Dialysis in acute renal failure: is more better?

An increasing body of evidence indicates therapy dose and intensity influence the outcome of dialyzed ARF patients. However, a number of unanswered questions remain on this issue. These questions need to be addressed in future prospective, controlled trials that assess the effect of dose and intensity on outcome both within and between the various ARF renal replacement therapies, with appropriate and clinically relevant control arms. Such investigations should provide guidelines ultimately for the dialytic management of critically ill patients with ARF.

Vancomycin assay performance in patients with end-stage renal disease receiving hemodialysis.

STUDY OBJECTIVE: To compare the performance of polyclonal fluorescence polarization immunoassay (pFPIA) with that of enzyme-multiplied immunoassay technique (EMIT) in patients receiving vancomycin and hemodialysis. SETTING: Outpatient hemodialysis center. PATIENTS: Seven subjects with end-stage renal disease treated with hemodialysis 3 times/week with a cellulose triacetate hemodialyzer. INTERVENTION: Subjects received vancomycin 1000 mg intradialytically during the first study session and 750 mg every other hemodialysis session thereafter for 4 weeks. MEASUREMENTS AND MAIN RESULTS: Blood samples were obtained throughout the study, and vancomycin serum concentrations were determined by pFPIA and EMIT. The mean +/- SD difference (estimate of bias) between assays was -1.10 +/- 1.35 mg/L. The limits of agreement (mean difference +/- 1.96 x SD) between them were -3.80-1.60 mg/L. CONCLUSION: Our data suggest that the manufacturer's changes in the vancomycin pFPIA eliminated overestimation of drug concentrations in patients undergoing high-permeability hemodialysis.

Cue-induced changes in basal local cerebral glucose utilization 13 days after morphine sensitization in the Fischer 344 rat: relevance for drug craving.

The present experiment tested the hypothesis that some persistent neural adaptation develops during the course of repeated sensitizing doses of morphine administered to rats. A sub-hypothesis was that this imprint would be of greater magnitude in the presence of morphine-conditioned cues. In order to test these hypotheses basal local cerebral metabolic rates for glucose (LCMR(glu)) were determined 13 days after the last of four 10-mg/kg doses of morphine administered in 36 h to Fischer 344 male rats. LCMR(glu) was determined using the 2-deoxy-D-[1-(14)C]glucose method (2-DG). Half of the rats, the conditioned group, were placed in the 2-DG chamber after each injection and half, the nonconditioned group, were placed in a neutral environment. A control group received only saline in lieu of morphine. All metabolic rates were determined in a nondrugged state. The major finding was large increases in metabolic rate throughout the forebrain in the sensitized rats. This was especially so in the conditioned group, 46 out of 93 areas examined had significant increases while in the nonconditioned group it was 25 out of 93. Both the core and shell of the nucleus accumbens showed significant elevations in metabolic rates in the presence of morphine cues but only the shell in the absence of the cues. There were no significant decreases in basal metabolic activity in any of the brain regions evaluated in either experimental group. The present finding suggests that changes in the brains of these morphine-sensitized rats may model the altered brain states responsible for drug craving in human drug addicts.

Small and middle molecular weight solute clearance in nocturnal intermittent peritoneal dialysis.

OBJECTIVES: To determine the dialysate-to-plasma (D/P) concentration ratios and peritoneal dialytic clearance (CI(D)) of substances with a wide range of molecular weights in subjects receiving a simulated nocturnal intermittent peritoneal dialysis (NIPD) session. DESIGN: Open-label single-dose study. SUBJECTS: Six end-stage renal disease patients undergoing peritoneal dialysis (PD). SETTING: Clinical research center of a university-affiliated hospital. INTERVENTIONS: Subjects received intravenous gentamicin and vancomycin on the first day of the study. Subjects received no PD until their return on the following day, when subjects underwent a simulated NIPD session utilizing four 2- to 2.5-L peritoneal dialysate dwells of 2 hours. Blood and dialysate samples were collected immediately before the session and after each dialysate dwell for determination of urea, creatinine, gentamicin, vancomycin, and beta2-microglobulin (beta2M) concentrations. Each solute's D/P concentration ratio and peritoneal CI(D) were calculated. MEASUREMENTS AND MAIN RESULTS: The (mean +/- SD) 2-hour D/P concentration ratios were 0.78 +/- 0.05 (urea), 0.49 +/- 0.11 (creatinine), 0.38 +/- 0.08 (gentamicin), 0.11 +/- 0.06 (vancomycin), and 0.07 +/- 0.03 (beta2M). Peritoneal CI(D) values (mL/min of dialysis) were 19.0 +/- 2.8 (urea), 12.1 +/- 3.5 (creatinine), 8.4 +/- 2.8 (gentamicin), 2.7 +/- 1.5 (vancomycin), and 1.7 +/- 0.8 (beta2M). The D/P concentration ratios and peritoneal CI(D) values for urea, creatinine, and gentamicin were significantly different from vancomycin and beta2M (repeated measures ANOVA, p < 0.05). Beta2-microglobulin peritoneal CI(D) was strongly related to gentamicin peritoneal CI(D) (r = 0.96, p < 0.05). CONCLUSION: Small molecular weight solutes have significantly greater D/P and peritoneal CI(D) than middle molecular weight solutes in NIPD. In NIPD, daily peritoneal CI(D) of beta2M is lower than that reported in continuous ambulatory PD. NIPD also results in lower drug CI(D) than that reported in continuous ambulatory PD studies.

Quantification of creatinine kinetic parameters in patients with acute renal failure.

BACKGROUND: Urea kinetic modeling (UKM) and creatinine (Cr) kinetic modeling (CKM) are used in the nutritional evaluation of end-stage renal disease (ESRD) patients. Both the UKM-derived normalized protein catabolic rate (nPCR) and the CKM-derived estimate of lean body mass (LBM) may also provide important information in critically ill acute renal failure (ARF) patients. Estimation of LBM may be particularly useful as previous data demonstrate that malnutrition adversely influences outcome in ARF patients. METHODS: Eleven critically ill ARF patients (age 52 +/- 21 years; mean +/- SD) treated with continuous venovenous hemofiltration (CVVH) were the study group. They were analyzed at steady state with a single-pool variable-volume model that determined the creatinine generation rate (GCr) by a methodology that we have previously described. RESULTS: The CVVH ultrafiltrate production rate was 913 +/- 49 ml/hr, yielding a blood Cr clearance of 15.2 +/- 0.9 ml/min and a steady state serum Cr of 3.4 +/- 1.7 mg/dl. Daily creatinine generation normalized to body wt (creatinine index: CI) was 6.3 +/- 0.8 and 10.6 +/- 3.0 mg/kg/day for females (N = 4) and males (N = 7), respectively (P < 0.05). Estimated mean LBM was 30.0 +/- 2.0 and 41.2 +/- 7.0 kg in females and males, respectively (P < 0.05), while the same parameter normalized to body wt was 0.50 +/- 0.05 and 0.52 +/- 0.10, respectively. These values are substantially lower than those previously reported for both normal and ESRD patients. Regression analysis demonstrated both GCr (r2 = 0.96; P < 0.001) and LBM (r2 = 0.96; P < 0.001) were significantly correlated with steady state serum Cr in a linear manner. However, no significant correlation (r2 = 0.06; P = 0.24) between nPCR and CI was observed. CONCLUSIONS: These data suggest critically ill ARF patients have severe somatic protein depletion. This malnourished state is likely due to deficits established prior to the development of ARF, such as those secondary to underlying chronic illnesses or prolonged hospitalization, and deficits related to acute hypercatabolism. Quantitative assessment of malnutrition in ARF patients with this CKM-based methodology may permit a better understanding of predisposing factors and, consequently, facilitate the development of interventions designed to prevent malnutrition in these patients.

Renal replacement therapy quantification in acute renal failure.

Recent results suggest that RRT delivery affects outcome in critically ill ARF patients. These data have generated interest in the use of RRT quantification methods, originally developed for ESRD patients, in ARF. However, the fundamental differences between ARF and ESRD, with respect to both patient and therapy characteristics, must be fully appreciated before making this extrapolation. These differences may render many of the simplified ESRD quantification formulae of little use in ARF. As is the case in ESRD, the use of clearance-based methods to compare disparate therapies is problematic in ARF. Although the optimal technique for RRT quantification in ARF remains to be defined, dialysate-side quantification may be the most rational approach for the future, as has been suggested for ESRD patients [43].

Safety and efficacy of pulse and daily calcitriol in patients on CAPD: a randomized trial.

BACKGROUND: Calcitriol therapy is the mainstay of therapy for the treatment of secondary hyperparathyroidism. Oral administration of calcitriol is necessary in CAPD patients, but no studies have directly compared different routes of administration in this patient population. METHODS: To determine if the peak serum calcitriol level (pulse therapy) is more important than the total delivered dose, we randomized CAPD patients with mild to moderate secondary hyperparathyroidism to receive either pulse (3.0 microg twice a week, n = 10) or daily (0.75 microg a day, n = 8) oral calcitriol in comparable weekly doses. The main comparison was the rate of decline of serum intact parathyroid hormone (PTH) levels to reach the desired end-point of 100 pg/ml. The patients were dialysed with low-calcium dialysate and received only calcium-containing phosphate binders. RESULTS: Pharmacokinetic analysis after a single dose of 3.0 microg (pulse) vs 0.75 microg (daily) revealed 1,25(OH)2-vitamin D levels to be higher in the pulse group at 3 and 6 h, but equivalent by 12 h. The area under the curve for 1 week of daily and 1 week of pulse therapy was equal. The patients in the 2 arms had equivalent basal serum levels of PTH (pulse = 562 +/- 291 vs daily = 454 +/- 113 pg/ml), calcium (pulse = 2.32 +/- 0.20 vs daily = 2.32 +/- 0.12 mmol/l) and phosphorus (pulse = 1.32 +/- 0.52 vs daily = 1.35 +/- 0.26 mmol/l). The time required for the PTH to decrease to 100 pg/ml and the rate of decline in PTH were similar (time: pulse = 14.2 +/- 6.8 weeks, daily = 12.2 +/- 7 weeks; rate: pulse = 7.4 +/- 4.2 vs daily = 8.4 +/- 4.2% PTH/week; P = NS). The serum calcium increased similarly in both groups. Hypercalcaemia (> 2.9 mmol/l) was rare (pulse = 3, daily = 2 episodes). CONCLUSIONS: This study demonstrates that pulse and daily calcitriol are similarly effective and safe for the treatment of mild to moderate secondary hyperparathyroidism in CAPD patients despite higher peak levels of 1,25(OH)2-vitamin D with pulse therapy.

Tunneled hemodialysis catheters: use of a silver-coated catheter for prevention of infection--a randomized study.

PURPOSE: To determine whether silver-coated tunneled hemodialysis catheters reduce infection and to determine the frequency of central venous thrombosis and stenosis with percutaneous placement of right internal jugular vein dialysis catheters by interventional radiologists. MATERIALS AND METHODS: Ninety-one patients were randomly assigned to a treatment (silver-coated catheter; n = 47) or control (identical catheter without silver coating; n = 44) arm. Baseline venography was performed. Catheter tips were cultured and venography was repeated at catheter removal. RESULTS: Mean duration of catheter placement was 92 days. Infection occurred in 11 patients (five in the treatment group, six in the control group). Tip cultures in 15 patients (eight treatment, seven control) were positive without clinical infection. Infection and colonization rates were slightly but not significantly higher in the treatment group than in the control group. Silver-coated catheters in two (4%) patients were removed due to reaction to the coating. Completion venograms (n = 72) showed new minor abnormalities in four (6%) patients and major abnormalities (stenosis, thrombosis) in three (4%) patients. Permanent venous abnormalities occurred in two (3%) patients. CONCLUSION: Silver coating does not confer a benefit against clinical infection or colonization. Interventional radiologic placement of tunneled dialysis catheters yields a low frequency of permanent central venous thrombosis and stenosis.

The role of renal replacement therapy quantification in acute renal failure.

The recognition that both morbidity and mortality are inversely related to delivered hemodialysis (HD) dose in end-stage renal disease (ESRD) patients has substantially changed clinical practices in the United States. A number of quantification techniques, which differ greatly in complexity and sophistication, are now used in ESRD patients. Investigators recently have attempted to extrapolate some of these ESRD quantification methods to the acute renal failure (ARF) setting. This review focuses on these recent attempts. Both patient-related and renal replacement therapy (RRT)-related differences in ESRD and ARF are discussed. In addition, the potential pitfalls of extrapolating certain ESRD quantification methods to RRT in ARF are discussed. Prescription considerations for both intermittent HD (IHD) and continuous RRT (CRRT) are presented. The optimal technique for RRT quantification in ARF remains to be determined.

In vitro cytotoxicity of silver-impregnated collagen cuffs designed to decrease infection in tunneled catheters.

PURPOSE: To examine in vitro the effects of silver-impregnated collagen cuff material from central venous catheters on human fibroblast growth. MATERIALS AND METHODS: In culture flasks, hybrid cells were exposed to silver-impregnated collagen cuff material, and human fibroblasts were exposed to silver-impregnated or silver-free collagen cuff material. After 72 hours of growth, cells were stained and digitally imaged, and the relative areas of cytotoxicity were determined. RESULTS: Flasks containing the silver-impregnated collagen cuff material and hybrid cells or human fibroblasts showed a marked local cytotoxic effect of the cuff material; cell-free zones surrounding the cuff material were demonstrated. No cytotoxic effect was seen in the flasks that contained silver-free cuff material (control group). Mean area of cleared cells was 312 mm2 +/- 130 (range, 156-624 mm2) in the flasks containing human fibroblasts and silver-impregnated cuff material and 0 mm2 in the corresponding control flasks (P < .0001). Mean radius of the area of cleared cells around the silver-impregnated cuff material in the flasks containing human fibroblasts was 9.8 mm +/- 2.0 (range, 7.0-14.1 mm). CONCLUSION: Silver-impregnated collagen cuff material demonstrates a local cytotoxicity on hybrid cells and human fibroblasts in vitro. This finding may explain the phenomena seen clinically of decreased anchorage and inadvertent removal of catheters with silver-impregnated collagen cuffs.

Persistent increases in basal cerebral metabolic activity induced by morphine sensitization.

To characterize the underlying neuroanatomic substrate of morphine (MS) sensitization, changes in the local cerebral metabolic rate for glucose (LCMRglu) were examined in 95 brain regions of male F-344 rats using the 2-deoxy-D-[1-14C]glucose method. The results of these experiments demonstrate that MS-induced sensitization is manifested by increases in basal metabolic activity that last for at least 6 days. Although changes in basal metabolic rate were found to be more extensive in the presence of conditioned cues, the increases in LCMRglu in nonconditioned sensitized rats indicate a basic underlying pharmacologic effect of MS sensitization on basal brain activity. Regions in which MS sensitization had a lasting pharmacologic effect include the shell of the nucleus accumbens, the prelimbic area of the prefrontal cortex, and the dorsolateral prefrontal cortex. Interestingly, the core of the nucleus accumbens and regions of the caudate were found to have an increased LCMRglu only in the presence of conditioned cues, indicating conditioned brain activity without observable changes in behavior. The previous administration of an MS-sensitizing treatment was also found to alter the cerebral metabolic response to a subsequent acute MS challenge (0.5 mg/kg, subcutaneously), most notably in forebrain systems. The more widespread activation of brain structures in the basal state in the presence of conditioned cues suggests that these MS-sensitized rats may model an altered brain state related to craving in the abstinent opiate addict.

Extracorporeal therapy requirements for patients with acute renal failure.

Renal replacement therapy (RRT) requirements for critically ill patients with acute renal failure (ARF) depend on numerous factors, including the degree of hypercatabolism, patient size, and desired level of metabolic control. However, the current practice at many institutions is to prescribe generally similar amounts of RRT to ARF patients essentially without regard for the above factors. In this study, a computer-based model designed to permit individualized RRT prescription to ARF patients was developed. The critical input parameter is the desired level of metabolic control, which is the time-averaged BUN (BUNa) or steady-state BUN (BUNs) for intermittent hemodialysis (IHD) or continuous RRT (CRRT), respectively. The basis for the model was a group of 20 patients who received uninterrupted CRRT for at least 5 days. In these patients, the normalized protein catabolic rate (nPCR) increased linearly (r = 0.974) from 1.55 +/- 0.14 g/kg per day (mean +/- SEM) on day 1 to 1.95 +/- 0.15 g/kg per day on day 6. The daily urea generation rate (G), determined from the above linear relationship, was utilized to produce BUN versus time curves by the direct quantification method for simulated patients of varying dry weights (50 to 100 kg) who received variable CRRT urea clearances (500 to 2000 ml/h). Steady-state BUN versus time profiles for the same simulated patient population treated with IHD regimens (K = 180 ml/min, T = 4 h) of variable frequency were generated by use of a variable-volume, single-pool kinetic model. From these profiles, regression lines of required IHD frequency (per week) versus patient weight for desired BUNa values of 60, 80, and 100 mg/dl were obtained. Regression lines of required CRRT urea K (ml/h) versus patient weight for desired BUNs values of 60, 80, and 100 mg/dl were also generated. For the attainment of intensive IHD metabolic control (BUNa = 60 mg/dl) at steady state, a required treatment frequency of 4.4 dialyses per week is predicted for a 50-kg patient. However, the model predicts that the same degree of metabolic control cannot be achieved even with daily IHD therapy in patients > or = 90 kg. On the other hand, for the attainment of intensive CRRT metabolic control (BUNs = 60 mg/dl), required urea clearance rates of approximately 900 ml/h and 1900 ml/h are predicted for 50- and 100-kg patients, respectively. This model suggests that, for many patients, rigorous azotemia control equivalent to that readily attainable with most CRRT can only be achieved with intensive IHD regimens. Following prospective clinical validation, this methodology may be a useful RRT prescription tool for critically ill ARF patients.

Outcome of tunneled hemodialysis catheters placed via the right internal jugular vein by interventional radiologists.

PURPOSE: To assess the outcome of interventional radiologic placement of tunneled hemodialysis catheters via the right internal jugular vein. MATERIALS AND METHODS: In 194 patients, the catheter was placed via the right internal jugular vein unless thrombosis was present. Real-time ultrasound-guided puncture and fluoroscopic guidance were used. Patients were followed up until catheter removal or death. Outcomes evaluated included infection, thrombosis, and catheter malfunction. RESULTS: In 175 patients, 250 consecutive catheters were placed via the right internal jugular vein with 100% success. All catheters functioned immediately after placement. Procedural complications were limited to clinically unimportant air embolus (n = 2). No instances of pneumothorax, hemothorax, or substantial bleeding complications occurred. Follow-up was available in 173 (99%) patients. Mean and median "catheter duration" were 87 and 56 days, respectively. Catheter-related symptomatic venous thrombosis or stenosis was not observed. The rate of infection was 0.08 per 100 catheter days, and the rate of malfunction that necessitated removal was 0.22 per 100 catheter days. Definite or possible catheter thrombosis that necessitated removal occurred at a rate of 0.16 per 100 catheter days. CONCLUSION: Interventional radiologic placement of tunneled hemodialysis catheters via the right internal jugular vein showed equal or better long-term results than those reported for surgical placement. Interventional radiologic placement should be the method of choice.

Dialysis prescription and kinetics in acute renal failure.

The recognition that both morbidity and mortality are inversely related to delivered hemodialysis (HD) dose in end-stage renal disease (ESRD) patients has substantially changed clinical practices in the United States. A number of quantification techniques, which differ greatly in complexity and sophistication, are now used in ESRD patients. Investigators recently have attempted to extrapolate some of these ESRD quantification methods to the acute renal failure (ARF) setting. This review focuses on these recent attempts. Both patient-related and renal replacement therapy (RRT)-related differences in ESRD and ARF are discussed. In addition, the potential pitfalls of extrapolating certain ESRD quantification methods to RRT in ARF are discussed. Prescription considerations for both intermittent HD (IHD) and continuous RRT (CRRT) are presented. Finally, recent data suggesting survival in critically ill ARF patients is directly correlated with delivered therapy dose are reviewed. The optimal technique for RRT quantification in ARF remains to be determined.

Effects of dialysis membrane on intradialytic vancomycin administration.

STUDY OBJECTIVE: To quantify the influence of hemodialyzers on vancomycin removal when the drug was infused during hemodialysis. DESIGN: Prospective, controlled, crossover study with three arms. SETTING: A university-affiliated medical center. PATIENTS: Eight subjects receiving outpatient hemodialysis. INTERVENTIONS: The three treatment arms were vancomycin 1000 mg infused after dialysis was completed (control), and the same dosages infused during the last hour of hemodialysis with a cellulose triacetate (CT) and a cellulose acetate (CA) hemodialyzer. MEASUREMENTS AND MAIN RESULTS: The areas under the curve from time zero to 44 hours (AUC0-44 hrs) for the three study arms were significantly different (p < 0.05), with the mean vancomycin AUC0-44 hrs being significantly lower when administered during CT and CA dialysis (73.7% and 87.2% of control; p < 0.05 vs control). The mean vancomycin peak concentration achieved during CT dialysis was significantly lower than for the CA and control arms (20.5, 23.9, 27.0 mg/L, respectively). Forty-four-hour postinfusion concentrations were similarly lower. CONCLUSION: Clinicians should recognize that the composition of the hemodialyzer significantly influences vancomycin serum concentrations when the drug is administered during hemodialysis.

Sleep apnea in renal failure.

Sleep apnea is a surprisingly common disorder in end-stage renal disease (ESRD) and chronic renal failure. The symptoms of sleep apnea frequently go unreported or may be misdiagnosed as uremia, depression, chronic illness, or insomnia. A review of the literature was performed to define the prevalence, morbidity, and treatment of sleep apnea syndrome in the ESRD patient. Sleep apnea occurs in at least 60% of ESRD patients. The known complications of sleep apnea include arrhythmias, pulmonary hypertension, and systemic hypertension. In addition, sleep apnea has been implicated in coronary artery disease and strokes. The contribution of sleep apnea to the high mortality from cardiac disease and stroke in peritoneal dialysis and hemodialysis patients is unknown. The causes of the increased prevalence of sleep apnea in ESRD patients are unknown and likely differ from the general population, but the treatment is similar. The literature suggests that modality of renal replacement therapy does not matter; however, large nocturnal volume peritoneal dialysis may worsen sleep apnea. Renal transplantation may be curative. In conclusion, sleep apnea may be an under-diagnosed disease in patients on dialysis. There are significant reasons to suspect that sleep apnea may worsen the morbidity and mortality of ESRD, and there are potential successful therapies.

Naloxone alters the local metabolic rate for glucose in discrete brain regions associated with opiate withdrawal.

The current 2-deoxy-D-[1-14C]glucose investigation was performed to test the hypothesis that endogenous opioids influence basal synaptic activity within discrete brain regions. To examine this hypothesis, the effects of naloxone (1.0 mg/kg s.c.) on local cerebral metabolic rate for glucose (LCMRglu) in 84 brain regions were compared to saline controls. The specificity of naloxone's effects for opioid receptors was assessed by the coadministration of the opiate agonist morphine in a separate group. In naloxone-treated rats, there was a significant decrease in LCMRglu in the locus coeruleus (LC) and an increase in the central nucleus of the amygdala (CAMY), supporting a tonic influence of endogenous opioids on these regions. These metabolic changes were reversed by coadministered morphine, indicating that naloxone's metabolic actions are specific for opioid receptors. Based on the role of the LC and CAMY in opiate withdrawal, the present results suggest a subthreshold naloxone precipitated withdrawal from endogenous opioids. Although morphine administered alone significantly reduced LCMRglu in 16 brain regions, these did not include the LC or the CAMY. These results identify brain regions in which synaptic activity is under tonic modulation by endogenous opioids.

Establishing a laparoscopic antireflux program: a private practice experience.

Laparoscopic antireflux surgery is an exciting new option in the treatment of gastroesophageal reflux disease. Establishing a successful antireflux program depends on obtaining proper education as well as generating a suitable flow of patients. Being outside a university-based referral center presents the surgeon in private practice with a unique challenge as to how to meet these goals effectively. A personal account of education, training, and patient recruitment methods is presented. Different modalities for stimulating interest in both the medical and lay communities, including methods for direct patient contact, such as advertising are described. This is followed by a review of a series of 57 antireflux procedures performed during the first year of this program in a private practice in a retirement community setting.

Falsely elevated serum vancomycin concentrations in hemodialysis patients.

Fluorescence polarization immunoassay (FPIA) is the most widely used clinical vancomycin assay in the United States. Questions exist regarding the accuracy of this polyclonal assay in patients with end-stage renal disease (ESRD). While several studies have reported discrepancies in vancomycin serum concentrations determined by FPIA compared with other vancomycin assays, no study has investigated the accuracy of vancomycin serum concentrations determined by FPIA in patients with ESRD undergoing maintenance hemodialysis. Therefore, we compared the assay performance of FPIA and enzyme multiplied immunoassay technique (EMIT) in six subjects with ESRD receiving high-efficiency hemodialysis. Subjects underwent 6 consecutive weeks of hemodialysis treatment with a cellulose acetate dialyzer (CA210) and received 1 g vancomycin intravenously once weekly during the last hour of dialysis. Vancomycin serum concentrations were determined by both EMIT and FPIA methodologies. From the serum concentration results of both assays, vancomycin dosing recommendations were calculated to achieve a desired steady-state peak concentration of 35 mg/L and trough concentration of 10 mg/L. Overall, vancomycin serum concentrations reported by FPIA were significantly higher than those reported by EMIT. The mean difference between assays in the peak serum concentrations at weeks 1, 4, and 6 was 7.5, 11.5, and 11.2 mg/L, respectively. The mean difference in trough serum concentrations at weeks 1, 4, and 6 was 4.2, 6.2, and 5.2 mg/L, respectively. The FPIA overestimation of the EMIT values (calculated as FPIA-EMIT) varied widely among study subjects with a range of 0.0 mg/L to 27.0 mg/L for peak serum concentrations and 0.0 mg/L to 12.8 mg/L for trough serum concentrations. The mean doses calculated based on FPIA results were significantly lower than the EMIT-derived doses. No significant difference was observed in the calculated dosing intervals. These results demonstrate that FPIA significantly overestimates vancomycin serum concentrations compared with EMIT in patients with ESRD undergoing high-efficiency hemodialysis. The overestimation by FPIA may result in significantly different vancomycin dosing recommendations, leading to underdosing and the potential for therapeutic failures. Due to the unpredictability of the overestimation by FPIA, we were unable to formulate vancomycin dosing guidelines for institutions that use FPIA. Therefore, we recommend that the EMIT vancomycin assay be used in patients with ESRD to ensure appropriate dosing.