C. Ciprofloxacin in peritonitis associated with continuous ambulatory peritoneal dialysis (CAPD)

C. Ciprofloxacin in peritonitis associated with continuous ambulatory peritoneal dialysis (CAPD)

Journal of Antimicrobial Chemotherapy (90) 6, Suppl. F, 63-7 A comparison between oral ciprofloxacin and intraperitoneal vancomycin and netilmicin in CAPD peritonitis J. S. Tapson', K. E. OTT*, J. C. George", E. StansfieW*, A. J. Bint* and M. K. Ward" Departments of'nephrology and b Clinical Microbiology, Royal Victoria Infirmary, Newcastle-upon-Tyne, Tyne and Wear NE4LP, UK This report describes a prospective, randomized comparison of oral ciprofloxacin and intraperitoneal vancomycin/netilmicin in the treatment of 50 consecutive episodes of CAPD peritonitis in 35 patients. Successful cure of peritonitis was achieved in 76% of subjects taking oral ciprofloxacin and 7% of those given intraperitoneal antibiotics. Satisfactory concentrations of ciprofloxacin in dialysate were achieved in all patients. Failure of ciprofloxacin was due to persistence of an isolate of intermediate sensitivity (), to persistence with acquisition of resistance (), and to relapse/reinfection in the remaining four cases (with resistant or moderately sensitive strains in three cases). Ciprofloxacin was well tolerated in the majority of cases. A significant rise in serum creatinine was noted in almost all patients taking oral ciprofloxacin. The advantages of oral drug administration indicate that oral ciprofloxacin is the preferred first-line treatment of CAPD-associated peritonitis. Introduction Peritonitis remains the major complication of continuous ambulatory peritoneal dialysis (CAPD) despite improved technology. Many different antimicrobial agents have been used to treat CAPD peritonitis. Antibiotics have usually been administered intraperitoneally or intravenously, and a variety of different dosing regimens have been suggested. No single regimen has been shown to be the most efficacious in clinical trials. There have been few reports of treatment with oral antibiotics (Knight et al., 8; Drew et al., 84; Searle & Raman, 85). Indeed a recent report from an ad hoc Advisory Committee on Peritonitis Management fails to recommend oral antimicrobial therapy for this complication (Keane et al., 89). However, phannacokinetic studies of oral treatment with the quinolone antibiotic ciprofloxacin have shown that peritoneal dialysate ciprofloxacin levels can be maintained above the MIC of most species causing CAPD peritonitis (Fleming et al., 87). A subsequent study demonstrated that a single course of oral ciprofloxacin was 76% successful as a first-line treatment for CAPD peritonitis caused by a wide range of organisms (Fleming et al., 90). However, studies comparing oral ciprofloxacin with traditional intraperitoneal antibiotic regimens have been lacking. If effective, oral treatment would offer the following potential advantages: simple drug administration, outpatient management, early commencement of home treatment, the avoidance of dialysate contamination and cost reduction. We have therefore performed a prospective, randomized study comparing the use of oral ciprofloxacin 63 0305-7453/90/6F063 + 09 $0.00/0 90 The British Society for Antimicrobial Chemotherapy

64 J. S. Tapson et al. with intraperitoneal vancomycin and/or netilmicin in 50 consecutive episodes of CAPD peritonitis presenting to the Renal Unit at the Royal Victoria Infirmary, Newcastleupon-Tyne. Patients and methods During the period from December 88 until March 90, 50 consecutive episodes of CAPD peritonitis (dialysate white cell counts greater than 00///, with or without other symptoms and signs) were entered into the study. All patients gave informed consent. Exclusion criteria were the presence of vomiting, chronic liver disease, a history of convulsions or allergy to compounds of the nalidixic acid/quinolone class, pregnancy or co-existing antibiotic therapy. No patient entered was taking theophylline and any oral medication containing magnesium or aluminium compounds was stopped. The 50 episodes of peritonitis occurred in 35 patients (0 males, 5 females); 5 patients had only one episode, seven had two episodes, two had three episodes and one lady hadfiveepisodes. No individual had experienced peritonitis during the 8 days before study entry. After the diagnosis of peritonitis was established each patient performed three two-litre dialysate 'flush' exchanges followed by four two-litre exchanges every day during the period of study. The patients were randomized using a predetermined randomization code, generated by a computer program, in blocks of ten subjects, five to each treatment regimen. Patients received ten days treatment with either oral ciprofloxacin or intraperitoneal vancomycin and/or netilmicin. Patients treated with ciprofloxacin took the antibiotic orally at the time of each dialysate exchange. Patients whose body weight was > 70 kg took 500 mg qds (n = 7) whilst those who weighed < 70 kg took 50 mg qds (n = 8). Patients randomized to receive intraperitoneal antibiotic therapy commenced treatment with vancomycin 30 mg added to each two-litre dialysate bag, and netilmicin 30 mg in alternate bags (unless Gram stain of the presenting dialysate indicated appropriateness of monotherapy). Vancomycin and netilmicin were prepared for the patients in pre-loaded syringes to ensure accurate intraperitoneal dosage. On day, all patients returned to the dialysis unit for re-assessment. A 0-ml sample of dialysate effluent was taken for microbiological evaluation from the morning drained bag. Aliquots were also taken from the morning and noon effluents for measurement of dialysate ciprofloxacin concentrations in those individuals receiving this antibiotic. Venous blood samples were taken in these patients immediately before (trough) and 90 min after (peak) the lunchtime dosage to measure plasma ciprofloxacin levels. From day, intraperitoneal antibiotics were modified in accordance with the sensitivities of isolates. Eighteen patients continued on vancomycin alone, three subjects received netilmicin alone, whilst four patients needed to continue adding both antibiotics to their dialysate bags. Patients taking oral ciprofloxacin returned to the unit on the fifth to seventh day of treatment for repeat dialysate and plasma ciprofloxacin levels (as above). All patients were reassessed after ten days of antibiotic treatment. Routine blood tests and dialysate cultures were performed on all patients. Those taking oral ciprofloxacin had dialysate and plasma concentrations re-measured (as above). Having completed ten days antibiotic treatment, patients were followed until day 8 to identify relapses. The sensitivity of isolates to the study antibiotics was routinely determined by a breakpoint method using DST agar and antibiotic concentrations of: ciprofloxacin

CAPD peritonitis 65 and 4mg/l, netilmicin 4 and 0mg/l, and vancomycin 8mg/l. MICs of ciprofloxacin were determined by an agar dilution method using DST agar with 5% lysed horse blood, and ciprofloxacin concentrations from O008 to 6 mg/. An inoculum of approx. KPcfu/spot was used with Staphylococcus aureus NCTC657 and Escherichia coli NCTC048 as controls. Blood and dialysate ciprofloxacin concentration were measured by a high performance liquid chromatography (HPLQ method using a Spherisorb ODSII column at 50 C with fluorescence detection and filters of 78 and 445 nm and a mobile phase of phosphoric acid and acetonitrile. The lower limit of detection was 0 mg/. Statistical analysis of pre- and post-treatment data was performed using Student's paired /-test. All means are expressed ± standard deviation. Patient characteristics Results The 5 episodes of peritonitis treated with oral ciprofloxacin occurred in patients with a mean age of 58-8 years (range 30-8) who had been on CAPD for an average of 9-6 months (range -89) and had experienced between 0 and 4 episodes of peritonitis during the previous year. There were no significant differences in these characteristics among the patients who received intraperitoneal antibiotics: 5 peritonitis episodes, mean age 55-3 years (range 30-76), time on CAPD 4-8 months (range -93) and 0-3 episodes of peritonitis during the previous year. Similarly, both groups of patients had an even distribution of associated diseases (angina pectoris, hypertension, CVA, etc) and concomitant appropriate medications. Clinical presentation Ah peritonitis episodes were associated with a cloudy dialysate effluent. In the ciprofloxacin group, 5 patients had abdominal pain, three had diarrhoea, two had nausea, one had chills and one patient had a fever. In the vancomycin/netilmicin group, individuals had abdominal discomfort, three had diarrhoea, one had nausea and six were pyrexial. Initial investigations There were no significant differences in haematology and biochemistry results at presentation between the two groups of patients (Table I). Dialysate white cell counts were similar in both groups. Table II documents the causative organisms isolated from the dialysate effluents at presentation from both groups of patients. Forty-two of these isolates were tested for sensitivity to ciprofloxacin by the breakpoint method: 88% were sensitive; 7% were moderately sensitive (E.coli, ; coagulase-negative staphylococcus ) and 5% were resistant (non-haemolytic streptococcus, ; coagulase-negative staphylococcus, ). Subsequent investigations Ciprofloxacin group. Table III shows the paired results of haematology and biochemistry before and after ciprofloxacin treatment. Significant increases in serum

66 J. S. Tapsoo et at Table L Initial investigations Ciprofloxacin group Vancomycin/ nctilmicin group Haemoglobin (g/dl) WCC(x 07) Platelets (x 0 9 /!) Na (mmol/) K (mmol/) Ca (mmol/) Urea (mmol/) Creatinine (/anol/) Urate (mmol/) Aspflrtntc trflp"" ''n!i«(u/) Alkaline phosphatase (u/) Dialysate WCC/ul 9- ±-7 9- ±3-7 304-9 ±74-3 36-9 ±4-4- ±-0-3±0-9 ±60 83-0 ±46-9 0-37±(M) 4-8 ±8-5 37-5 ±80-3 905 ±3793 9-7±-l 0-3 ±5 357-9± 0-36-9±3-4- ±0-7 -3 ±0-8 ±5-9 845-9 ±78-0 037 ±0-07 8-9±5-8 6-7 ±55-3 47 ±344 creatinine, urate and aspartate transaminase were noted after treatment. The rise in serum creatinine was particularly marked with a mean increment of 0-6/imol/l. Vancomycin/netilmicin group. Table IV shows the paired data for haematology and biochemistry before and after intraperitoneal antibiotics. Only serum urate changed significantly after treatment Clinical response Nineteen of the 5 peritonitis episodes were finally cured by ciprofloxacin. In two episodes ciprofloxacin failed to resolve the peritonitis. After satisfactory response, a further four patients subsequently relapsed. Eighteen of the 5 episodes treated with intraperitoneal antibiotics were finally cured. In the remaining seven patients, peritonitis resolved only to relapse after antibiotic therapy had finished. Table II. Isolates from dialysate at presentation Ciprofloxacin group Vancomycin/ nctilmicin group Coagulase-negative staphylococus Staph aureus Micrococcus Diphtheroid Streptococcus sp. Coliform E. coli Klebsiella sp. Pseudomonas sp. No growth 0 5 0 0 6 3 3 0 0 0 4

CAPD peritonitis 67 Table HI. Investigations before and after treatment with ciprofloxacin No. of patients Before After Haemoglobin (g/dl) WCC(x 07) Platelets (x 07) Na (mmol/) K (mmol/) Ca (mmol/) Urea (mmol/) Creatinine Qnnol/) Urate (mmol/) Aspartate transaminasc (u/) Alkaline phosphatase P<<HK),3-0-6. h P <OO. Bacteriological response 0 7 7 7 9-6-9 8-83-6 33-770-0 37-73-3 4-00-7-3405 -60 805-356- 0-370O3 4-98-7 37-48-4 9-80 8-63-4 33-77-9 38-3-9 4-0-7-370-8 06-7 96-356-' 0-4005* -6* 8-879-9 Ciprofloxacin group. Thirteen of the 5 peritonitis episodes were successfully treated with ciprofloxacin with bacteriologically proven eradication of the isolated organism. In six other successful clinical responses to oral ciprofloxacin bacteriological response was indeterminate because of a failure to obtain a positive culture from pre-treatment dialysate effluent. Six episodes of peritonitis were treatment failures on ciprofloxacin. In two cases, both due to coagulase-negative staphylococci, the organisms persisted despite treatment. One of these infections was due to a strain of intermediate sensitivity to ciprofloxacin at presentation (MIC, mg/), and the other was due to two separate strains, both of which acquired resistance to the antibiotic by day 5 of treatment (MIC increased from 0-5 to mg/). In two other cases, peritonitis returned following Table TV. Investigations before and after vancomycin/netilmicin Haemoglobin (g/dl) WCC(x 07) Platelets (x 0*/l) Na (mmol/) K (mmol/) Ca (mmol/) Urea (mmol/) Creatinine (/nnol/l) Urate (mmol/) Aftpartptr fnin<uift inax (u/) Alkaline phosphatase No. of patients 8 8 7 0 0 5 Before 9-7-3 0-85-4 348-07-5 36-83-4 4-0-7-40- 806-0 809-668-5 0-350-07-46-8 33-75-5 treatment After l(k>i- 9-3-4 370-50- 38-04- 4-0-9-40- 7-66-5 833-977-5 0-40-08* 9-744-9 3-543-4 with 'P<O00\.

68 J. S. Tapson et al. successful eradication of a ciprofloxacin sensitive organism or a ciprofloxacin intermediately sensitive organism after completion of ten days treatment. One patient redeveloped a cloudy bag on day, owing to recurrence of infection by the original diphtheroid, which remained moderately sensitive to the antibiotic (MIC = mg/). The other patient experienced a clinical relapse on day 4 due to recurrence of a coagulase-negative staphylococcus which maintained ciprofloxacin sensitivity. Two other cases relapsed after successful eradication of ciprofloxacin-sensitive S. aureus isolates with re-infection by a different organism (a coagulase-negative staphylococcus, resistant to ciprofloxacin, isolated in both cases, on days 7 and 0, respectively). Vancomycin/netilmicin group. Fifteen of the 5 peritonitis episodes were successfully treated with intraperitoneal antibiotics with eradication of the presenting isolates. In three other successful clinical responses bacteriological evaluation was indeterminate because of negative culture results from the presenting cloudy dialysate effluent Seven episodes of peritonitis were treatment failures on the vancomycin/netilmicin regimen. In all cases clinical resolution of peritonitis was followed by recurrence of cloudy bags after completion of the ten days intraperitoneal antibiotics. In two of these cases bacteriological response was indeterminate either because of a negative culture at presentation or from the recurrent cloudy bag (which occurred on days and 3 respectively). In the remaining five episodes of peritonitis, bacteriological eradication occurred after ten days intraperitoneal vancomycin. Recurrence of the vancomycinsensitive staphylococcus occurred in each case (on days, 5,, 5 and 7 respectively. Ciprofloxacin concentrations Figure shows the mean trough and peak plasma ciprofloxacin concentrations for the 8 individuals taking 50mgqds and the seven individuals taking 500mgqds. Figure shows dialysate ciprofloxacin concentrations in the two groups of patients taking this medication. Adverse effects The treatment was well tolerated. No patient treated with intraperitoneal antibiotics developed adverse reactions. Four patients on ciprofloxacin had minimal side effects: nausea (3), lethargy (), anorexia () and myalgia (). In only one of these patients were high plasma ciprofloxacin levels present (mean trough value 9-8 mg/, peak value -7 mg/). This patient took 500 mg ciprofloxacin qds. Discussion The intraperitoneal use of vancomycin with an aminoglycoside continues to be recommended for first-line treatment of peritonitis complicating CAPD (Spencer & Fenton, 84; Keane et al., 89). The use of such antibiotics complicates treatment regimens. Many patients find it difficult to add antibiotics to their CAPD bags and admission to hospital for training and/or treatment is occasionally necessary. Instillation of antibiotics may result in contamination of dialysate. In addition, although aminoglycoside ototoxicity after a single ten-day course is rare, repeated treatment may be required for

CAPD peritonitis 69 T 0 I a. (7) (7) T P / \ (5) I I T P' 6 (i4) ( ' 4) T 'C500 #C50 P 0 Doys Figure. Plasma ciprofloxacin concentrations. T, trough; P, peak; CSOO, 300mgqds dotage; CS0, 50 mg qdi dosage; ( ) number of patients in group. [mg/d 0 E 6 8 Mo i 4 &oprof (5 _ (6) ( i / (6) (5) (5) i i 6 (4)»C500 Flgnre. Dialytate dprofloxacin concentrations. CSOO, SOOmgqds dote; CS0, S0mgqds dose; ( ) number of patienu in group. Days (4) l 0

70 J. S. Tapson et al recurrent peritonitis, and the dangers of ototoxicity increase. Finally, the use of intraperitoneal antibiotics is expensive. For these reasons the treatment of CAPD peritonitis would be greatly simplified by an oral antibiotic active against the commonly encountered pathogens that is able to reach bactericidal concentrations rapidly in the peritoneal dialysate. Previous reports that describe oral therapy have used cephalosporins (Knight et al., 8; Drew et al., 84; Searle & Raman, 85). Ciprofloxacin is a bactericidal antimicrobial agent. It has a broad spectrum of activity including Gram-negative and certain Gram-positive organisms, being particularly active against members of the Enterobacteriaceae and Pseudomonas species. In addition cross-resistance between ciprofloxacin and the /?- lactams and aminoglycosides is not a problem. Preliminary studies have demonstrated that therapeutic concentrations of ciprofloxacin may be achievable in the peritoneal dialysis fluid after oral administration to patients undergoing CAPD (Shalit et al., 86). A subsequent study demonstrated the efficacy of oral ciprofloxacin in treating 33 unselected episodes of CAPD peritonitis (Fleming et al., 90). Our prospective, randomized comparison of oral ciprofloxacin and intraperitoneal vancomycin/netilmicin confirms that this oral antibiotic is a useful alternative in the management of bacterial peritonitis in CAPD patients. Overall a single course of oral ciprofloxacin was 76% successful as first-line treatment in 5 episodes of CAPD peritonitis. This compares with a success rate of 7% for 5 episodes in a comparable group of individuals treated with intraperitoneal antibiotics. The use of intraperitoneal antibiotics resulted in a clinical cure of all episodes of peritonitis by day 0 of treatment with clear dialysate effluent in all cases. Failure of therapy was due to relapse after discontinuation of antibiotics. Clear dialysate effluent was achieved after ten days oral ciprofloxacin treatment in 3 patients. However, four of these subjects subsequently redeveloped cloudy dialysate after oral treatment had been discontinued. The return of peritonitis was due to recurrence/re-infection by resistant or moderately sensitive strains in three of these cases. Dialysate ciprofloxacin levels were maintained above mg/l (which exceeded the MIC for 95% of the original isolates causing peritonitis in this study) in all but three patients. These three patients achieved a clinical cure. In the six patients who were treatment failures, dialysate ciprofloxacin concentrations exceeded -7 mg/ in all cases. Treatment failure of ciprofloxacin was due to organisms with reduced sensitivity in all but one case, rather than an inability to achieve satisfactory antibiotic levels in peritoneal dialysate. Ciprofloxacin was well tolerated in most cases. In only one patient were adverse effects felt to be a consequence of high plasma ciprofloxacin levels. A significant rise in plasma urate levels was noted in both treatment groups. Elevation of plasma transaminases during ciprofloxacin has been reported in some cases previously (Arcieri et al., 86). In none of our patients, however, did this serum enzyme level rise out of the normal reference range for our laboratory. A rise in serum creatinine has also been previously reported in a few patients (Arcieri et al., 86). Seventeen of the patients taking ciprofloxacin, for whom paired data were available, showed a rise in serum creatinine. The explanation of this change remains uncertain, but the effect was transient, and return to pre-treatment levels occurred in all patients at follow up. Oral ciprofloxacin treatment is also cost-effective. At present prices (April 90), ten days treatment with ciprofloxacin 50mgqds costs 6.7 (exclusive of VAT). This compared with 3.9 for vancomycin prepared in pre-loaded syringes, taken from

CAPD peritonitis 7 3x5OOmg vials, and 7.4 for netilmicin in pre-loaded syringes, taken from x50mg vials. Acknowledgements We are grateful to Dr L. O. White, Southmead Hospital, Bristol for performing ciprofloxacin assays in the early part of the study. We also thank the nursing staff of the Regular Dialysis Unit, Royal Victoria Infirmary for their help and co-operation with this study, and Miss C. Lee for secretarial assistance. References Arcieri, G., August, R., Becker, N., Doyle, C, Griffith, E., Gruenwaldt, G. et al. (86). Clinical experience with ciprofloxacin in the USA. European Journal of Clinical Microbiology 5, 0-5. Drew, P. J., CaseweU, M. W., Desai, N., Houang, E. T., Simpson, C. N. & Marsh, F. P. (84). Cephalexin for the oral treatment of CAPD peritonitis. Journal of Antimicrobial Chemotherapy 3, 53-9. Fleming, L. W., Moreland, T. A., Scott, A. C, Stewart, W. K. & White, L. O. (87). Ciprofloxacin in plasma and peritoneal dialysate after oral therapy in patients on continuous ambulatory peritoneal dialysis. Journal of Antimicrobial Chemotherapy, 493-503. Fleming, L. W., Phillips, G., Stewart, W. K. & Scott, A. C. (90). Oral ciprofloxacin in the treatment of peritonitis in patients on continuous ambulatory peritoneal dialysis. Journal of Antimicrobial Chemotherapy 5, 44-8. Keane, W. F., Dale Everett, E., Fine, R. N., Golper, T. A., Vas, S., Peterson, P. K. et al. (89). Continuous ambulatory peritoneal dialysis (CAPD) peritonitis treatment recommendations: 89 update. Peritoneal Dialysis International 9, 47-56. Knight, K. R., Polak, A., Crump, J. & Maskell, R. (8). Laboratory diagnosis and oral treatment of CAPD peritonitis. Lancet ii, 30-4. Sanders, C. C, Sanders, W. E. & Goering, R. V. (87). Overview of precunical studies with ciprofloxacin. American Journal of Medicine 8, Suppl. 4A, -. Earle, M. & Raman, G. V. (85). Oral treatment of peritonitis complicating continuous ambulatory peritoneal dialysis. Clinical Nephrology 3, 4-4. Shalit, I., Greenwood, R. B., Marks, M. I., Pederson, J. A. & Frederick, D. L. (86). Pharmacoldnetics of single-dose oral ciprofloxacin in patients undergoing chronic ambulatory peritoneal dialysis. Antimicrobial Agents and Chemotherapy 30, 5-6. Spencer, R. C. & Fenton, P. A. (84). Infective complications of peritoneal dialysis. Journal of Hospital Infection 5, 33-40.