Scientific Publications - Work Done by Microbiology Reader Bioscreen C

Antimicrobial Agents and Chemotherapy, 1996, vol. 40, (9), pp. 2099-2105

Antimicrobial activity  of human pancreatic juice  and its interaction with antibiotics

Minelli, E.B., Benini, A., Bassi, C., Abbas, H., Falconi, M., Locatelli, F., de Marco, R. and Pederzoli, R.
 

ABSTRACT

Pancreatic juice (PJ) should be a factor of variability in the antimicrobial activity of antibiotics eliminated by the pancreas during pancreatic infections. We studied its effects on the activity of antimicrobial drugs with different mechanisms of action. Samples of pure PJ were collected from 16 patients with stabilized external pancreatic fistulas. The antimicrobial activity of the juice at different concentrations (from 1.25 to 100%) alone and in combination with mezlocillin, imipenem, ceftriaxone, gentamicin, ofloxacin, and ciprofloxacin was studied by a microbiological method (continuous turbidimetric recording of bacterial growth). The human PJ showed dose-dependent antimicrobial activity that increased directly with the concentration. The activity of the antibiotics at bactericidal concentrations were not modified by the PJ, while the combination with subinhibitory concentrations produced the following variable and different effects: (i) additivity with mezlocillin, ceftriaxone, gentamicin, and ciprofloxacin and autonomy (no interaction) with imipenem and ofloxacin against Providencia rettgeri and (ii) additivity with ceftriaxone, ofloxacin, gentamicin, imipenem, and mezlocillin and autonomy with ciprofloxacin against Escherichia coli. In the presence of PJ, fluoroquinolones showed constant positive effects, while beta-lactams showed more variable antimicrobial activity. Antibiotic concentrations and PJ pharmacodynamics are the main factors determining the final effect of the interaction in vitro. These results may be useful in choosing antibiotics for the treatment of pancreatic infections when they are supplemented with the pharmacokinetic data for each drug.

INTRODUCTION

Pancreatic juice (PJ), like other biological fluids such as serum, saliva, pleural exudates, and amniotic fluid (18, 25, 28, 30), possesses an intrinsic antibacterial activity against several bacteria (2, 17, 25). At the same time, it is one of the main pathways of antibiotic secretion from the pancreas (3, 7, 9, 10). Taken together, these characteristics have a clinical impact, considering that the main mortality factor in acute pancreatitis today is the infection of necrotic tissue (3).

The majority of pathogens responsible for pancreatic infections are gram-negative enteric bacteria. Escherichia coli is the most frequent pathogen. Other bacteria include Klebsiella pneumoniae, Pseudomonas aeruginosa, Proteus mirabilis, Staphylococcus aureus, and Enterococcus (Streptococcus) faecalis (8).

Antimicrobial prophylaxis in severe pancreatitis is still a matter of debate because it proved to be ineffective in early clinical trials (7). To date the management of infected necrosis and pancreatic abscesses has been based primarily on surgical debridement and on the presumptive use of antimicrobial drugs with good activity against the most common microorganisms responsible for pancreatic infections and with proven ability to penetrate into the pancreas even when the disease is in full spate (4, 7, 10). The only clinical experience based on a rational approach to antimicrobial prophylaxis has recently shown a decrease in pancreatic infections from 30 to 12% during severe pancreatitis (21).

Nevertheless, there are still many outstanding problems such as the timing and length of treatment, drug dosages, and resistance. With regard to resistance in particular, the potential interaction between intrinsic antibacterial activity and antibiotic concentration within the PJ may provide useful information for clinical practice. Moreover, the characteristics of pancreatic fluid such as pH . 8 and electrolyte and enzyme concentrations may be important factors regarding the variability of antibacterial drug activity. After reaching the site of infection, the antimicrobial activities of antibiotics could be affected by environmental factors. Therefore, it seems helpful to evaluate whether the effect of this interaction may be clinically relevant.

The aim of the study described here was to investigate the antimicrobial activity of pure human PJ and its effect on the activities of different antibiotics, evaluating the effect, if any, as no interaction (autonomy), additivity, synergism, and/or antagonism. Escherichia coli and Providencia (Proteus) rettgeri were used as the test microorganisms because they are pathogens commonly isolated from patients with pancreatic infections (3, 7).

MATERIALS AND METHODS

Human PJ.

Sixteen patients (10 males and 6 females; mean age, 47.2 years; age range, 20 to 64 years; mean weight, 62.7 kg; weight range, 44.7 to 81 kg) were selected for the study. Informed consent was obtained from all patients.

Human PJ was collected from external surgical drains in sterile containers from patients with pure, stabilized, high-output external pancreatic fistulas (800 to 1,200 ml/24 h) after surgery for severe acute pancreatitis (eight patients) and extirpative procedures for carcinoma of the head of the pancreas. This juice, obtained without pharmacological stimulation, was collected after normalization of pancreatic exocrine function and may be considered similar to that produced by the pancreas under normal conditions. Patients with pancreatic cancer had never received antibiotic treatment, while for patients in the acute pancreatitis group, all antibiotic treatments were withdrawn at least 48 h before sampling. Residual antibiotic contamination was checked for in all samples.

Samples.

After PJ sample collection the pH was measured and the PJ sample was subdivided into small aliquots (2 ml each) and frozen at -80°C until assay (at most, within 3 weeks of collection). The antimicrobial activity of PJ from each subject was assayed individually.

Antimicrobial activity assay.

The antimicrobial activities of PJ, antimicrobial drugs, and their combinations were simultaneously evaluated by continuous monitoring of bacterial growth with an automated analyzer (Bioscreen; Labsystems, Helsinki, Finland). The absorbance was measured at 10-min intervals by a turbidimetric method with vertical light photometry (at 580 nm) and was recorded with a computerized system with special software for the analysis of growth curves. The results were analyzed by examining the growth curves and evaluating the area under the growth curve (AUGC). The AUGC was calculated automatically by the software. Pure PJ, antibiotics at different concentrations, medium (Mueller-Hinton broth [MHB]; Difco), and microbial suspensions were automatically dispensed in that order at the beginning of the experiments on the basis of our previously prepared dispensing protocol. The different volumes of the components used were dispensed into a maximum of 200 test cuvettes arranged in a honeycomb fashion in microtiter plates with flat bottoms (400-µl final volume per well). Wells containing different volumes of sterile saline, phosphate buffer (pH 8.3), MHB, and a microorganism served as controls for antibiotics and PJ alone and in combination in order to reproduce the same volume and pH conditions used for the samples. PJ was replaced in the control wells with the same volume of phosphate buffer with an alkaline pH that was the same as that of each juice sample (generally pH 8.0 to 8.5) or by MHB at pH 8.3; MHB was used at pH 7.0 in the other wells. Wells with MHB alone served as absolute controls (blanks). E. coli ATCC 25922 and the clinical isolate P. rettgeri (Sanelli) were used as test microorganisms. The inoculum (from an overnight culture in MHB diluted in MHB) was 10 µl per well, corresponding to 10⁷ CFU/ml. The growth chamber was maintained at 37°C with agitation between absorbance measurements for 20 to 24 hours. Each sample (antibiotic, PJ, microorganism alone, and combination) was assayed three or more times.

Preliminary experiments were performed to compare the effects of PJ on the MICs of different antibiotics by the broth microdilution method (MHB), colony counting, and the agar well diffusion method (petri dishes with Mueller-Hinton Agar [MHA]) (4). Serial twofold microdilutions (100 µl) of mezlocillin, imipenem, gentamicin, and ciprofloxacin were prepared (in parallel) in MHB and pure PJ, respectively (96-well microtiter plates; Greiner, Nurtingen, Germany). A total of 10 µl of E. coli and P. rettgeri from an overnight culture diluted 1:100 was added to each well. After 8 and 20 h of incubation at 37°C, the MIC endpoints were determined to be the lowest concentration at which there was no visible turbidity in the wells (with a magnifying mirror; PBI, Milan, Italy). The MICs for E. coli in MHB and PJ were as follows: mezlocillin, 4 and 2 mg/liter, respectively; gentamicin, 4 and 2 mg/liter, respectively; imipenem, 4 and 2 mg/ liter, respectively; and ciprofloxacin, 0.5 and 0.062 mg/liter, respectively. The MICs for P. rettgeri in MHB and PJ were as follows: mezlocillin, 2.5 and 0.62 mg/liter, respectively; gentamicin, 40 and 10.0 mg/liter, respectively; and cipro- floxacin, 0.5 and 0.25 mg/liter, respectively. The determination of the interaction in solid medium appeared to be less sensitive than that by the microtiter method, although we were able to detect the positive effects of PJ on antibiotic activity. The MIC tests (micro- and macrodilution methods) based on all-or-none responses (i.e., the presence or absence of significant growth) and assessed at only one time point as well as colony counting appeared to be of little use for quantifying the effects of the combination. These tests cannot be used to define the effects of antimicrobial concentrations other than the MIC or MBC or to develop the graded dose-response curves necessary to quantify the antimicrobial action. Therefore, we decided to use the measurement of bacterial growth over time. The conventional pour plate assay was performed in parallel with the turbidimetric method for most of the tests performed over 20 h and for tests performed at some other observation times. Aliquots from each sample were collected at 0, 5, and 20 h. Bacterial counts (CFU per milliliter) were determined by serially diluting the sample 10-fold in cold sterile saline and inoculating it (in duplicate) onto drug-free MHA with subsequent incubation at 37°C for 18 to 24 h. The relationship between the logarithm of the number of viable bacteria determined by plate counting and the absorbance and AUGC was linear for controls and cultures treated with antibiotic, PJ, or both. Linear regression analysis of the data (from preliminary experiments) yielded the following correlation coefficients between colony counts and absorbance for E. coli: control, r = 0.906 (P < 0.013); for PJ, r = 0.667 (P = 0.05); for the six antibiotics, range from r = 0.795 (P ≤ 0.001) for mezlocillin to r = 0.897 (P ≤ 0.001) for imipenem; and for the antibiotic-PJ combinations, range from r = 0.995 (P = 0.064) for the imipenem plus PJ to r = 0.994 (P = 0.04) for ciprofloxacin plus PJ. The results obtained with P. rettgeri were as follows: control, r = 0.929 (P = 0.007); PJ, r = 0.704 (P = 0.003); all antibiotics, r = 0.838 (P ≤ 0.001); and antibiotic-PJ combinations, r = 0.894 (P ≤ 0.001). The relationship between colony counts and AUGC yielded the following correlation coefficients for E. coli: control, r = 0.915 (P = 0.01); PJ, r = 0.910 (P ≤ 0.001); all antibiotics, r = 0.78 (P ≤ 0.001); antibiotic-PJ combinations, r = 0.737 (P ≤ 0.001). The values for P. rettgeri were as follows: control, r = 0.902 (P = 0.014); PJ, r = 0.88 (P ≤ 0.001); all antibiotics, r = 0.816 (P ≤ 0.001); and antibiotic-PJ combinations, r = 0.897 (P ≤ 0.001) (single data are available upon request). The parameters determined by the turbidimetric method correlated well with those obtained by plate counting over a wide range of bacterial concentrations. These results are in accord with those reported by other investigators (15, 26). Incubation with PJ and the subinhibitory concentrations of antibiotics induced no appreciable morphological changes (determined by Gram staining). Only mezlocillin at one-eighth to one-fourth the MIC induced elongation and the formation of filaments, although this occurred only in a small proportion of E. coli cultures.

The slope and the shape of the growth curves recorded continuously also yielded qualitative information about the inhibitory effects of different concentrations of antibiotics and PJ, both alone and in combination, on bacterial growth (1, 31).

Drugs.

The antimicrobial agents used in the study were mezlocillin and ciprofloxacin (Bayer, Milan, Italy), ofloxacin (Glaxo S.p.A., Verona, Italy), imipenem (Merck Sharp & Dohme, Rome, Italy), ceftriaxone (Roche, Milan, Italy), and gentamicin (Schering-Plough, Milan, Italy).

Stock solutions of the different antibiotics were prepared in sterile distilled water and were stored in small volumes at -80°C. For each experiment dilutions were made in MHB. Standard concentrations of imipenem were prepared fresh each day in morpholinethansulfonate acid (MES)–ethylene glycol (1:1 mixture) (19) and were then diluted with MHB and used.

Antibiotics were used at sub-MICs (range, 1/2 to 1/32 the MIC), chosen to delay bacterial growth for 4 to 6 h. After this period the microorganisms showed normal growth rates with few morphological changes. In some experiments inhibitory concentrations were used and bacterial growth was inhibited for a longer period (up to 24 h). The concentrations used for each antibiotic are reported in the Results (see Table 1). The MICs, determined by Bioscreen (19), were 0.5 mg/liter for gentamicin, 0.4 mg/liter for ofloxacin, 1.0 mg/liter for mezlocillin, 0.02 mg/liter for ceftriaxone, 1.5 mg/liter for imipenem, and 0.008 mg/liter for ciprofloxacin for P. rettgeri and 0.5 mg/liter, for gentamicin, 0.4 mg/liter for ofloxacin, 2 mg/liter for mezlocillin, 0.02 mg/liter for ceftriaxone, 0.2 mg/liter for imipenem, and 0.004 mg/liter for ciprofloxacin for E. coli ATCC 25922.

The stability of the antibiotics was evaluated by the agar well diffusion method (4) with Bacillus subtilis ATCC 6633 (spore suspension; Difco) in Iso-Sensitest agar (0.02 ml/100 ml of agar; Oxoid). Different drugs prepared in H2O or buffer, broth (MHB), and PJ (final concentrations, 10 mg/liter) were maintained at 4 and 37°C for 24 h; aliquots were collected at time zero and 0.5, 5, and 24 h after incubation and were then assayed (50 ml per well). After overnight incubation at 37°C, the inhibition zones were measured.

PJ.

PJ was tested at different progressive dilutions from 100 to 1.25% to assess its intrinsic antimicrobial activity. On the basis of these results PJ was used at 40 and 10% concentrations in the experiments in combination with antibiotics.

Interaction of PJ and antibiotics.

The interaction of PJ and the antibiotics was determined by using sub-MICs of the antibiotics in the presence of 40 and 10% PJ. The concentrations of both components were selected to produce a delay in bacterial growth of 4 to 6 h in order to analyze the qualitative and quantitative effects. The 10% concentration of PJ allowed us to study the effect of the PJ composition without appreciable antimicrobial activity.

In one set of experiments the effect of antimicrobial combinations on bacterial growth was also determined by using fixed concentrations of PJ (40 and 10%) against increasing drug concentrations, and fixed subinhibitory concentrations of each antibiotic were tested against increasing PJ concentrations (13, 16).

Evaluation of results.

The effect of antimicrobial activity was expressed as (i) AUGC from 0 to 20 h (AUGC_0–20; absorbance x time [in hours]), i.e., the difference (ΔAUGC) between the AUGC of each component alone and in combination versus the AUGC of the controls (1) or as percent reduction (26), and (ii) hours of inhibition, i.e., the delay in bacterial growth in reaching 0.6 absorbance units in the presence of the components alone and in combination compared with that for the controls (5) plus the duration of inhibition calculated as the bacterial growth reduction of the sample in comparison with that of the control growth (100%) for the remaining hours of incubation. The inhibition value was obtained by the following equation:

inhibition (hours) = h_{Abs 0.6} + [(h_end - h_{Abs 0.6}) x (End Abs S/End Abs C)]

where h_{Abs 0.6} is the time to reach an absorbance of 0.6, h_end is the duration of the experiment, End Abs S is the absorbance of the sample at the end of the experiment, and End Abs C is the absorbance of the control at the end of the experiment.

Interaction was described according to a general definition of antimicrobial combinations (13, 16) as follows: autonomy or indifference, when two components did not interact with one another or the effect corresponded to the activity of the more active drug alone; additivity, the sum of the separate effects of PJ and drugs; synergism, the combined effect of the drug and PJ was greater than the expected results (additivity); and antagonism, the combined effect of the drug and PJ was less than their independent effects.

In order to study the effect of the interaction between an antibiotic and PJ, the data were analyzed by two-way analysis of variance (ANOVA) (12) by using ΔAUGC as the variable response and PJ and the subinhibitory concentrations of each drug as variability factors. The model used in this analysis was

---------- 8< ----------

DISCUSSION

Our data confirm that pure human PJ exerts intrinsic antimicrobial activity against gram-negative bacteria. This activity is concentration dependent, is not altered by heat treatment (is not due to serum complement factors), and may presumably be ascribed to substances with different molecular weights (above 10,000). Furthermore, our data demonstrate that PJ cooperates with bactericidal drugs such as beta-lactams, fluoroquinolones, and aminoglycosides to produce an additive antimicrobial effect, despite a wide intersubject variability in antimicrobial activity.

All PJ samples were inhibitory at high concentrations (80 to 100%), with the effect being bactericidal (no growth for 24 h) in some cases and bacteriostatic in others. Patient characteristics, PJ pH, and underlying pancreatic disease were not found to be relevant variability factors with regard to the final antimicrobial activity of PJ. Therefore, different, additional factors must be responsible for the interindividual differences in the antimicrobial activity of PJ. Our results correspond to those obtained in the dog by Rubinstein et al. (25) concerning heat stability, dialyzability, and antimicrobial activity, whereas the molecular weights of substances with antibacterial activity seem to be higher than those of substances from dog PJ, estimated to be less than 4,000. Since PJ presents characteristics such as an alkaline pH, high concentrations of electrolytes, and the presence of proteins, it is interesting that the combination with different antibiotics resulted in a positive antimicrobial effect against E. coli and P. rettgeri. In a preliminary study using an MIC method we observed a cooperative effect between PJ and mezlocillin (2). Now, we can define as additive the effect between high PJ concentrations and subinhibitory concentrations of antibiotics (range from 1/2 to 1/40 the MIC). Therefore, various antibiotics were effective in the presence of PJ at a wide range of concentrations. Their behaviors did not appear to be modified by PJ, as shown by the analysis of the shapes of growth curves and dose-response curves. The drugs seemed to be stable under our experimental conditions as well as in different systems (11, 16, 27, 32).

PJ operates jointly with antibiotics with different mechanisms of action; the effect with fluoroquinolones and aminoglycosides appears to be fairly constant, while the results obtained with cell wall inhibitors are more variable and unpredictable. No macroscopic morphological changes (determined by Gramstaining) were observed in bacteria exposed to human PJ alone; protein synthesis inhibition appeared to be the main mechanism of action of PJ (25). Fluoroquinolones and aminoglycosides exhibit optimal activity at alkaline pH (16), and their main target is the inhibition of DNA and protein synthesis at different levels (14, 27). These factors seem to favor positive cooperation between PJ and antibiotics which interfere with DNA and protein synthesis. Chlorquinaldol, clioquinol, chloramphenicol, and trimethoprim-sulfamethoxazole showed synergism with duodenopancreatic secretions (17), while substances interfering with the phospholipids of the cytoplasmic membrane (colistin) or lipoprotein (bicozamycin) (29) presented autonomy or antagonism.

The substantial ability of PJ to cooperate with antibiotics was confirmed by a different investigation and evaluation approach. The continuous monitoring of bacterial growth makes it possible to evaluate the kinetics and the final effect of the drug-pancreatic juice interaction as well as to analyze the effects of graded concentrations and of the timing and duration of antimicrobial activity for each component of the combination (13). Since the antibiotics were added at the beginning of the experiment, before logarithmic growth, the inhibition of bacterial growth was observable as soon as the bacterial growth started. A number of limitations of other methods, such as MIC determination by the checkerboard technique, the fractional MIC technique (the fractional inhibitory concentration), and the rate of bacterial killing or serum bactericidal tests, are partly eliminated. This technique combined with computerized analysis of the data is probably a reasonable alternative for an in vitro evaluation of the interaction which is closer to clinical conditions.

Our results indicated that the variable and different effects are concentration dependent and specific for each drug; these conditions can occur in vivo, since the antibiotic concentrations change constantly at the site of infection according to pharmacokinetic characteristics and therapeutic regimen. The attainment of a determined concentration of antibiotic at the site of infection is not enough to ensure that the patient will be cured of the infection; different factors must be evaluated and integrated with the in vivo results to interpret antimicrobial concentrations in the context of therapeutic efficacy. PJ cooperates with host defense mechanisms against bacterial infections. If the positive effect observed in vitro also occurs in vivo during the pancreatic infection, we can postulate that the antimicrobial activities of antibiotics present in the pancreas can be enhanced and prolonged by PJ. This means that it may be possible to adopt therapeutic regimens with antimicrobial drugs capable of penetrating the pancreas (4, 6, 9, 14, 21–23) at therapeutic concentrations without high doses and at normal administration intervals.

In conclusion, the present study demonstrates the efficacy of antibiotics in the presence of PJ in the normal (nonimmunocompromised) host and the antibiotic-enhancing activity of pure human PJ. The present data provide further evidence that antimicrobial activity in vivo cannot be predicted merely on the basis of levels of antibiotics in the pancreas and the in vitro susceptibilities of microorganisms. Therefore, the processing of pharmacokinetic data in relation to PJ excretion, drug antimicrobial activity (MIC and antibacterial spectrum), and interaction with PJ should be considered when choosing antimicrobial drugs, dosage regimen, and/or administration modality in the course of necrotizing pancreatitis.

ACKNOWLEDGMENT

This work was supported by a grant from the Italian Ministry of the University and Scientific Research (MURST 40%).

 REFERENCES

1. Barriere, S. L., E. Ely, J. E. Kapusnik, and J. G. Gambertoglio. 1985. Analysis of a new method for assessing activity of combination of antimicrobials: area under the bactericidal activity curve. J. Antimicrob. Chemother. 16:49–59.

2. Bassi, C., R. Fontana, S. Vesentini, G. Cavallini, L. Marchiori, M. Falconi, S. Corra`, and P. Pederzoli. 1991. Antibacterial and mezlocillin-enhancing activity of pure human pancreatic fluid. Int. J. Pancreatol. 10:293–297.

3. Bassi, C., and P. Pederzoli. 1994. Antibiotics in necrotizing pancreatitis, p. 93–98. In E. L. Bradley III (ed.), Acute pancreatitis: diagnosis and therapy. Raven Press, New York.

4. Bassi, C., P. Pederzoli, S. Vesentini, M. Falconi, A. Bonora, H. Abbas, A. Benini, and E. M. Bertazzoni. 1994. Behavior of antibiotics during human necrotizing pancreatitis. Antimicrob. Agents Chemother. 38:830–836.

5. Bertazzoni, E. M., and G. Panzetta. 1986. Antibacterial interaction of cefuroxime and serum from uraemic patients. J. Antimicrob. Chemother. 18: 365–373.

6. Bertazzoni Minelli, E. Pharmacokinetics of antibiotics in the pancreas, p. 301–312. In P. Pederzoli, G. Cavallini, and C. Bassi (ed.), Facing the pancreatic dilemma. Springer-Verlag, Berlin.

7. Bradley, E. L., III. 1989. Antibiotics in acute pancreatitis. Current status and future directions. Am. J. Surg. 158:472–477.

8. Bradley, E. L., III. 1994. Overview of clinical experiences, p. 313–322. In P. Pederzoli, G. Cavallini, and C. Bassi (ed.), Facing the pancreatic dilemma. Springer-Verlag, Berlin.

9. Bu¨chler, M., P. Malfertheiner, H. Friess, and H. Beger. 1989. The penetration of antibiotics into human pancreas. Infection 1:20–25.

10. Bu¨chler, M., P. Malfertheiner, H. Friess, R. Isenmann, E. Vanek, H. Grimm, P. Schlegel, T. Friess, and H. G. Beger. 1992. Human pancreatic tissue concentration of bactericidal antibiotics. Gastroenterology 103:1902–1908.

11. Clissold, P. S., P. A. Todd, and D. M. Campoli-Richars. 1987. Imipenem/ cilastatin. A review of its antibacterial activity, pharmacokinetic properties and therapeutic efficacy. Drugs 33:183–241.

12. Dunn, O. J., and V. A. Clark. 1987. Applied statistics analysis of variance and regression, 2nd ed. John Wiley & Sons, Inc., New York.

13. Eliopoulos, E., and R. C. Moellering. 1991. Antimicrobial combinations, p. 432–492. In V. Lorian (ed.), Antibiotics in laboratory medicine, 3rd ed. The Williams & Wilkins Co., Baltimore.

14. Kucers, A., and N. M. Bennet. 1989. The use of antibiotics, 4th ed. Heinemann Medical Books, Oxford.

15. Li, R. C., D. E. Nix, and J. J. Shentag. 1993. New turbidimetric assay for quantitation of viable bacterial densities. Antimicrob. Agents Chemother. 37:371–374.

16. Lorian, V. (ed.). 1993. Antibiotics in laboratory medicine, 3rd ed. The Williams & Wilkins Co., Baltimore.

17. Mett, H., K. Gyr, O. Zak, and K. Vosbeck. 1984. Duodeno-pancreatic secretions enhance bactericidal activity of antimicrobial drugs. Antimicrob. Agents Chemother. 26:35–38.

18. Miglioli, P. A., E. Bertazzoni Minelli, and T. Berti. 1987. Enhancement of antibiotic antibacterial activity by experimental pleural exudates. Drugs Exp. Clin. Res. 13:161–166.

19. Norrby, S. R., K. Alestig, F. Ferber, J. L. Huber, K. H. Jones, F. M. Kahan, A. P. Meisinger, and J. D. Rogers. 1983. Pharmacokinetics and tolerance of N-formimidoyl thienamycin (MK 0787)) in humans. Antimicrob. Agents Chemother. 23:293–299.

20. Olsson-Liljequist, B., and B. M. Hoffman. 1991. In-vitro activity of clarithromycin combined with its 14-hydroxymetabolite A-62671 against Haemophilus influenzae. J. Antimicrob. Chemother. 27(Suppl. A):11–17.

21. Pederzoli, P., C. Bassi, S. Vesentini, and A. Campedelli. 1993. Antibiotic prophylaxis of septic complications in acute necrotizing pancreatitis. A randomized multicenter clinical trial with imipenem. Surg. Gynecol. Obstet. 176:480–483.

22. Pederzoli, P., M. Falconi, C. Bassi, S. Vesentini, F. Orcalli, F. Scaglione, A. Messori, and N. Martini. 1987. Ciprofloxacin penetration in pancreatic juice. J. Antimicrob. Chemother. 33:397–401.

23. Pederzoli, P., F. Orcalli, M. Falconi, L. Bozzini, and N. Martini. 1986. Penetration of mezlocillin into pancreatic juice. J. Antimicrob. Chemother. 17:397–398.

24. Pritchard, M. N., and C. Shipman, Jr. 1990. A three-dimensional model to analyze drug-drug interaction. Antiviral Res. 14:181–206.

25. Rubinstein, E., Z. Mark, J. Haspel, G. Ben-ari, Z. Dreznik, D. Mirelman, and A. Tadmor. 1985. Antibacterial activity of the pancreatic fluid. Gastroenterology 88:927–932.

26. Skytta¨, E., and T. Mattila-Sandholm. 1991. A quantitative method for assessing bacteriocins and other food antimicrobials by automated turbidometry. J. Microbiol. Methods 14:77–88.

27. Smith, J. T., and C. S. Lewin. 1988. Chemistry and mechanisms of action of the quinolone antibacterials, p. 23–82. In V. T. Andriole (ed.), The quinolones. Academic Press, London.

28. Stern, C. M. M. 1981. Bactericidal glycopeptide in human amniotic fluid. J. Antimicrob. Chemother. 8:1–4.

29. Tanaka, N., M. Iseki, T. Miyoshi, H. Aoki, and H. Imanaka. 1976. Mechanism of action of bicyclomycin. J. Antibiot. 29:155–168.

30. Taylor, P. W. 1983. Bactericidal and bacteriolytic activity of serum against gram-negative bacteria. Microbiol. Rev. 47:46–83.

31. Tisdale, J. E., M. T. Pasko, and J. M. Mylotte. 1989. Antipseudomonal activity of simulated infusions of gentamicin alone or with piperacillin assessed by serum bactericidal rate and area under the killing curve. Antimicrob. Agents Chemother. 33:1500–1505.

32. Todd, P. A., and D. Faulds. 1991. Ofloxacin. A reappraisal of its antimicrobial activity, pharmacology and therapeutic use. Drugs 42:825–876.

(Full Text online - PDF)

Back to Automation in Microbiology main page

© 2005 Transgalactic Ltd (manufacturer of Bioscreen C software) | Privacy Statement | P.O. Box 1393, 00101 Helsinki, Finland, phone: +358 9 85172920, fax: +358 9 8749481, e-mail: microbiology@bionewsonline.com
 

Last modified: May 25, 2005