Adverse effects of dialyzers manifesting during the dialysis session.

The intradialytic symptoms that can be linked to components of the extracorporeal circuit of greatest clinical importance are the Type A (anaphylactoid) reactions. Most of these are IgE-mediated reactions due to ethylene oxide and are preventable by adequate degassing of the dialyzer by the manufacturer and by adequate rinsing of the dialyzer just prior to use. AN69-associated reactions are a second group, and are probably mediated by membrane-induced bradykinin generation coupled with ACE-inhibitor induced prolongation of bradykinin half-life. Type A reactions occur in a reuse setting, alos, and these may be related to some as yet poorly understood interaction between bleach, reuse sterilants, and certain dialyzer membranes, again, with ACE inhibitors playing an amplifier role. There is no compelling evidence linking membrane-induced complement activation to type A dialyzer reactions. However, there is a large body of evidence in animal models that exposure to complement fragment-releasing membranes can increase the pulmonary artery pressure and increase thromboxane formation. Thus, at least in principle, a case can be made for using unsubstituted cellulose membranes with caution in patients with a history of atopy or eosinophilia, particularly if acetate dialysate is to be used. Such a caution, however, must be viewed as conjectural in the absence of definitive evidence. Type B dialyzer reactions (mild back and chest pain 20-60 min into the dialysis session) is a phenomenon that is in the process of vanishing. The reason why is unclear. These reactions may have been due to some sort of dialyzer contaminant, or they may have been due to complement fragment release and required the use of acetate dialysate as a cofactor. In any event, recent well designed studies fail to find any differences in symptoms between unsubstituted cellulose and synthetic membranes. Membrane-induced complement fragment release also may play a minor role in dialysis hypoxemia, but evidence is conflicting in this area. Again, the use of acetate dialysate appears to be an important cofactor. Post-dialysis events which may be conceivably linked to the delayed effects of complement fragment releasing membranes need to be evaluated in controlled studies. Studies suggesting increased post-dialysis catabolism with use of unsubstituted cellulose membranes need to be confirmed in dialysis patients, and symptomatic correlates should be sought and evaluated.

A comparison of cutaneous testing and ELISA testing for assessing reactivity to ethylene oxide-human serum albumin in hemodialysis patients with anaphylactic reactions.

In five patients who have experienced anaphylaxis and in 29 patients who have not had such episodes during hemodialysis, we have performed two immunologic studies: cutaneous testing with ethylene oxide-human serum albumin (ETO-HSA) and ELISA for IgE against ETO-HSA. Four of five patients with reactions had positive cutaneous tests, whereas only one nonreactor had a positive skin test (p less than 0.0002). The same four of five patients with reactions also had positive ELISA results, whereas three nonreactors has positive ELISA results (p less than 0.003). In this group of patients, the positive predictive value of cutaneous testing (80%) is somewhat higher than that of ELISA testing (57%). However, the sensitivity, specificity, and negative predictive values are similar. We conclude that cutaneous testing with ETO-HSA probably offers a small advantage over IgE against ETO-HSA as determined by ELISA.

Plate, coil, and hollow-fiber cuprammonium cellulose dialyzers: discrepancy between incidence of anaphylactic reactions and degree of complement activation.

During the past 10 years, the incidence of severe anaphylactic reactions during dialysis [type A first-use syndrome (FUS)] at our center has been much lower when using cuprammonium cellulose plate (CC-P) dialyzers (0/37, 750 dialyses) or coil (CC-C) dialyzers (0/32, 500) than when using cuprammonium cellulose hollow-fiber (CC-F) dialyzers (8/21,022 dialyses, p less than 0.005 by Chi-square). To determine if the difference in type A FUS incidence between the three dialyzer types could be explained by differences in complement activation, we compared plasma concentrations of C3a des-arginine (des arg) in patients undergoing dialysis with these three varieties of dialyzers. Plasma C3a des arg values increased markedly in the dialyzer outflow blood with the three dialyzer configurations. The levels were similar with the dialyzer types when results were corrected for membrane surface area. Also, the degree of leukopenia was not markedly different with the three dialyzer types. Our findings suggest that complement activation per unit surface area is similar during dialysis with plate, coil, and hollow-fiber cuprammonium cellulose dialyzers. The lack of correlation between the degree of complement activation and the incidence of type A FUS suggests that membrane-induced complement activation is not of primary importance to type A dialyzer hypersensitivity reactions.

IgE against ethylene oxide-altered human serum albumin in patients with anaphylactic reactions to dialysis.

We have measured total antibody and IgE directed against ethylene oxide-altered human serum albumin (ETO-HSA) in the sera of 24 patients who have experienced anaphylaxis during hemodialysis and of 41 patients who have not had such episodes during hemodialysis. ETO is used to sterilize dialyzers and other medical equipment. The geometric mean level of IgE to ETO-HSA in patients with reactions (0.9 ng ETO-HSA bound to IgE per milliliter of serum) is significantly higher than in nonreacting patients (0.1 ng/ml, p less than 0.0001). Sixteen of 24 patients with reactions had detectable levels of IgE to ETO-HSA, whereas only three of 41 nonreacting patients had detectable levels (p less than 0.0001 chi-square). The geometric mean level of total antibody to ETO-HSA is also significantly higher in patients with reactions (270 ng ETO-HSA bound per milliliter) than in nonreacting patients (31 ng/ml, p less than 0.0001). Fourteen of 24 patients with reactions but only four of 39 nonreacting patients had total antibody binding of ETO-HSA (p less than 0.0001 chi-square). These data extend our previous observations on a small group of 13 patients receiving hemodialysis (seven patients with reactions, and six nonreacting patients) and clearly demonstrate an association between the presence of IgE or total antibody to ETO-HSA and immediate anaphylactic reactions in this group of 65 patients receiving hemodialysis.

Severe anaphylactoid reactions to cuprammonium cellulose hemodialyzers.

Twenty-one severe reactions to hemodialysis occurred in approximately 260,000 dialysis treatments at three centers within a 10 1/2-year period. Reactions typically appeared within minutes of initiating dialysis, and were characterized by cardiopulmonary, mucocutaneous, and/or gastrointestinal tract symptoms highly suggestive of anaphylaxis. Four respiratory arrests and one death resulted. Analysis of dialyzer use patterns and of each patient's dialyzer exposure history strongly implicated hollow-fiber dialyzers made of cuprammonium cellulose (CC) as a cause of these reactions. No obvious factors could be found to identify predisposed patients. Less than optimal rinsing of the CC hollow-fiber dialyzers prior to use may have been responsible for some, but not all, of these reactions.

Amantadine hydrochloride pharmacokinetics in hemodialysis patients.

To study the fate of amantadine hydrochloride in patients with renal failure, we gave 100 mg orally to 12 such patients immediately after hemodialysis. Plasma levels did not decrease between 24 and 44 hours after drug ingestion, suggesting an extremely poor total body clearance. Apparent volume of distribution was 5.1 +/- 0.2 (SEM) L/kg of body weight. Between 44 and 48 hours, as a result of 4 hours of hemodialysis, the mean plasma drug level decreased from 268 to 225 ng/mL (P less than 0.001). Dialyzer clearance was 67.0 +/- 3.9 mL of plasma per minute. The total quantity of drug removed by the dialysis treatment, however, was only 3.9 +/- 0.25 mg. The average half-life of amantadine in eight patients studied while receiving maintenance hemodialysis was 24.3 +/- 2.4 h of dialysis administered over approximately 13 days. Plasma half-life in six nonuremic control subjects was 12.2 +/- 1.6 h. Amantadine is poorly excreted in dialysis patients and has a large volume of distribution. The amount removed by a single dialysis is only a small fraction of the total body store.