improved by the fifth to seventh day after therapy started. Drug related adverse experiences were seen in 9% of patients. None necessitated cessation of therapy. No significant biochemical or haematological abnormalities occurred.

One hundred and four patients with complicated urinary tract infections (prolonged severe chronic infections or with complicated postoperative conditions) were treated for ten days with pefloxacin 400 mg bid. Bacteriological eradication of the initial pathogen was achieved in 98% of the patients. After six weeks 93% of the patients were still free of the initial infecting microorganism. Clinical improvement was achieved in 77% of the patients five to seven days after cessation of treatment. The side-effects which were definitely related to pefloxacin occurred in 9% of patients and were mostly of a gastro-intestinal, a neurological, or an allergic nature. No significant biochemical or haematological adverse reactions occurred.

Two thousand seven hundred bacterial strains, isolated during 1981 from urine specimens both from patients in hospital and general practice, were examined for resistance to trimethoprim. The incidence of such resistance was 13% in the hospital isolates and 5.8% in the general practice strains. High level resistance (MIC greater than 1 mg/ml) was present in 75-90% of the resistant strains. With the exception of Staphylococcus epidermidis and Proteus mirabilis, almost all the trimethoprim-resistant strains were also resistant to sulphonamide. In 410 strains from sputum specimens the incidence of resistance was 5.4%. Disc testing for the determination of sensitivity of urine isolates is best carried out using a 5 micrograms disc, while for sputum isolates our present experience indicates that a 1.25 micrograms disc is most appropriate. Our results indicate that the rate of increase in the acquisition of resistance to trimethoprim appears to have slowed down compared with the period 1975-1977, and that almost all this resistance is now high level. The introduction of trimethoprim alone for therapeutic use appears to have little if any effect on the incidence of trimethoprim resistance.

We compared the bactericidal activities of norfloxacin and nine other oral urinary tract antibiotics against 344 Gram-negative bacilli causing bacteriuria in chronically catheterized patients. Norfloxacin killed all organisms at 32 mg/l, 91% of isolates at 2 mg/l, and was the most active antibiotic tested.

One hundred and twenty-two patients with uncomplicated urinary tract infections were treated with either 400 mg bd norfloxacin or 160/800 mg bd cotrimoxazole for 7 days. Follow-up examinations showed norfloxacin to be equally effective as cotrimoxazole in the eradication of bacteriuria and symptoms. Norfloxacin caused fewer and less severe side effects.