Antibiotic sensitivity of the Bacteroides fragilis group in Europe. European Study Group.

Rates of resistance to 12 antibiotics were determined for 1,289 isolates of Bacteroides fragilis group submitted in 1988-1989 by 22 laboratories in 15 European countries. There was no resistance to metronidazole (breakpoint 8 mg/l) and only one isolate was resistant to chloramphenicol (breakpoint 8 mg/l). Resistance was uncommon for imipenem (0.3% at greater than 4 mg/l), amoxicillin/clavulanate (1% at greater than 8 mg/l), cefoxitin (3% at greater than 32 mg/l), mezlocillin (6% at greater than 64 mg/l) and clindamycin (9% at greater than 4 mg/l). Resistance was the rule for ampicillin (93% at greater than 4 mg/l), ciprofloxacin (56% at greater than 4 mg/l) and tetracycline (64% at greater than 4 mg/l). Bacteroides fragilis, the commonest species, was generally the most sensitive: resistance of this organism was uncommon for cefotetan (4% at greater than 32 mg/l) and ceftazidime (12% at greater than 32 mg/l) to which the other species were more often resistant. There were small but significant differences between laboratories and countries for many of the antibiotics. Regionally the most striking differences were for clindamycin where resistance in Bacteroides fragilis was most common in the South and for tetracycline where resistance in Bacteroides fragilis, Bacteroides thetaiotaomicron and Bacteroides uniformis was least common in the North.

Fosfomycin versus ampicillin in the treatment of acute pyelonephritis.

In a controlled open study of 38 patients with acute pyelonephritis, treatment with fosfomycin 8 g b.i.d. was compared to treatment with ampicillin 2 g t.i.d., both for one week. The most common pathogen in both groups was Escherichia coli, susceptible to fosfomycin. However, 17% were resistant to ampicillin. The success rate was 44% in the fosfomycin treated group and 28% in the ampicillin group. The difference was not significant (p greater than 0.20). Peak concentrations of fosfomycin in serum were 395 mg/l and in tissue fluid 85 mg/l. Urine concentrations ranged from 6990 to 24,320 mg/l.

An open, comparative study of sulbactam plus ampicillin vs. cefotaxime as initial therapy for serious soft tissue and bone and joint infections.

In an open, randomized, comparative study, patients with bone, joint, or soft-tissue infections were treated with sulbactam sodium plus ampicillin (sulbactam/ampicillin, 1 g of sulbactam and 2 g of ampicillin three times a day) or with cefotaxime (2 g three times a day) for two weeks as initial therapy. Thirteen patients were entered into the sulbactam/ampicillin group and nine into the cefotaxime group. Two weeks after the end of therapy, clinical cure or improvement was observed in all 13 patients treated with sulbactam/ampicillin and in seven of nine patients treated with cefotaxime. Staphylococcus aureus was not eliminated from one patient in each group. Recurrence of S. aureus infection occurred in two patients in the cefotaxime group within two weeks after the end of therapy. No adverse effects warranting discontinuation of antibiotic therapy were observed. The combination of sulbactam sodium plus ampicillin was at least equivalent to cefotaxime for the initial treatment of acute serious soft tissue and bone and joint infections.

The use of antibiotic serum levels to predict concentrations in tissues.

A review of the literature shows that antibiotic concentrations in tissues and tissue fluids are often quoted as being different in profile to concurrent serum levels. To study the relationship between serum and tissue concentrations we analysed published studies where different experimental models were tested simultaneously. In some models serum levels predicted tissue levels while in others they did not. The factors likely to be responsible for the differences were examined. The most important of these factors was tissue geometry (surface area to volume ratio; SA/V). Serum levels predicted tissue levels in models where the SA/V was high (greater than 60) but not where the SA/V's were low (less than 10); here the antibiotic concentrations were lower and more prolonged than serum levels. These observations can be extrapolated to the clinical situation. In most situations involving prophylaxis or treatment of infections in non-specialised tissues (naturally high SA/V), serum levels will closely reflect levels in extracellular tissue fluid where most bacterial infections are located.

Comparison of skin blisters and implanted cotton threads for the evaluation of antibiotic tissue concentrations.

The skin blister technique and a new technique using subcutaneously implanted cotton threads for the evaluation of antibiotic tissue concentrations were compared. Six volunteers received 2 g ampicillin and, after at least one week, 1 g of cloxacillin intravenously. Samples of serum, blister fluid and thread fluid were analysed by an agar well diffusion technique. The half-life of ampicillin and cloxacillin was prolonged in blister fluid compared to serum and thread fluid. Our conclusions are that the thread technique might reflect the kinetics of antibiotics in extracellular fluid where the penetration mostly depends on serum protein binding of the drug. The blister technique on the other hand might resemble the conditions in different subcutaneously implanted cages, fibrin clots and even abscesses where factors such as time and volume of the extracellular compartments also strongly influence the distribution of antibiotics.

Influence of serum protein binding and mode of administration on penetration of five cephalosporins into subcutaneous tissue fluid in humans.

The penetration of five cephalosporins into interstitial fluid was investigated by a new method employing cotton threads implanted subcutaneously. Penetration after short bolus injections, calculated as area under interstitial fluid level curve divided by area under serum level curve x 100, was 47.0% for cephradine, 30.6% for cefuroxime, 26.7% for cefotaxime, 24.6% for cefoxitin, and 9.6% for cefoperazone. There was an inverse correlation between the degree of penetration and serum protein binding with r = -0.97. The area under interstitial fluid level curves was the same whether the drugs were administered as short bolus injections or short time infusions.

Penetration of ceftazidime into extracellular fluid in patients.

Eight patients were treated with ceftazidime, four of whom had impaired renal function with creatinine clearances below 60 ml/min per 1.73 m2. The levels of ceftazidime in serum and extracellular fluid were determined. Half-life in serum and extracellular fluid was 2.5 and 2.2 h respectively for patients with normal renal function. For patients with impaired renal function the half-life of the drug was 7.9 and 115 h, respectively. The penetration of ceftazidime into extracellular fluid was around 50% in both groups. The levels achieved in both serum and extracellular fluid were satisfactorily high and prolonged, suggesting that a dosage regimen of 1 g twice daily is satisfactory.