Prevention of catheter-associated Gram-negative bacilluria with norfloxacin by selective decontamination of the bowel and high urinary concentration

Transcription

1 Journal of Antimicrobial Chemotherapy (1989) 23, Prevention of catheter-associated Gram-negative bacilluria with norfloxacin by selective decontamination of the bowel and high urinary concentration E. J. Vonaard*, H. A. L. Clasener*, J. V. Zambon*, H. J. M. Joosten' and A. J. A. van Griethuysen* Departments of'pharmacy, b Microbiology and'surgery, Canisius-Wilhelmina Hospital, St Annastraat 289, 6500 GS Ngmegen, The Netherlands Oral norfloxacin prevented Gram-negative bacilluria in female patients with hip fractures, who needed medium-term transurethral cathcterization. This was shown in a placebo-controlled double-blind study of 34 patients. Seventeen of these received a suspension containing 200 mg norfloxacin and 500 mg amphotericin B, twice daily. In the placebo group, six cases of Gram-negative bacilluria had occurred by day 7, as compared with no cases during a median time of catheterization of 23 days in the group on medication. Bacteriuria, either by Gram-positive cocci or by Gram-negative bacilli, was observed in 50% of patients on placebo by day 7; in the treatment group this was the case by day 17 (./>< 0-001). Subsequent bacteriuria with Gram-positive cocci was eliminated by nitrofurantoin (50 mg qid) within four days. Norfloxacin is very suitable for the prevention of Gram-negative bacilluria, because it decontaminates Gram-negative bacilli from the bowel, reaches high concentrations in urine and rarely produces resistant variants. Introduction The indwelling urinary catheter is the leading cause of nosocomial urinary tract infection and the most common predisposing factor in fatal Gram-negative sepsis in hospitals (Schaeffer, 1986; Kunin, 1987). Systemic antimicrobial prophylaxis reduces the rate of infection in catheterized patients, but this favourable effect is usually limited to a few days because of the emergence of resistant strains. Therefore, systemic antimicrobial prophylaxis is generally rejected as a method for the prevention of urinary tract infections in patients with median-term or long-term catheterization (Turck & Stamm, 1981; Schaeffer, 1986; Kunin, 1987). The major pathway of infection of the urinary tract in patients with properly managed closed drainage systems is the migration of bacteria in the space between the catheter and the urethral mucosa (Schaeffer, 1986; Kunin, 1987). Therefore, catheter-associated bacteriuria might be prevented by reduction or elimination of colonization of the periurethral mucous membrane by potentially pathogenic micro-organisms. Application of poly-antimicrobial ointment to the urethral meatus, however, was of marginal or no benefit (Kunin, 1987; Stickler & Chawla, 1987). Colonization of the periurethral mucous membrane is derived from the faecal flora (Winberg et al., 1973; Daifuku & Stamm, 1984; Kunin & Steele, 1985; Stamey, 1987). Therefore, decontamination of the digestive tract from potentially pathogenic micro-organisms might be useful in the prevention of catheter-associated bacteriuria /89/ $02.00/ The British Society for Antimicrobial Chemotherapy

2 916 E. J. Vollaard el al Norfloxacin is an effective agent for eliminating Gram-negative bacilli from the intestinal tract (Clasener, Vollaard & van Saene, 1987). Furthermore, it reaches urinary levels far above the MIC of most uropathogens (Wolfson & Hooper, 1988). In this way it might prevent bacteriuria originating from a contaminated drainage system, and clear the urine of bacteria that were present beforehand, or that entered while the catheter was in situ. Enterobacteriaceae rarely develop resistance to norfloxacin (Kresken & Wiedemann, 1988; Wolfson & Hooper, 1988). Therefore, it seemed timely to reconsider the possibility of preventing Gram-negative bacilluria in patients needing urinary catheters. As a study population we chose female patients admitted to the surgical department with hip fractures. These patients are old and many of them need catheterization for at least two weeks. Female sex, old age, duration of catheterization and orthopaedic surgery are important risk factors in the acquisition of catheter-associated bacteriuria (Shapiro el al., 1984; Kunin, 1987). In an unpublished pilot study in patients with indwelling catheters in nursing homes, Gram-negative bacilluria was prevented by norfloxacin, but candiduria developed in four of seven patients within a month; therefore, oral amphotericin B was also given to patients in the study group, to eliminate Candida albicans from the gut. Patients Patients and methods All female patients admitted to the surgical department with hip fractures, who received a transurethral catheter, and who had no history of allergy to the study drugs, were enrolled in the trial. Most patients received the catheter on the day before operation. Peri-operative prophylaxis for all patients consisted of two doses of cefazolin 1 g iv. A closed catheter drainage system was used. Patients were excluded from analysis when the catheter was removed within six days, and when they received antibiotics for other indications than for urinary tract infection. Permission for the performance of this study was obtained from the Ethical Committee of the Canisius-Wilhelmina Hospital. Trial procedure Placebo suspension or active medication, according to the randomized list of patients in the pharmacy, was started immediately following insertion of the catheter. The active suspension contained 200 mg norfloxacin and 500 mg amphotericin B in 10 ml. The suspension was prepared by the hospital pharmacy from norfloxacin tablets (Noroxin, MSD) and amphotericin B suspension (Fungizone, Squibb). The placebo was a suspension of identical appearance that had no antimicrobial activity, except for the preservative agent methylhydroxybenzoate (01%) which was present in both suspensions. Dosage was 10 ml suspension twice daily by mouth. Therapy Patients, who developed bacteriuria, despite norfloxacin prophylaxis, were treated with nitrofurantoin 50 mg orally four times daily, unless the infecting micro-organism was

3 Norfloxadn SDD in prophylaxis of badunria 917 insensitive to this agent, or the patient was allergic to nitrofurantoin. In that case cephradine 500 mg orally four times daily was used. At the same time, prophylaxis with norfloxacin and amphotericin B was continued. Control patients who developed bacteriuria, were treated with norfloxacin and amphotericin B, orally. Urine cultures Urine was taken by antiseptic catheter puncture, immediately after introduction of the catheter, then three times a week, and immediately before removal of the catheter. One-/d samples of urine were inoculated on selective solid media (see faeces cultures). All urine samples in which bacteria could be detected (detection limit: 10 3 cfu/ml) were considered to be positive. Identities and sensitivities of bacteria were determined by standard methods (Lennette et al., 1985). Bacteria were considered resistant to norfloxacin if the MIC exceeded 4 mg/1. Two successive positive cultures were required for the diagnosis of bacteriuria; if only one sample was positive, it was considered to be contaminated. Patients were excluded from analysis if the first two urine samples were positive. Faeces cultures Faecal concentrations of uropathogenic micro-organisms were determined as often as faeces could be obtained. Serial 1/10 dilutions of faeces were made in Thioglycollate Medium (BBL). One-^1 volumes of each dilution were inoculated on solid media selective, respectively, for Enterobacteriaceae (Eosin Methylene-blue Lactose Sucrose Agar, Merck); for enterococci (5% sheep blood in Blood Agar Base (Oxoid) with nalidixic acid 50 mg/1); for staphylococci (Mannitol Salt Agar, BioMerieux); and for yeasts (Sabouraud Dextrose Agar (Gibco) with chloramphenicol 125 mg/1). The solid selective media were also inoculated with 100-//1 volumes of the first 1/10 dilutions of faeces, lowering the limit of detection to loocfu/g of faeces. Micro-organisms isolated were identified and their antibiotic sensitivities were determined by standard methods. Statistical analysis Age and number of days with catheter in place were compared by Wilcoxon's test. Distributions of survival, defined as number of days between day of catheterization and the first day of bacteriuria, were compared by the SAS procedure LIFETEST (SAS Institute Inc., 1985). Results Between 1 June 1986 and 1 July 1987, 58 patients were enrolled in the study. Twentyfour patients were excluded from analysis (Table I). Seventeen patients were evaluable in each group. The median age of the patients on norfloxacin medication was 85 years (range 76-96) (Table II). The median age of the control patients was 85 years (range 74-90) (Table III). The median number of days of catheterization was slightly lower in the placebo-group: 14 (6-68) versus 23 (6-52) (P=00T).

4 918 E. J. VoUaanl et al Table I. Patients excluded from analysis Active Reason for exclusion Placebo medication Non-compliance Catheter removed within 6 days Antibiotic use Study-medication not started First two cultures positive Total number of patients Urine cultures Contaminated samples (positive samples preceded and succeeded by negative samples) occurred infrequently. In three patients on active medication Gram-positive flora was isolated ( cfu/ml) and in one sample of one placebo-patient C. albicans was found (10 3 cfu/ml). In some patients who acquired bacteriuria, the first positive sample contained 10 3 or 10* cfu/ml. All other positive samples contained at least 10 5 cfu/ml. The results of the study are shown in Tables II and III, and Figure 1. Fifty per cent of patients in the placebo-group were bacteriuric by day 7, compared with day 17 in the norfloxacin group. The difference was highly significant (/ < 0-001). Eleven patients in the medication group developed bacteriuria caused by Staphylococcus epidermidis, sensitive to nitrofurantoin. In five patients the catheter was removed before therapy could be started. Five patients were treated with nitrofurantoin, which sterilized the urine within four days. One patient who was Table II. Evaluable patients on norfloxacin-amphotericin B prophylaxis Age Catheter removal on day First day of bacteriuria Micro-organism

5 Norfloxadn SDD in prophylaxis of badhnria 919 Table m. Evaluable patients on placebo Age Catheter removal on day First day of bacteriuria Micro-organism Ent. faecalis Proteus vulgaris S. aureus S. aureus Pr. morgani. Klebsiella oxytoca S. cpidermidis Escherichia coli Ps. aeruginosa Ent. faecalis Ent. faecalis Ps. aeruginosa Ps. aeruginosa believed to be hypersensitive to nitrofurantoin was treated successfully with cephradine 500 mg orally four times daily. Of the 17 patients in the placebo group, 12 acquired bacteriuria. In six patients bacteriuria was caused by Gram-negative bacilli, and in six by Gram-positive cocci (Table III). In three patients the catheter was removed before therapy could be started. Changing to active medication (norfloxadn and amphotericin B) in the remaining nine patients resulted in sterilization of the urine in seven. In the two other patients, Pseudomonas aeruginosa and S. aureus were not eliminated by this treatment Doyi culhelw In place Fignre 1. Days of catheterization before bacteriuria occurred in the treatment and placebo groups.

6 920 E. J. Vollaard et al Faeces cultures Faeces could not be obtained from all patients. Most patients had a very low frequency of defaecation. In the 11 patients on placebo from whom faeces were obtained before norfloxacin and amphotericin B were started, the median concentration of Gram-negative bacilli was 10 7 cfu/g faeces (range 10*-10'). In six of these 11 patients, C. albicans was present in the faeces, in a median concentration of 10* cfu/g faeces (range ). In the 14 patients started on active medication, from whom faecal samples were obtained, faecal concentrations of Gram-negative bacilli and C. albicans were suppressed to below the detection limit of 10 2 cfu/g of faeces. Three of these patients had one positive faecal sample for Gram-negative bacilli, preceded and succeeded by negative cultures. These transient Gram-negative bacilli (Acinetobacter anitrains 10 6 /g and 10 3 /g, and Ps. cepacia 10*/g) were sensitive to norfloxacin. Enterococcus faecalis was suppressed to a variable extent. Discussion Catheter-associated infections in the placebo-group were acquired with an average of about 7% a day during the first ten days; this is within the expected limits of 4-7-5% (Kunin, 1987). This indicates a good standard of closed-system catheter- care, certainly if the high risk of patients in this study is taken into account. Thus, the results in the group on medication are an important improvement. In this study, norfloxacin decontaminated the bowel of our patients from Gram-negative bacilli and prevented Gram-negative bacteriuria in 17 patients during 478 days of catheterization. We believe that norfloxacin is better than other drugs previously used to suppress catheter-associated bacteriuria, because it eliminates uropathogenic Gram-negative bacilli from the intestine and rarely produces resistant variants. In this way, it eliminates the most important source of bacteriuria. (Winberg et al., 1973; Daifuku & Stamm, 1984). Therefore, the number of Gram-negative bacilli which reach the bladder will be strongly reduced. At the same time, high concentrations of norfloxacin in the urine will protect against Gram-negative bacteriuria from other sources. Norfloxacin does not prevent bacteriuria caused by 5. epidermidis. However, 5. epidermidis was eliminated from urine with oral nitrofurantoin or cephradine, with the catheter in situ, within four days. Following sterilization of the urine the catheter was replaced, because strongly attaches itself to the catheter surface and will not be removed from this place by nitrofurantoin (Kunin & Steele, 1985). The catheter was replaced after sterilization of the urine, to avoid the risk of bacteraemia (Gillespie, 1986). Especially in geriatric patients, the danger of neuropathy following nitrofurantoin has to be taken into account. Therefore, we do not include nitrofurantoin in the prophylactic regimen, and we used it therapeutically in the low dosage of 50 mg four times daily. Also, at this low dosage practically no short-term side effects occur. Candiduria did not occur, either in the gtoup on medication, or in the group on placebo. In view of our observations in catheterized patients in nursing homes, we still employ amphotericin B in our prophylactic regimen, but its value needs further investigation. As Platt has shown in retrospective and prospective studies, asymptomatic bacteriuria may be associated with increased morbidity and mortality in catheterized

7 Norfloxadn SDD in prophylaxis of bacffluria 921 patients (Platt, 1987). He also suggests that other authors could not show a bacteriuria-associated increase in morbidity and mortality, because bacterium could not be prevented in the treated groups (Platt, 1987). Therefore, effective preventive regimens are a prerequisite to investigate the consequences of asymptomatic bacteriuria in catheterized patients. Effective regimens for the prevention and treatment of bacteriuria are also needed for those patients who suffer from recurring symptomatic urinary tract infection during catheterization and for patients who have recently received an implant. The latter type of patients should be protected from haematogenous infection of the implant (d'ambrosia, Shoji & Heater, 1976). In this study, using norfloxadn 200 mg twice daily, it was found that it is possible to postpone bacteriuria from 7 days to 17 days, to prevent Gram-negative bacilluria altogether during a median time of 23 days of catheterization, and to obtain a further reduction of bacteriuric days by treatment with nitrofurantoin of Gram-positive bacteriurias. Acknowledgements The authors thank Dr R. de Graaf of the Department of Statistical Consultation, University of Nijmegen for statistical advice. The authors are grateful to the nursing staff of the surgical department for the collection of samples of faeces and urine, to A. J. H. M. Janssen and A. H. J. Sanders-Reijmers for technical assistance, and to Ms C. van Lingen for typewriting the manuscript. This investigation was supported in part by Merck, Sharp and Dohme, The Netherlands. References d'ambrosia, R. D., Shoji, H. & Heater, R. (1976). Secondarily infected total joint replacements by hematogenous spread. Journal of Bone and Joint Surgery (American Volume) 58, Clasener, H. A. L., Vollaard, E. J. & van Saene, H. K. F. (1987). Long-term prophylaxis of infection by selective decontamination in leukopenia and in mechanical ventilation. Reviews of Infectious Diseases 9, Daifuku, R. & Stamm, W. E. (1984). Association of rectal and urethral colonization with urinary tract infection in patients with indwelling catheters. Journal of the American Medical Association 252, Gillespie, W. A. (1986). Antibiotics in catheterized patients. Journal of Antimicrobial Chemotherapy 18, Kresken, M. & Wiedemann, B. (1988). Development of resistance to nalidixic acid and the fluoroquinolones after the introduction of norfloxacin and ofloxacin. Antimicrobial Agents and Chemotherapy 32, Kunin, C. M. (1987). Detection, Prevention and Management of Urinary Tract Infections, 4th edn. Lea and Febiger, Philadelphia PA. Kunin, C. M. & Steele, C. (1985). Culture of the surfaces of urinary catheters to sample urethral flora and study the effect of antimicrobial therapy. Journal of Clinical Microbiology 21, Lennctte, E. H., Balows, A., Hausler, W. J. & Shadomy, H. J. (Eds). (1985). Manual of Clinical Microbiology, 4th edn. American Society for Microbiology, Washington, DC. Platt, R. (1987). Adverse consequences of asymptomatic urinary tract infections in adults. American Journal of Medicine 82, Suppl. 6B, SAS Institute Inc. (1985). SAS Users Guide: Statistics, Version 5 Edition. SAS Institute Inc, Cary NC.

8 922 E. J. Vollaartl et al Schaeffcr, A. J. (1986). Catheter-associated bacteriuria. Urologic Clinics of North America 13, Shapiro, M., Simchen, E., Izraeli, S. & Sacks, T. G. (1984). A multivariate analysis of risk factors for acquiring bacteriuria in patients with indwelling urinary catheters for longer than 24 hours. Infection Control 5, Stamey, T. A. (1987). Recurrent urinary tract infections in female patients: an overview of management and treatment Reviews of Infections Diseases 9, Suppl. 2, S Stickler, D. J. & Chawla, J. C. (1987). The role of antiseptics in the management of patients with long-term indwelling bladder catheters. Journal of Hospital Infection 10, Turck, M. & Stamm, W. (1981). Nosocomial infection of the urinary tract. American Journal of Medicine 70, Winberg, J., Bergstrom, T, Lincoln, K. & Lidin-Janson, G. (1973). Treatment trials in urinary tract infection (UTT) with special reference to the effect of antimicrobials on the fecal and periurethral flora. Clinical Nephrology 1, Wolfson, J. S. & Hooper, D. C. (1988). Norfloxacin: a new targeted fluoroquinolone antimicrobial agent. Annals of Internal Medicine 108, (Received 26 September 1988; revised version accepted 18 January 1989)