Ampicillin and Sulbactam Pharmacokinetics and Pharmacodynamics in Continuous Ambulatory Peritoneal Dialysis (CAPD)

Transcription

1 Peritoneal Dialysis International, Vol. 10, pp , /90 $ Printed in the USA All rights reserved Copyright 1990 Peritoneal Dialysis Bulletin, Inc Ampicillin and Sulbactam Pharmacokinetics and Pharmacodynamics in Continuous Ambulatory Peritoneal Dialysis (CAPD) Brad G. Blackwell,1 James E. Leggett,2 Curtis A. Johnson,3 Stephen W. Zimmerman,4 and William A. Craig2 10epartment of Pharmacy, Lutheran General Hospital, Park Ridge, Illinois, 20epartment of Medicine, Veterans Administration Hospital, Madison, Wisconsin, 3School of Pharmacy, University of Wisconsin, Madison, Wisconsin, 4School of Medicine, University of Wisconsin, Madison, Wisconsin The fixed combination antibiotic ampicillin/sulbactam may provide a new, safe, and effective method of treating dialysisrelated bacterial peritonitis. The pharmacokinetics of this antibiotic combination were determined in patients receiving continuous ambulatory peritoneal dialysis (CAPD). The pharmacodynamic activity of this drug was also determined by use of mean bactericidal titers against selected bacterial strains. Six noninfected CAPD patients in a randomized twoway crossover study were given a fixed dose of ampicillin (2 gm) and sulbactam (1 gm) either intravenously or intra peritoneally. The mean peak ampicillin and sulbactam serum concentrations following intravenous dosing were and 87.5 μg/ml, respectively. The mean peak serum concentrations of ampicillin and sulbactam following intraperitoneal dosing were 48.0 and 27.8 μg/ml, respectively. Absolute bioavailabilities of the intraperitoneal ampicillin and sulbactam doses were 60% and 68%. Both drugs exhibited similar distribution and elimination characteristics. Renal failure markedly reduced drug elimination. Intraperitoneal administration of ampicillin/sulbactam provided satisfactory inhibitory and bactericidal antibiotic titers for most organisms in dialysate at 6 h but not 24 h. Ampicillin/sulbactam (2 gm/1 gm} should be administered every 12 h to patients with peritoneal dialysis-related peritonitis. KEY WORDS: Ampicillin/sulbactam; pharmacokinetics; pharmacodynamics; peritonitis; bacterial susceptibility. A Pproximately two-thirds of peritonitis episodes associated with peritoneal dialysis are caused by gram-positive organisms (1). The combination of ampicillin/sulbactam, a beta-lactamase inhibitor, is highly active in vitro against staphylococcal and streptococcal species, as well as many gram negative bacteria causing PD-peritonitis (2-4), and, therefore, is of interest as a possible effective and nontoxic treatment of peritonitis. Sulbactam has limited in Correspondence to: Curtis A. Johnson, School of Pharmacy, University of Wisconsin, 425 North Charter Street, Madison, WI Received July 25, 1989; accepted September 12, trinsic antibacterial activity with the notable exception of N. ganarrhaeae (5,6). However, it does expand the spectrum of activity of beta-lactam antibiotics, such as ampicillin, to include beta-lactamase producing organisms (2). Clinical experience has shown ampicillin to be well-tolerated with few adverse effects. This study was conducted to evaluate the potential use of ampicillin/sulbactam in the treatment of continuous ambulatory peritoneal dialysis (CAPD) related bacterial peritonitis. To establish dosing guidelines, this study characterized the intraperitoneal bioavailability, distribution, and elimination of a single dose of a fixed combination of ampicillin/sulbactam following two-way crossover intravenous and intraperitoneal administration to noninfected CAPD patients. In addition, ampicillin/sulbactam's antimicrobial activity in dialysate effluent was determined by in vitro susceptibility studies against selected bacterial isolates. METHODS PATIENT POPULATION Eight noninfected CAPD patients between the ages of 33 and 83 participated in this study. The weight of the patients ranged from 54 to 89.5 kg. Prior to the study, no patient had received antibiotics for 7 days, but all patients were allowed to take any other medications prescribed for the management of chronic renal failure. No patient took any medication known to affect the metabolism or elimination of ampicillin or sulbactam. All patients had the anemia of chronic renal failure and had normal hepatic function as determined by routine liver function tests. Written informed consent was obtained from all patients. DOSING Ampicillin 2 gm/sulbactam 1 gm was administered as a single dose in a randomized cross-over fashion, intravenously (IV) and intraperitoneally (IP). There was a washout period of at least 10 days between doses.

2 SAMPLES Prior to drug administration, an intravenous cannula was inserted into a forearm vein. Dextrose 5% in water was infused continuously at a rate of 30 ml/hr. All blood samples were drawn from this cannula after a 5 ml sample was removed to prevent any dilutional effect of the intravenous fluid. For intravenous drug administration, a temporary intravenous catheter was inserted into the opposite arm. The IV dose was added to 50 ml of Dextrose 5% in water, and infused over 15 min. Blood samples were drawn at the end of the infusion, time 0. Prior to drug administration, each patient's indwelling peritoneal catheter was equipped with a 3-way stopcock. Sampling of the dialysate was performed by withdrawing 50 ml of dialysate into a syringe and then forcefully rein fusing this sample. This procedure was repeated twice prior to each sample withdrawal to thoroughly mix the dialysate within the peritoneal cavity. A 5 ml sample was then withdrawn for analysis. The IP dose was injected through the medication port into the dialysate bag just prior to administration. The dialysate was infused over 15 min. The 0 time sample was collected at the end of the IP infusion. Dialysate exchanges were performed every 6 h during the study. All patients received 2.08 liters of dialysate (Dianeal, Baxter, Deerfield, IL) containing 1.5% hydrous glucose. No additives other than heparin or insulin were permitted. Blood samples were collected at 0,0.25,0.5, 1, 1.5, 2, 4, 6, 8, 12, 18, 24, and 48 h following the IV and IP doses. Additional samples were also collected at 3 and 5 h following the IP dose. The blood was centrifuged and serum was obtained. Dialysate samples were collected at 0,0.25,0.5,1,2,4,6,6.5, 7,8, 10,12,13,14, 16, 18, 19, 20, 21, 24, 30, 36, 42, and 48 h following the IV and IP doses. Additional samples were collected at 3 and 5 h following the IP dose. After each 6 h exchange, the dialysate was drained and the dialysate volume measured. Urine was collected during the intervals of 0-2, 2-4, 4-8, 8-12, 12-24, and h, and samples were saved for analysis. All serum, dialysate, and urine samples were immediately stored at -70 C until assayed. DRUG ANALYSIS Simultaneous ampicillin/sulbactam concentrations in serum, dialysate, and urine were determined by high-performance liquid chromatography (HPLC), using an adaptation of a previously published method (7). The analyses were performed by Pfizer Quality Control, Groton, Connecticut. An internal standard of cefazolin was used for the serum assays while external standards were used for urine and dialysate analysis. Final ampicillin and sulbactam concentrations were determined by linear regression analysis using peak height ratios of sulbactam and ampicillin to the standards. No interfering substances were de tected in blank uremic serum, dialysate, or urine. Ampicillin and sulbactam recovery was 92% in all specimens. Assay precision for ampicillin and sulbactam was 5% relative standard deviations. Semilogarithmic plots were constructed of the ampicillin and sulbactam serum concentrations versus time for both IV and IP dosing. The slope of the terminal elimination phase, kβ' was determined by linear regression. The elimination half life in serum, t112, was calculated as: t112 = (In 2)/kβ. The area under the serum concentration time curve to the last measured time t ( mg.hr ), following IV and IP dosing, was cal liter culated by the linear trapezoidal rule. The residual area from the last measurable serum concentration to infinity was estimated by dividing the last measured serum concentration by kβ. The absolute bioavailabil P ity of the IP dose was calculated by AUCo &infin:i.total AUC O &infin: IV body clearance (CLTB) following IV dosing was determined by: Dialysis clearance (CLD) was determined by: CLD = CLTB X the fraction of the IV dose recovered in dialysate. Renal clearance (CLR) following the intravenous dose was determined by: CLR = CLTB X fraction of the IV dose recovered in urine. Volume of distribution (V β) was calculated: SUSCEPTIBILITY TESTING The 6 and 24 h (peak and trough) dialysate samples following IV dosing were utilized to determine the antimicrobial activity of ampicillin/sulbactam in dialysate. Susceptibility testing was performed using 2 standard laboratory strains and 3 clinical isolates: Escherichia coli A TCC 25922, Escherichia coli-ampi - cillin resistant UW35, Klebsiella pneumoniae UCLA 5166, Staphylocaccus epidermidis-ampicillin sensitive isolated from a patient with CAPD peritonitis, and Staphylocaccus aureusampicillin resistant. These organisms were selected because they represent bacteria commonly associated with PD-peritonitis. Minimum inhibitory (MICs) and bactericidal (MBCs) concentrations in 50% fresh dialysate-50% Mueller-Hinton broth mixture (D-MH) for ampicillin and sulbactam alone, and at an ampicillin/sulbactam ratio of 1:1 and 2:1, were determined in standard fashion using a microdilution technique (8). Bacterial in

3 ocula of approximately 5 X 105 colony-forming units (CFUs)!mL in Mueller-Hinton broth were placed in microdilution wells (0.1 ml volume) with ampicillin and sulbactam in fresh dialysate. The microdilution plates were incubated at 35 C for h. From each clear well, 25 μl was plated on Mueller-Hinton agar for CFU determinations. A 99.9% reduction in CFU from initial inoculum was used as the bactericidal endpoint. Inhibitory and bactericidal titers were determined following serial dilutions in fresh dialysate of the 6 and 24 h samples in microdilution plates. Bacteria were added as above and plates were incubated and processed as indicated previously to determine bactericidal endpoints. In addition, a predicted inhibitory! bactericidal titer was calculated by dividing observed ampicillin concentrations in dialysate at 6 and 24 h by the MIC (same as MBC) for each organism. Linear regression analysis was performed on the predicted and observed mean bactericidal titers for all patients and all organisms. Titers > 1:1024 were arbitrarily assigned a value of 1:2048. Titers < 1:2 were not included in the calculation. RESULTS Six patients completed the study. One patient was removed because of a presumed hypersensitivity reaction to the antibiotics, another was removed because of the development of cloudy peritoneal fluid. The mean serum and dialysate ampicillin and sulbactam concentrations following IV dosing are presented in Figures 1 and 2, respectively. The pharmacokinetic parameter values for IV ampicillin and sulbactam are presented in Table 1. Ampicillin and sulbactam distributed rapidly following IV infusion. They crossed the peritoneal membrane and appeared to achieve near equilibrium between serum and dialysate by 6 h. The total amount of ampicillin recovered in the dialysate during the 48 h. of sample collection after IV dosing was 11.3 ± 2.2%; that.of sulbactam, 15.2 ± 1.9% of the dose. The total amount of ampicillin and sulbactam recovered in the urine during the 48 h. of sample collection was 7.2 ± 8.0% and 6.7 ± 6.9% of the dose, respectively. The elimination half life of ampicillin was 9.5 h, that for sulbactam was 9.7 h. The mean serum and dialysate drug concentrations following IP dosing are presented in Figures 3 and 4. The pharmacokinetic parameter values following IP dosing are given in Table 1. Both drugs crossed the peritoneal membrane. Peak serum concentrations of ampicillin ( 48.0 ± 7.6 μg!ml) and sulbactam (27.8 ± 4.1 μg!ml) were reached 5-6 h after the peritoneal administration of the drugs. The absolute bioavailability of both drugs following IP dosing was about 60%. The elimination half lives of ampicillin and sulbactam in serum were 9.6 and 9.4 h, respectively. SUSCEPTIBILITY TESTING Peak ampicillin and sulbactam concentrations in dialysate occurred at 6 h following IV dosing and declined over the 48 h of the study. The ampicillin MICs in the presence of sulbactam were identical at concentration ratios of 1:1 and 2:1. For S. aureus, the ampicillin MIC in the presence of sulbactam was 2 μg!ml; for S. epidermidis, the MIC was 0.06 μg!ml; for K. pneumaniae, the MIC was 4 μg!ml; for the standard strain of E. cali A TCC 25922, the MIC was 2 μg!ml; and for ampicillinresistant E. coli UW35, the MIC was 64 μg!ml. The MICs for ampicillin alone were as follows: S. aureus, 512 μg!ml; S. epidermidis, 0.06 μg/ml; K. pneumoniae, 512 μg!ml; the standard strain of E. coli ATCC 25922,2.0 μg!ml; and ampicillin-resistant E. cali UW35, >512 μg!ml. Ampicillin!sulbactam bactericidal titers for the 6 and 24 h samples were identical to inhibitory titers for each organism. The mean observed reciprocal titers are presented in Table 2. Regression analysis revealed a highly significant correlation between predicted and observed bactericidal titers for all patients and all organisms (r2 = 0.95, y = x).

4 DISCUSSION The pharmacokinetic parameter values of ampicillin observed in this study are similar to those previously reported for patients with renal dysfunction and also for patients on CAPD (9, 10). It has been shown that renal clearance plays a major role in the elimination of ampicillin. Approximately 85% of the dose is eliminated in the urine and the half life is reported to be 1 h in patients with normal renal function (11, 12). Dialysis clearance contributed minimally to removal of ampicillin, as evidenced by recovery of only 11% of the dose in the dialysate after 48 h. Urinary excretion accounted for only 7% of the dose by 48 h. Intraperitoneal drug administration was well tolerated. From these pharmacokinetic data, it appears that ampicillin may be dosed IP or IV every h in patients receiving CAPD. Previous reports have shown that renal clearance is also the primary method of sulbactam elimination (13-16). In patients with normal renal function, renal clearance represents approximately 80% of total clearance, with 75-85% of the dose being recovered in the urine within 24 h (13). An elimination half life of 1 h has been observed (13). Following IV administration of sulbactam to CAPD patients, only 15% of the dose was recovered in the dialysate in 48 h. Approximately 7% of the dose was recovered in the urine at 48 h. Coadministration of ampicillin and sulbactam has been shown to have no effect on the pharmacokinetics of either drug (13). The pharmacokinetics of ampicillin and sulbactam are almost identical in patients with normal renal function (13). Our data indicate the pharmacokinetic parameters of ampicillin and sulbactam are virtually identical in patients on CAPD. The present study demonstrates a marked reduction in total ampicillin and sulbactam clearance in patients on CAPD. Reduced elimination of ampicillin and sulbactam led to a half life of approximately 9.5 h for each drug. Therefore, it should be possible to dose ampicillin and sulbactam in combination every h to patients undergoing CAPD. To be effective in the treatment of peritonitis, an antibiotic must be pharmacologically active in peritoneal fluid. Verbrugh, et at., have observed that tobramycin had only 10% of its bactericidal activity in dialysate when compared to Mueller-Hinton broth (17). However, the beta-lactam imipenem lost no activity. In another study, the beta-lactam cefoperazone also retained its bactericidal activity in dialysate (18). To determine the pharmacodynamic activity of ampicillin/sulbactam in peritoneal fluid, the present study included a determination of ampicillin and ampicillin/sulbactam MICs and MBCs in antibiotic-containing dialysis effluent. The results of our study indicate that ampicillin, in combination with sulbactam, maintains its predicted inhibitory and bactericidal activity in dialysate. A comparison of the peak and trough inhibitory and bactericidal titers in dialysate revealed that higher titers were obtained with IP dosing than with IV dosing. Intraperitoneal dosing provided therapeutic

5 (> 1:8) inhibitory and bactericidal titers at 6 h for all organisms except the strain of E. coli, which had a high degree of ampicillin resistance. Intravenous dosing provided therapeutic titers for only S. epidermidis and the ampicillin-sensitive strain of E. coli. The titers at 24 h following either route of administration were therapeutic only for S. epidermidis, which was exquisitely sensitive to the ampicillin/sulbactam combination. Linear regression revealed a good correlation between observed bactericidal titers and titers predicted from dialysate drug concentrations and organism MICs to ampicillin/sulbactam, (r2 = 0.95). During episodes of CAPD-associated peritonitis, it should be possible to predict therapeutic efficacy from the ampicillin/sulbactam concentrations achieved in dialysate if the MIC of the causative organism is known. Clinical trials with infected patients are needed to confirm this hypothesis. In most cases, ampicillin/sulbactam should be dosed IP every 12 h to treat CAPD-associated peritonitis. For peritonitis caused by highly susceptible organisms, bactericidal concentrations can be sustained with 24-h dosing. Intravenous administration of this antibiotic combination should also be efficacious for certain bacterial strains, as evidenced by the pharmacodynamic data presented in this report. However, for less susceptible strains, intraperitoneal dosing may be required. A limitation of this study is that noninfected patients were studied. Because the effects of peritonitis on the pharmacokinetics of ampicillin and sulbactam remain unknown, it will be necessary to confirm our results in studies with infected patients. Future studies should also examine the pharmacokinetic behavior of the drugs following multiple dosing. From the data obtained in this single-dose study, it appears unlikely that either ampicillin or sulbactam will accumulate to a potentially toxic concentration following multipledosing every 12 h to CAPD patients. The susceptibility data from the present study strongly suggest ampicillin/sulbactam is active against many pathogens causing PDrelated peritonitis. Because of the difference in bactericidal titers achieved following each route of administration, it may be necessary to use intraperitoneal dosing of ampicillin/sulbactam in most cases of peritonitis. Efficacy studies in infected patients will be necessary to confirm the susceptibility data presented in this paper. ACKNOWLEDGMENT This work was supported by a grant by Pfizer Pharmaceuticals. The authors would like to acknowledge the assistance of James B. Noffsinger of Pfizer Quality Control, Croton, Connecticut, for performing all ampicillin and sulbactam assays. REFERENCES I. Vas SI. Peritonitis. In: Nolph KD, ed. Peritoneal dialy sis, 2nd ed. Boston: Martinus Nijhoff; 1985:

6 2. Fu KP, Neu HC. Comparative inhibition of lactamases by novellactam compounds. Antimicrob Agents Chemother. 1979; 15: Lees L, Milson JA, Knirsch AK. Sulbactam plus ampicillin: Interim review of efficacy and safety for therapeutic and prophylactic use. Rev Infect Diseases. 1986; 8:S Retsema JA, English AR, Girard A, et al. Sulbactam/ ampicillin: In vitro spectrum, potency, and activity in models of acute infection. Rev Infect Diseases. 1986; 8:S Retsema JA, English AR, Girard AE. CP in combination with penicillin or ampicillin against pencillinresistant staphylococcus, Hemophilus influenzae, and bacteroides. Antimicrob Agents Chemother. 1980; 16: Caine V A, Foulds G, Handsfield HH. Therapeutic trial and pharmacokinetics of sulbactam for uncomplicated gonorrhea in men. Antimicrob Agents Chemother. 1984; 26: Haginaka J, Wakai J. Liquid chromatographic determi nation of penicillins by postcolumn alkaline degradation. Anal Chem. 1985: 57: Jones RN, Barry AL, Gavan TL, et al. Susceptibility tests: Microdilution and macrodilution broth procedures. In: Lennette EH, Balows A, Hausler WJ, Washington JA, eds. Manual of clinical microbiology, 4th ed. Washington DC: American Society for Microbiology; 1985: Bennett WM, Aronoff GR, Morrison G, et al. Drug prescribing in renal failure: Dosing guidelines for adults. Amer J Kidney Dis. 1983; 3: Jackson EA, McLeod DC. Pharmacokinetics and dosing of antimicrobial agents in renal impairment. Amer J Hasp Pharm. 1974; 31: Kucers A, Bennett NM. The use of antibiotics, 3rd ed. London: William Heinemann Medical Books Ltd. 1979: Wright AJ, Wilkowsek CJ. The penicillins. Mayo Clin Proc. 1983; 58: Foulds G, Stankewich JP, Marshall DC, et al. Pharma cokinetics of sulbactam in humans. Antimicrob Agents Chemother. 1983; 23: Brown RM, Wise R, Andrews JM, Hancox J. Comparative pharmacokinetics and tissue penetration of sulbactam and ampicillin after concurrent intravenous administration. Antimicrob Agents Chemother. 1982; 21: Schad VB, Guenin K, Straehl P. Single-dose pharmacokinetics of intravenous sulbactam in pediatric patients. Rev Infect Diseases. 1986; 8:S Rogers HJ, Bradbrook ID, Morrison PG, Spector RG, Cox DA, Lees LJ. Pharmacokinetics and bioavailability of sultamicillin estimated by high performance liquid chromatography. J Antimicrob Chemother. 1983; 11: Verbrugh HA, Keane WF, Conroy WE, Peterson PK. Bacterial growth and killing in chronic ambulatory dialysis fluids. J Clin Microbiol. 1984; 20: Johnson CA, Zimmerman SW, Reitberg DP, Whall TJ, Leggett JE, Craig W A. Pharmacokinetics and pharmacodynamics of cefoperazone/sulbactam in patients on continuous ambulatory peritoneal dialysis. Antimicrob Agents Chemother. 1988; 32:51-56.