Vancomycin for staphylococcal shunt site infections in patients on regular haemodialysis

Chemical Peritonitis due to Intraperitoneal Vancomycin (VANCOLED)

Six consecutive patients treated with intraperitoneal vancomycin for catheter infections developed cloudy dialysate in the exchange to which the drug was added. In all six patients, the dialysate was clear before the addition of vancomycin. All had elevated dialysate leukocyte and polymorphonuclear counts after the intraperitoneal vancomycin. All had sterile cultures. Two continued to have cloudy dialysate while continuing on 500 mg vancomycin per one exchange per day. In one of these patients, the catheter did not appear infected when removed. All episodes occurred after we had changed from Lilly's Vancocin to Lederle's Vancoled. During this time, the per cent of peritonitis episodes with sterile cultures rose from 0% in the previous 3 months to 50%. In all episodes with sterile cultures Vancoled had been administered via the dialysate. Fourteen patients had received intraperitoneal Vancocin in the previous nine months for the same indications; none developed cloudy dialysate. We believe that these cases represent chemical peritonitis due to Vancoled.

Vancomycin revisited: a reappraisal of clinical use

Vancomycin has been used for decades to treat serious systemic gram positive infections. Extensive use over time has demonstrated vancomycin is not nephrotoxic even when used in high dosage, i.e., twice the usual dose. Since vancomycin is not nephrotoxic, there is no rationale for dosing vancomycin based on serum vancomycin levels. Since vancomycin is eliminated by GFR, vancomycin dosing should be based on creatinine clearance. Vancomycin obeys "concentration dependent" kinetics and higher than usual doses may be useful in some infections (eg, osteomyelitis). Widespread vancomycin use has resulted in increased VRE prevalence worldwide. Among staphylococci, vancomycin induced cell wall thickening results in "permeability mediated" resistance to vancomycin, as well as other anti-staphylococcal antibiotics. "Permeability mediated" resistance accounts for the common clinical observation that MRSA infections treated with vancomycin often resolve slowly or not at all. Other effective MRSA antibiotics are available (eg, linezolid, daptomycin, minocycline, or tigecycline) and are more reliably effective, do not increase staphylococcal resistance or increase VRE prevalence.

Antimicrobial therapy of methicillin resistant Staphylococcus aureus infection

Methicillin-resistant Staphylococcus aureus (MRSA) is now one of the commonest causes of nosocomial infection worldwide. The mainstay of treatment until now has been the glycopeptides (vancomycin and teicoplanin). They are not without toxicity and need parenteral administration and monitoring of levels. The increasing frequency of MRSA infections, coupled with the emergence of glycopeptide resistance in S. aureus has made the introduction of new drugs active against Gram-positive organisms essential. New agents active against Gram-positive organisms represent either genuinely novel classes of antimicrobials (e.g., oxazolidinones and lipoproteins) or those derived from existing classes (e.g., tetracyclines, glycopeptides, streptogramins and cephalosporins). Some of these newer antibiotics appear to be effective against multi-resistant organisms including MRSA.

Vancomycin administration in continuous ambulatory peritoneal dialysis: the risk of ototoxicity

A prospective study was undertaken in 16 patients with chronic renal failure on continuous ambulatory peritoneal dialysis, with 22 episodes of peritonitis treated with vancomycin, a known ototoxic agent. Twelve patients had one episode each, and four had recurrent peritonitis. Each treatment course consisted of two infusions of vancomycin (30 mg/kg body weight) in 2 L of peritoneal dialysate administered at 6-day intervals. Serum vancomycin analyzed by enzyme immunoassay showed a mean trough level of 11.00 microg/ml on day 6 and mean serum levels of 33.8 and 38.6 microg/ml about 12 hours after administration on days 1 and 7, respectively. Similar levels, well within the therapeutic range, were encountered with repeated vancomycin therapy for recurrent episodes of peritonitis, suggesting that no changes occurred in the pharmacokinetic profile of the drug. Pure-tone audiometry, electronystagmography, and clinical assessment performed during each course of treatment showed no evidence of ototoxicity even on repeated courses of vancomycin therapy. The results suggest that vancomycin therapy when given in appropriate concentrations as a single therapeutic agent is both effective and safe. We believe, however, that vancomycin administered in combination with an aminoglycoside may produce ototoxic effects that may be greatly aggravated, possibly because of synergism.

Preoperative vancomycin prophylaxis decreases incidence of postoperative hemodialysis vascular access infections

The role of vancomycin in the treatment of infected arteriovenous chronic dialysis access is well established. However, the role of preoperative vancomycin administration in preventing infection in newly placed, revised, or surgically thrombectomized grafts has not been determined. We performed a prospective randomized study to examine whether vancomycin prophylaxis can decrease the incidence of postoperative graft infections. Over a 5-year period, 206 patients undergoing 408 permanent vascular access procedures were randomized into two groups. Group 1 (206 procedures) received a single intravenous dose of 750 mg of vancomycin approximately 6 to 12 hours before vascular access placement procedures, while group 2 (202 procedures) did not. Patients were evaluated for access infection within the following 30 days and before use of the access for chronic dialysis. Access infection developed in two patients (1%) in group 1 and in 12 patients (6%) in group 2 (P = 0.006). All 14 infections occurred in upper extremity polytetrafluoroethylene grafts. We conclude that the use of preoperative single-dose intravenous vancomycin prophylaxis for hemodialysis vascular graft procedures reduces the risk of postoperative access infection.

Comparison of the effects of three haemodialysis membranes on vancomycin disposition

Polysulfone (PSF) and polyacrylonitrile (PAN) were recently introduced haemodialysis (HD) membranes. The effect of each on vancomycin disposition was compared with cuprophan (SCE) in 12 chronic HD patients who received 14 infusions. Vancomycin (1 g) was infused over 1 hour, followed by three 4-hour HD sessions over 5 days, beginning 1 hour after the end of infusion. The intradialytic clearances of vancomycin were 73, 54 and 15 ml/min for PSF, PAN and SCE, respectively. At the end of the third HD session, vancomycin concentration dropped to subtherapeutic level (< 7.5 micrograms/ml) only in patients dialysed with PSF and PAN. The corresponding elimination half-lives (t1/2 beta) were 61, 60 and 86 hours for the three membranes, respectively. According to these findings, vancomycin should be given every three HD sessions for PSF and PAN. The dosage interval should be extended up to every 5 HD sessions for patients on SCE. The peak (mean +/- S.D.) obtained one hour after the end of infusion was 34.2 +/- 11.4 micrograms/ml, which is within the therapeutic range.

Vancomycin redistribution: dosing recommendations following high-flux hemodialysis

Although increased vancomycin clearance has been reported with highly permeable hemodialysis membranes (such as polysulfone), failure to consider post-dialysis redistribution could lead to unnecessary dosage supplementation. In protocol 1, twelve hemodialysis patients admitted for vascular access thrombectomy received 15 mg/kg of vancomycin as surgical prophylaxis. Post-operatively, patients underwent high-flux hemodialysis (HFHD) for two hours using a Fresenius F-80 polysulfone dialyzer (QB = 417 +/- 49, QD = 800 ml/min). Vancomycin's intradialytic clearance increased 13-fold compared to the patient's endogenous clearance (120 +/- 59 vs. 9 +/- 8 ml/min, respectively) yet dialysate recovery indicated that only 17% of body stores were removed (179 +/- 70 mg). Although serum vancomycin levels decreased 33% during HFHD, vancomycin levels increased in all patients following dialysis and the post-rebound values reached 87% of the pre-dialysis concentration. In protocol 2, eight outpatients receiving maintenance HFHD with a F-80 dialyzer (Kt/V = 1.29 +/- 0.08) were given 20 mg/kg of vancomycin immediately following dialysis on Monday; pre- and post-levels were measured during the next three dialysis treatments. The predialysis serum vancomycin levels were > 7.5 micrograms/ml (9.7 +/- 1.0 micrograms/ml; range 8.0 to 11.0) in all patients the following Monday. Thus, vancomycin clearance is increased during HFHD, but redistribution post-HD minimizes changes in serum levels. We recommend a 20 mg/kg i.v. loading dose and subsequent doses of 15 mg/kg every seven days; to account for individual variability, weekly vancomycin levels should be drawn before dialysis.

Drug dosage in patients during continuous renal replacement therapy. Pharmacokinetic and therapeutic considerations

The advantages of continuous haemofiltration and haemodialysis over intermittent haemodialysis for the treatment of acute renal failure are well recognised. In intensive care patients, 4 different continuous procedures, arteriovenous and venovenous haemofiltration (CAVH and CVVH) or haemodialysis (CAVHD and CVVHD), are employed. These effective detoxification treatments require knowledge of their influence on drug disposition. Data on kinetics of drugs during continuous treatment are scarce and limited almost exclusively to the oldest and least effective procedure (CAVH). Selected dialysis membranes may adsorb drugs, as in the case of aminoglycosides. In addition, elimination of substances with large molecular weights may vary depending on the pore size of the membrane, as in the case of vancomycin. Thus, even if drug dosages can be based on pharmacokinetic studies, selection of a dialysis membrane not studied may cause unpredictable drug concentrations. With these limitations in mind and considering the available literature on pharmacokinetics in patients with renal failure, general guidelines for drug dosage during continuous renal replacement therapy can be given. In haemofiltration, drug protein binding is the major factor determining sieving, i.e. the appearance of the drug in the ultrafiltrate. In haemodialysis, diffusion is added to ultrafiltration, and therefore the saturation of the combined dialysate and ultrafiltrate will decrease further with increasing dialysate flow rate. In continuous haemofiltration or haemodialysis the extracorporeal clearance can be calculated by multiplying the saturation value (estimated or, better, measured) with the ultrafiltrate and dialysate flow rate. Dividing the extracorporeal clearance by the total clearance (including the nonrenal clearance) gives the fraction of the dose removed due to extracorporeal elimination. Whether dosage recommendations available for anuric patients have to be modified or not can be decided on the basis of this value. In case of high nonrenal clearance, the degree of saturation is without clinical significance. Based on these considerations guidelines have been constructed for the effect of extracorporeal elimination on more than 120 different drugs commonly used in intensive care patients.

Simplified dosing and monitoring of vancomycin for the burn care clinician

Vancomycin has excellent activity against Gram-positive bacteria and is often selected for use in the infected burn patient. Because of multiple-compartment pharmacokinetics, vancomycin serum concentrations can decrease dramatically in a short time period following the end of an intravenous infusion. This accounts for the widely divergent recommendations for serum vancomycin peak concentrations, e.g. from 15 mg/l up to 80 mg/l, when the time for blood sampling following the end of intravenous infusion is different. It is in general not necessary to monitor vancomycin peak concentrations, not only because its toxic potential is overrated but also because potential toxicity and therapeutic efficacy are correlated with trough concentrations. Post-distribution 'peak' concentrations are generally only useful for determining the optimal dosing interval for patients with impaired renal function. A dosing and monitoring paradigm for vancomycin therapy in burned adults has been devised for burn care clinicians. It provides suggested dose and dosing intervals based on body weight and creatinine clearance, with specific recommendations for regimen modification based upon the results of trough serum concentration determinations.

Rebound of plasma vancomycin levels after haemodialysis with highly permeable membranes

Vancomycin is usually given only once a week to haemodialysis (HD) patients. If highly permeable dialysis membranes are used, however, high clearance values have been reported, so the aim of the study was to determine whether high clearance of vancomycin resulted in sufficient drug elimination to induce subtherapeutic plasma levels after one week. In 18 chronic HD patients, treated with polysulfone dialyzers (1.2 m2), the pharmacokinetics of vancomycin were studied after administration of 1 g. Concentrations were determined by fluorescence polarisation immunoassay. At a blood flow of 219 ml.min-1, HD clearance of vancomycin was 62.3 ml.min-1. Immediately after dialysis plasma concentrations were 38% lower than predialysis levels. However, marked rebound in the vancomycin level was observed 5 h later, resulting in plasma levels only 16% lower than prior to dialysis. 3 HD treatments in 1 week removed about one third of the initial dose. After one week 15 of 18 patients still had a therapeutic plasma level (greater than 4 micrograms.ml-1). In conclusion, polysulfone membranes show high clearance of vancomycin. However, transfer of drug from blood to dialysate appears to be faster than from tissues to blood. Because of a marked rebound in plasma level after treatment, therapeutic drug concentrations will still be present in most patients after one week.

Resistance to vancomycin and teicoplanin: an emerging clinical problem

Vancomycin and teicoplanin are glycopeptides active against a wide range of gram-positive bacteria. For 30 years following the discovery of vancomycin in 1956, vancomycin resistance was not detected among normally susceptible bacteria recovered from human specimens. Since 1986, however, bacteria resistant to vancomycin or teicoplanin or both have been described. Strains of the genera Leuconostoc, Lactobacillus, Pediococcus, and Erysipelothrix seem inherently resistant to glycopeptides. Species and strains of enterococci and coagulase-negative staphylococci appear to have acquired or developed resistance. There are at least two categories of glycopeptide resistance among enterococci, characterized by either high-level resistance to vancomycin (MIC, greater than or equal to 64 mg/liter) and teicoplanin (MIC, greater than or equal to 8 mg/liter) or lower-level vancomycin resistance (MIC, 32 to 64 mg/liter) and teicoplanin susceptibility (MIC, less than or equal to 1 mg/liter). The two categories appear to have similar resistance mechanisms, although genetic and biochemical studies indicate that they have arisen independently. Among coagulase-negative staphylococci, strains for which vancomycin MICs are up to 20 mg/liter or teicoplanin MICs are 16 to 32 mg/liter have been reported, but cross-resistance between these glycopeptides varies. The selective advantage accorded to glycopeptide-resistant bacteria and the observation that high-level resistance in enterococci is transferable suggest that such resistance may be expected to increase in incidence. Clinicians and microbiologists need to be aware of this emerging problem.

Clinical pharmacokinetics of vancomycin

Vancomycin utilisation has increased dramatically in the last 10 years due to the increasing clinical significance of infections with methicillin-resistant staphylococci. Recent studies have focused on characterising the disposition of vancomycin in patients and assessing the relationship between serum concentrations and therapeutic as well as adverse effects. Although vancomycin is not appreciably absorbed from the intact gastrointestinal tract, several recent case reports have documented the attainment of therapeutic and potentially toxic vancomycin serum concentrations following oral administration to patients with pseudomembranous colitis. The disposition of parenterally administered vancomycin has been best characterised by a triexponential model. The half-life of the initial phase (t1/2 pi) is approximately 7 minutes, that of the second phase (t1/2 alpha) is approximately 0.5 to 1 hour, while the terminal elimination half-life (t1/2 beta) ranges from 3 to 9 hours in subjects with normal renal function. The volume of the central compartment (Vc) in adults is approximately 0.15 L/kg while the steady-state volume of distribution (Vdss) ranges from 0.39 to 0.97 L/kg. More than 80% of a vancomycin dose is excreted unchanged in the urine within 24 hours after administration, and the concentration of vancomycin in liver tissue and bile has been reported to be at or below detection limits. Vancomycin renal clearance approximates 0.5 to 0.8 of simultaneously determined creatinine or 125I-iothalamate clearances, suggesting that the primary route of renal excretion is glomerular filtration. Recently, non-renal factors such as hepatic conjugation have been proposed as an important route of vancomycin elimination. However, these data are difficult to reconcile with other studies showing minimal non-renal clearance of vancomycin in subjects with end-stage renal disease. As yet, the disposition of vancomycin in patients with hepatic disease has not been adequately defined. Only limited data are available regarding the concentrations of vancomycin in biological fluids other than plasma. The penetration of vancomycin into cerebrospinal fluid (CSF) in patients with and without meningitis has been quite variable. Although early studies suggested that adequate CSF concentrations may not be achieved in subjects with uninflamed meninges, more recent investigations have reported contradictory results. Therapeutic concentrations of vancomycin, i.e. greater than 2.5 mg/L, have, however, been reported in ascitic, pericardial, pleural and synovial fluids. Tissue concentrations of vancomycin have exceeded simultaneous serum concentrations in heart, kidney, liver and lung sp

Vancomycin: an update

Vancomycin is a narrow-spectrum glycopeptide antibiotic with potent antistaphylococcal activity. It is primarily active against gram-positive organisms. Bacterial resistance rarely develops due to its numerous modes of action. The toxic potential of vancomycin is less significant than previously thought. "Red neck syndrome" seems to be the most common side effect and appears to be caused by rapid intravenous infusion. It is characterized by erythema at the base of the neck and the upper back; hypotensive episodes may also occur. Nephrotoxicity and ototoxicity are rare. Relationships between toxicities and serum concentrations have not been established. The disposition of vancomycin after intravenous administration proceeds biphasically--rapid distribution followed by elimination. The drug is excreted primarily unchanged in the urine by glomerular filtration. Vancomycin clearance is reduced and elimination half-life is prolonged in patients with renal insufficiency. Various methods have been published to aid in dosing the drug in these patients. Vancomycin is the drug of choice in the treatment of methicillin-resistant staphylococcal infections. It is also useful in the treatment of gram-positive endocarditis and has been used as alternative therapy in the treatment of prophylaxis of gram-positive infections in penicillin-allergic patients. Oral vancomycin is the preferred therapy in antibiotic-associated colitis.

Vancomycin prevents polytetrafluoroethylene graft infections in pediatric patients receiving chronic hemodialysis

Polytetrafluoroethylene (PTFE) grafts have been a useful addition to the pediatric hemodialysis vascular access armamentarium. In this study, 17 pediatric patients underwent 331 total months of hemodialysis via PTFE grafts. There was a statistically significant (P less than .025) decrease in the incidence of graft infections in 12 patients (235 patient-months) while receiving prophylactic parenteral vancomycin compared with 9 patients (96 patient-months) while receiving no vancomycin (0% v 44%). Vancomycin side effects were uncommon and mild. Vancomycin is a safe and effective agent for the prevention of PTFE graft infections in pediatric patients receiving chronic hemodialysis.

Staphylococcus epidermidis sepsis in pediatric patients: clinical and therapeutic considerations

This report concerns 60 children with documented Staphylococcus epidermidis sepsis. There were 34 boys and 26 girls, ages 2 weeks to 15 years. The primary diagnosis included malignancy (13), congenital (13) or acquired (11) gastrointestinal disorders, prematurity (7), cardiac defect (5), hydrocephalus (2) and miscellaneous (9). Clinical presentation included fever (54), tachycardia (15), lethargy (20), hypotension (8), irritability (6), increased gastric residuals (6) and apnea/bradycardia (3). A documented source of sepsis was noted in 56 patients, including percutaneous central venous catheters (23), Broviac catheters (17), umbilical arterial catheters (6), wound (3), V-P shunt (2), cardiac defect (2), cholangitis (1), chest tube (1) and peripheral arterial line (1). There were six sepsis-related deaths, four in premature infants. Two of six infected subclavian catheters were treated successfully with vancomycin. Infection was successfully cleared in 20 of 23 infected Broviac catheters with vancomycin through the line. However, six were eventually removed for tract infection (1), persistent fever (2), and Candida sp. infection (3). Although once considered a non-pathogenic skin contaminant, S. epidermidis has emerged as a serious pathogen in hospitalized, immunosuppressed, premature and malnourished pediatric patients. Indwelling catheters enhance the likelihood of infection in these patients. Aggressive antimicrobial therapy is vital in this potentially lethal infection. Vancomycin proved efficacious in this series.

Comparison of fluorescence polarization immunoassay and bioassay of vancomycin

Human serum samples were analyzed for vancomycin concentrations by two different methods: the fluorescence polarization immunoassay and the disk plate bioassay. Each assay method offered acceptable precision. The correlation between both assay methods was excellent (correlation coefficient = 0.985). Excluding technical time, the bioassay was the least expensive method to perform but was more labor intensive than the fluorescence polarization immunoassay.

Pharmacokinetics of vancomycin in patients undergoing continuous ambulatory peritoneal dialysis

The pharmacokinetics of vancomycin were studied in four patients on continuous ambulatory peritoneal dialysis. After a single intravenous infusion of 10 mg/kg of total body weight, multiple blood, urine, and dialysate samples were collected during a 72-h evaluation period. The steady-state volume of distribution was 0.73 +/- 0.07 (mean +/- standard deviation) liters/kg with a beta half-life of 90.2 +/- 24.2 h. The continuous ambulatory peritoneal dialysis clearance of vancomycin was 1.35 +/- 0.35 ml/min, and the serum clearance was 6.4 +/- 1.1 ml/min. Peritoneal dialysate concentrations of vancomycin were rapidly attained after the intravenous infusion and averaged 2.2 +/- 0.7 mg/liter throughout the 72-h observation period. A loading dose of 23 mg/kg followed by a maintenance dose of 17 mg/kg every 7 days should attain and maintain therapeutic serum and dialysate concentrations. More frequent dosing may be necessary for less susceptible organisms.

Pharmacokinetics of vancomycin in patients with various degrees of renal function

The pharmacokinetics of vancomycin were characterized in 56 patients with different degrees of renal function after an intravenous dose of 18.4 +/- 4.7 mg kg-1 (mean +/- standard deviation). Seven subjects had a creatinine clearance (CLCR) of greater than 60 ml min-1 (group I), 13 had a CLCR of 10 to 60 ml min-1 (group II), and 36 had a CLCR of less than 10 ml min-1 (group III). Serial serum samples (range, 3 to 8) were collected during the 168 h after drug administration. The serum concentration-time profile in all patients demonstrated monoexponential decay. The mean half-lives were 9.1, 32.3, and 146.7 h in groups I, II, and III, respectively. A significant decline in serum clearance (CLS) was also noted (62.7 to 28.3 to 4.87 ml min-1 in groups I, II, and III, respectively). The steady-state volume of distribution varied from 0.72 to 0.90 liter kg-1. There was no significant relationship between the steady-state volume of distribution and CLCR. The observed relationship between CLS and CLCR (CLS = 3.66 + 0.689 CLCR; r = 0.8807) can be utilized to devise dosage schedules for patients with any degree of renal impairment. This relationship was utilized to develop a nomogram for initial and maintenance dosing of vancomycin.

Evaluation of an automated fluorescence polarization immunoassay for vancomycin

An automated fluorescence polarization immunoassay for the determination of vancomycin levels in serum was evaluated. The vancomycin assay is a homogeneous competitive inhibition immunoassay based on changes in fluorescence polarization that occur with antibody binding. This assay was compared with a liquid chromatographic assay and an agar well diffusion bioassay method by using clinical serum specimens and controls. Linear regression analysis of the data obtained on clinical specimens by the three methods resulted in correlation coefficients of 0.97 for the fluorescence polarization immunoassay versus the liquid chromatographic assay (n = 60), 0.90 for the fluorescence polarization immunoassay versus the bioassay (n = 57), and 0.90 for the liquid chromatographic assay versus the bioassay (n = 57). Repetitive analysis of control sera containing 7, 35, and 75 micrograms of vancomycin per ml by the fluorescence polarization immunoassay yielded coefficients of variation of 3.0, 1.7, and 2.3, respectively. No interference was demonstrated in spiked hemolytic, lipemic, or icteric sera, and the assay was free of matrix effects. The automated fluorescence polarization immunoassay system offers a rapid, efficient, and accurate method for monitoring vancomycin serum levels for both toxicity and efficacy.

Vancomycin pharmacokinetics in patients with peritonitis on peritoneal dialysis

Vancomycin pharmacokinetics were studied in four patients with peritonitis undergoing chronic intermittent peritoneal dialysis. Serum levels exceeding 4.0 micrograms/ml were maintained for 8 and 13 days after a single 1-g intravenous dose. Vancomycin serum concentrations measured before, during, and upon completion of dialysis revealed no appreciable decline. Peritoneal fluid concentrations in two patients exceeded 4.0 micrograms/ml for more than 12 days.

Vancomycin

While vancomycin is thus not as nephrotoxic as once feared, its use by the parenteral route should be avoided if possible when other nephrotoxic drugs are being given. Used properly, vancomycin appears efficacious and can be given with safety to infants, children, and adults. Vancomycin is incompatible with many other drugs in intravenous solutions, especially chloramphenicol, adrenocorticosteroids, and methicillin.

In vitro evaluation of a vancomycin radioimmunoassay and observations on vancomycin pharmacokinetics in dialysis patients

A recently marketed radioimmunoassay (RIA) for vancomycin (Monitor Science Corp.) was evaluated in vitro and in vivo. The RIA and bioassay results on 15 spiked serum samples found no significant difference between the two assays, although RIA results were significantly less variable. The coefficient of variation (RIA) was 5.5%. Vancomycin showed no significant degradation in plasma at 4 degrees C and -20 degrees C for up to 36 days. Five dialysis patients had a total of 62 serum levels determined. When vancomycin 1 g was given weekly to two anuric adults, marked accumulation occurred. This finding was consistent with vancomycin's long terminal half-life (19.8 and 17.8 days) obtained from a new computer subroutine which obtains least square estimators of model parameters from multiple dose data. Despite serum levels well above the MIC, two patients remained bacteremic. Peritoneal levels of vancomycin in two patients with suspected peritonitis equalled corresponding serum levels.

Vancomycin therapy for serious staphylococcal infections in chronic hemodialysis patients

Vancomycin therapy during 7 episodes of serious staphylococcal infections in chronic hemodialysis patients was monitored by a sensitive bioassay technique. One gm of vancomycin was given during dialysis at a mean dosage interval of 7 days for a mean duration of 48 days. Serum peak and trough vancomycin levels were monitored during therapy. Accumulation of vancomycin occurred in 1 patient on prolonged therapy; progressive rising through levels required a reduction in vancomycin dosage. Pre and post-dialysis vancomycin levels in one patient were unchanged. Vancomycin was effective in eradication of all staphylococcal infections and bacteremias. Three A-V shunt infections required surgical revision; 2 A-V fistula infections were salvaged with vancomycin therapy alone. We conclude that 1 gm vancomycin every 7 days is an effective regimen for serious staphylococcal infections in chronic hemodialysis patients. Monitoring of vancomycin levels insures maintenance of adequate levels and prevents toxic accumulation.

Drug administration in renal failure

Renal failure impairs urinary excretion of drugs and may also modify drug action by alternations in protein binding, distribution, biotransformation and, possibly, by retention of active metabolites. Dialysis adds another variable by altering the blood levels of those drugs soluble in plasma water and therefore available for diffusion or ultrafiltration. Renal insufficiency clearly modifies decisions about the choice and dose of a wide variety of drugs. Although data are accumulating at a rapid rate, available information about the use of drugs in patients with kidney disease is rather limited. The following is a summary of recent information on the use of a variety of drugs frequently utilized in patients with impaired renal function. The guidelines presented here are not absolute, but they are intended to be practical and reasonable, based on current information for adult patients of average size with kidney disease.

Renal excretion of vancomycinin in kidney disease

The renal climination of vancomycin has been determined in 18 patients. In 4 anuric patients in intermittent haemodialysis the dosage of vancomycin necessary to treat infection with penicillin-resistantstrains of Staphylococcus aureus was determined. In 14 patients with varying degrees of renal insufficiency vancomycin, creatinine and 125-iothalamate clearances were measured and found to be closely correlated. After administration of the initial vancomycin dose and attainment of the serum concentration desired, the maintenance dose can be calculated on the basis of the GFR.

The flora of renal haemodialysis shunt sites

During investigations of the microbial flora of the skin over haemodialysis shunt sites it has not proved possible to predict clinical infection by a preceding colonization of the shunt site with a pathogenic organism. The normal non-pathogenic flora of the sites is not specifically related to the flora of other sites on the body though Staphylococcus aureus on a shunt site appeared to be acquired principally from the nose when the shunt was in the arm or from the perineum when the shunt was in the leg. Cimino shunt sites had a greater density of organisms than did Scribner shunt sites; this may be related to the disinfection procedures.

Cloxacillin and sodium fusidate in management of shunt infections

The effectiveness of two oral antibiotics, cloxacillin and sodium fusidate, has been evaluated in the treatment of shunt infections among 37 patients allocated at random to two treatment groups. Both proved to be safe bactericidal agents giving adequate serum M.I.C. when taken by mouth. Treatment should always be started on the basis of the clinical presentation without waiting for the bacteriologist's report. The commonest infecting organism is Staphylococcus aureus. Nine shunts were lost in this study, eight through Staphylococcus aureus infection. The nasal carrier state is of considerable importance in perpetuating these shunt infections.