Strategies to increase the use of autogenous arteriovenous fistula in end-stage renal disease.

The purpose of this retrospective study was to review our experience with a consecutive group of end-stage renal disease (ESRD) patients using simple strategies to increase the use of autogenous vascular access, and determine whether the current trend of using synthetic bridge-graft fistula (BGF) rather than autogenous arteriovenous fistula (AVF), could be reversed, despite an aging population and broadening criteria for hemodialysis. All patients for vascular access surgery had careful preoperative clinical examination of the arm veins with outflow occlusion to determine the venous anatomy and continuity. Where no veins were apparent or their continuity in doubt, selective preoperative venography was performed. Where veins were unsatisfactory for forearm AVF, new or modified surgical procedures to use both the basilic and cephalic veins in the upper arm were performed. Intraoperative angioscopy was used to monitor vein quality and surgical technique. Ninety-eight primary vascular access procedures were performed in 76 patients, 75 (76.5%) AVF (forearm, n = 41; upper arm, n = 34) and 23 (23.5%) BGF. Forty-one of 76 (54%) had already had at least one previous access procedure prior to this study. More than one access procedure was needed in 16 patients. Preoperative venography was performed in 22 (22.4%) and intraoperative angioscopy in 45 (45.9%) of the 98 procedures. The number of revisions required to maintain patency was significantly higher for BGF (37 revisions in 14/23) than AVF (16 revisions in 13/75) (p < 0.0001, Poisson test) with an annualized secondary revision rate of 1.168 for BGF and 0.173 for AVF (p < 0.0001, Poisson test). AVF had both longer primary (p = 0.0001, log rank test) and secondary patency (p = 0.038, log rank test) than BGF. AVF as the primary vascular access can be significantly increased and the current trend of using BGF reversed with the use of simple clinical strategies to evaluate the suitability of the arm veins for vascular access.

Nutritional assessment with bioelectrical impedance analysis in maintenance hemodialysis patients.

Protein energy malnutrition is common among persons with ESRD and contributes substantially to morbidity and mortality. The usual methods of nutritional assessment, such as anthropometry, can be misleading because of altered tissue hydration. Bioelectrical impedance analysis (BIA) has been recommended by some as a practical nutritional assessment tool but has not been validated in patients with ESRD. Thirty-three stable patients on maintenance hemodialysis were evaluated in an ambulatory clinical research center with simultaneous BIA, dual-energy x-ray absorptiometry, and deuterium oxide (D2O) and sodium bromide (NaBr) isotope dilution studies. Standard determinations of total body water (TBW) and body cell mass (BCM) were obtained and compared with values estimated by BIA. Two separate outpatient BIA measurements were also obtained approximately 2 wk before and after the clinical research center evaluation. BCM estimated by BIA was directly correlated (r = 0.92, P < 0.0001) with BCM determined by DEXA and NaBr. TBW estimated by BIA was directly correlated (r = 0.96, P < 0.0001) with TBW determined by deuterium oxide dilution. The reactance to resistance ratio (Xc/R) derived from BIA was inversely correlated (r = -0.73, P < 0.0001) with the extracellular water/TBW ratio determined by NaBr/D2O. Bland-Altman analyses showed that for TBW, BIA was in excellent agreement with D2O dilution. BCM was modestly underestimated by BIA compared with the dual-energy x-ray absorptiometry/NaBr standard and was adjusted by linear regression. The coefficients of variation on repeated BIA measurements were below 4%, demonstrating test-retest reliability. BIA is a valid and reliable method of nutritional assessment in maintenance hemodialysis patients.

The role of routine angioscopy in vascular access surgery.

PURPOSE: The purpose of this retrospective study is to describe our techniques, review our experience, and determine the feasibility, safety, and role of the routine use of angioscopy during primary and revision vascular access surgery. METHODS: Between February 1991 and October 1993, intraoperative angioscopy was routinely performed in 84 consecutive operations (51 patients) for vascular access surgery. We reviewed the videotaped recordings of the angioscopic studies together with the clinical data according to a predetermined protocol. RESULTS: There were 43 primary procedures (36 autogenous arteriovenous fistulas and 7 bridge graft fistulas) and 41 revision procedures for failed vascular access (7 autogenous arteriovenous fistulas and 34 graft bridge fistulas). In 20.9% of the primary vascular access procedures, abnormal endoluminal findings were noted. Based on these findings, only one additional intervention was performed. In revision vascular access surgery, abnormal endoluminal findings were noted in 92.7%, resulting in additional surgical interventions in 65.9% of the procedures. In the revised synthetic bridge graft fistulas, stenosis of the midgraft (n = 9) as a result of needle insertion for dialysis was more common than at venous anastomosis (n = 4). Detection and correction of endoluminal abnormalities resulted in a 30-day patency of 66.6% as opposed to 33.3% when none was detected (p < or = 0.012, Fisher's exact test). CONCLUSIONS: Routine angioscopy is technically feasible and can be performed safely in anuric patients during vascular access surgery. It provides additional and useful intraoperative information that may significantly alter the surgical procedure. Routine angioscopy may also provide new insights into the pathophysiology of vascular access failure.

Effects of acetate and bicarbonate dialysate in stable chronic dialysis patients.

The effects of acetate and bicarbonate dialysate on the biochemical and clinical parameters of 16 stable chronic hemodialysis patients were investigated in a double-blind crossover study. A central delivery system was used for both types of dialysates with identical sodium concentrations (138 mEq/liter) and osmolality in a single-pass dialysate flow. The results indicate that dialysis with bicarbonate leads to significantly less hypoxemia (P less than or equal to 0.001) and hypotensive episodes (P less than or equal to 0.002) than with acetate. Pre- to post-dialysis blood pressure changes were also more marked during acetate dialysis. Older patients with recurrent hypotension on acetate benefit most from bicarbonate dialysate. This group of patients appears to metabolize acetate more slowly and has a significantly lower post-dialysis bicarbonate concentration (P less than or equal to 0.005) than asymptomatic patients during dialysis with acetate dialysate.

Bacteremic infection in hemodialysis.

This is a retrospective study of 133 episodes of bacteremic infection in 112 hemodialysis patients. The frequency of bacteremic infection was 9.5% in patients with chronic renal failure and 10.9% in patients with acute renal failure. In patients with acute renal failure, pneumonia and intra-abdominal abscess were the most frequent sources of septicemia. Sepsis was usually due to Gram-negative organisms and mortality was high. In patients with chronic renal failure, infection of the shunt or fistula was the most common cause, was frequently due to Staphylococcus organism, and had a more favorable survival rate. Gram-negative septicemia from a nonaccess source in patients with chronic renal failure was associated with a higher mortality. Bacterial endocarditis and septic pulmonary emboli occurred in 3.6% of septic episodes and 0.35% of patients at risk and had very low mortality. A low threshold for obtaining blood cultures and early antibiotic treatment are believed to be important in the treatment of bacteremic infections in patients undergoing long-term hemodialysis.

Vestibular toxicity of gentamicin. Incidence in patients receiving long-term hemodialysis therapy.

Twenty-three patients on long-term hemodialysis regimens who received gentamicin sulfate were reviewed retrospectively to assess the incidence of ototoxicity and to identify potential risk factors. Dosage of gentamicin sulfate was 1.0 to 1.5 mg/kg intravenously three times weekly. Serum gentamicin levels were monitored in 21 cases. Seven patients developed signs and symptoms of vestibular dysfunction. Statistically significant differences were found between the ototoxic and nonototoxic groups with respect to age (P less than .001), total dose (milligrams per kilogram) (P less than .001), and duration of therapy (P less than .001). The total dose per kilogram of body weight contributed most heavily to ototoxicity, and regression analysis suggests that the critical cumulative dose is about 17.5 mg/kg. The two groups did not differ with respect to mean peak and valley serum levels. We conclude that this population is at high risk of developing gentamicin-related vestibular dysfunction specifically when the cumulative dose exceeds 17.5 mg/kg.

Initial clinical evaluation of the Hospal artificial kidney system.

A high flux artificial kidney system, which permits control over ultrafiltration, has been developed and patients are now being treated with it at a large outpatient kidney center. The basic operating characteristics have been described and some preliminary clinical results mentioned. Characteristics of the Hospital Artificial Kidney System are: 1. Higher clearances for middle molecules than conventional systems; 2. Adequate clearances for small solutes such as urea; 3. Predictable control of ultrafiltration with discrepancies between actual weight loss and theoretical weight loss on the order of 0.7 to 1.3 ml/min, when the system is operated in the designated manner.

A longitudinal study of bone disease after successful renal transplantation.

A retrospective study of 100 patients followed for 1-4 years after successful renal transplantation was undertaken to assess the amelioration of previously present metabolic bone disease and to determine the risk factors associated with the development of osteonecrosis. 42% of patients showed some evidence of bony abnormality. Following transplantation, there was slow but progressive resolution in the X-ray changes of hyperparathyroidism but not of osteoporosis. 14% of patients developed osteonecrosis in the posttransplant period with the femoral head the most common site involved (72% of patients). Osteonecrosis could not be related to average steroid dose, number of steroid pulses, or the preexistence of metabolic bone disease.

Normal renocortical blood flow in experimental acute renal failure.

Renal cortical blood flow of rats with postischemic, myohemoglobinuric, and mercury-induced acute renal failure was measured by the hydrogen washout technique using implanted platinum electrodes. Total renal blood flow was determined by venous cannulation in separate series of rats. The values obtained with the two methods were in excellent qualitative agreement (r=0.99, P less than 0.001), although venous cannulation gave values that were constantly lower than those calculated for whole kidney from the cortical flow rate and assumed cortical mass. Myohemoglobinuria produced by glycerol injection caused cortical blood flow to fall from a control value of 7.37+/-0.23 (SEM) ml/min X g of cortex to approximately one-half that value for four hours after injection (P less than 0.001). Flow rates 12 and 24 hr after glycerol injection were 85% (P less than 0.001) and 90% (P less than 0.05) of control, respectively. Cortical flow was reduced to 5.49+/-0.39 (SEM) ml/min X g of cortex four hours after release of one hour's total bilateral renal arterial occlusion (P less than 0.001), but rose to normal within 24 hr. Poisoning with 4.7 mg/kg of body wt of mercuric chloride produced a cortical blood flow value that was 30% higher than control 24 hr after injection (P less than 0.01), while a 12 mg/kg of body wt dose gave a normal flow value. Inulin clearance was severely depressed in all models at all study times. Thus, in contrast to human acute renal failure, marked renal cortical ischemia is not an essential feature of these different forms of murine acute renal failure.

The long-term follow-up of 195 patients with renal failure: a preliminary report.

Radiographic and bone mineral (BM) data were collected over a three-year period on 195 patients with chronic renal failure. Most women maintained BM on dyalysis, whereas 44% of the men lost BM (p less than 0.05). Following transplantation, 86% of the patients either maintained or restored BM. After parathyroidectomy, only half of the women and 34% of the men gained BM. Normal radiographs may be associated with low BM values, but there is a correlation between decreasing BM and increasing renal osteodystrophy in women (p less than 0.05).

Controlled ultrafiltration (UF) with hemodialysis (HD): analysis of coupling between convective and diffusive mass transfer in a new HD-UF system.

A convertible and versatile dialysis system consisting of a PMMA hollow fiber unit and UFR controller is shown to be useful for evaluating the effect of solute and fluid removal rates on adequacy of dialysis. Post-dilution mode where high UFR is applied to the patient with matching or comparable dilution rate is concluded to be most effective modification for enhancing the clearance for "middle molecules", without sacrificing the removal rates for the small solutes.

Factors in the dialysis regimen which contribute to alterations in the abnormalities of uremia.

While hemodialysis therapy in its present form is capable of sustaining life, dialysis patients are not metabolically normal and we are unable to say what technical factors contribute adequate therapy. Recent efforts to resolve these problems have led to the assumption that substances in the molecular weight range of 800 to 3000 daltons may be pathogenic in uremia and these may not be effectively removed by dialysis. Accordingly, four groups of patients (ten each) underwent changes in their routine which were theoretically designed to alter independently the concentration of small (urea) and "middle" molecules in the blood. In two groups, the concentration of urea was theoretically increased or decreased while the concentration of so-called middle molecules was maintained unchanged. In the remaining two groups, middle molecule concentration was theoretically increased or decreased while small molecule concentration was unchanged. Patients were evaluated prior to and after completing altered dialysis therapy. The results suggest three related conclusions. First, the uremic syndrome may be viewed as a constellation of abnormalities which can be subgrouped by association so that azotemia may be correlated with neuropathic disease and hypertension with weight gain or body size, for example. Second, those physiologic variables which changed after altered dialysis tended to deteriorate with increasing concentration of small molecules in the blood and remained independent of theoretical changes in middle molecules. Finally, when patients are relatively under-dialyzed, they may spontaneously modulate the reduced removal of metabolites such as urea by decreasing the dietary intake of nutrients.

Performance and clinical use of a convertible hemodialysis (HD) - ultrafiltration (UF) system.

A hemodialysis-ultrafiltration system has been developed and evaluated. It may be used either as a conventional hemodialyzer or as a nondialytic ultrafilter. When used as a dialyzer an ultrafiltration controller is required to fix ultrafiltration rate at the desired level. Rates of low molecular solute removal are compatible with those observed in commercially available artificial kidney systems, higher clearances are obtained for middle molecules, and the rate of ultrafiltration can be controlled within narrowly prescribed limits.

The estimation of blood flow to hemodialyzers by bubble time.

Reliable estimation of blood flow to hemodialyzers is essential to the accurate prescription of therapy. Customarily an air bubble is timed as it traverses a racetrack of known volume and blood flow is calculated from the formula Qb (ml/min) = (volume/BT) (60 sec/min). The linearity of this relationship has recently been questioned. Air bubble velocities were statistically fit to measured blood flows over 6 hematocrit values. From the resulting relationship, Qb (ml/min) = 62 (vol/L) (L/BT)0.96, one can generate a table relating velocity to flow for clinical use.