No abstract

Vancomycin is often administered to hemodialysis patients at long dosage intervals because its removal by hemodialysis is considered to be negligible. We and others, however, have demonstrated significant removal of vancomycin by high-flux hemodialysis. This report describes our experience with 89 courses of vancomycin using a revised regimen with a loading dose followed by 500 mg doses after each dialysis treatment, and compares results with 41 courses using single weekly dosing. All patients were dialyzed with high-flux membranes using volumetric ultrafiltration and bicarbonate dialysate. Serum vancomycin levels were obtained two hours after completion of infusion (peak) and immediately prior to dialysis (trough) and were measured by Abbot TDx fluorescence polarization immunoassay. Duration of multiple-dose therapy was 11 +/- 8 days, with mean total dose 3.6 +/- 1.8 g. Initial doses of 20 mg/kg rapidly and reliably established therapeutic pre-dialysis serum levels (10 to 25 micrograms/ml). In patients treated with multiple dosing 431 pre-dialysis levels were obtained. The mean level was 15.9 +/- 5.7 micrograms/ml; 55 levels (13%) were less than 10 micrograms/ml and 22 (5%) were above 25 micrograms/ml. In patients treated once weekly, 77% of levels were below 10 micrograms/ml by five days after administration, and 84% at one week. No patient developed demonstrable ototoxicity. Twenty-five patients were treated for > or = two weeks, five for > or = four weeks, and two for > five weeks, with no evidence of toxic accumulation. Mean peak level was 20.1 +/- 4.6 micrograms/ml, with a mean difference from preceding pre-dialysis level of 7.2 +/- 2.2 micrograms/ml. We conclude that in high-flux hemodialysis, a 20 mg/kg loading dose of vancomycin followed by 500 mg doses after each dialysis treatment achieves predictable, adequate and safe therapeutic levels, does not lead to unacceptably high peaks, and does not accumulate during long treatment courses. By contrast, once-weekly vancomycin dosing resulted in subtherapeutic serum levels after five to seven days, and should be abandoned in the high-flux setting.

Variation in numbers of land plant species on islands in the Galapagos Archipelago can be predicted on the basis of elevation, area of the adjacent island, distance from the nearest island, and distance from the center of the archipelago, but not on the basis of the area of the host island. Multiple linear regression (y = bx(1) + bx(2) . . .) gives better "goodness of fit" than curvilinear analysis (y = bx(z)). The variation in number of species on large islands can be predicted more accurately than the variation on small ones. Ecologic diversity and isolation are the natural regulators of species abundance.

Serum albumin and creatinine rose by 12 to 13% during the first half year of hemodialysis in a stable cohort. The slope of serum albumin versus time predicted survival, but it was not as predictive as the absolute albumin concentration. The pattern of correlations of baseline urinary protein and creatinine excretion with the respective monthly serum values of albumin and creatinine and their slopes is consistent with the hypothesis that as residual renal function declines, progressive retention of protein and creatinine contributes to the respective rises in serum albumin and creatinine.

Levels of vancomycin in serum are traditionally believed to be unaffected by hemodialysis. By both in vivo and in vitro techniques, the effects of a newer, more permeable dialyzer membrane on vancomycin concentrations were investigated. Six patients who were receiving vancomycin and undergoing maintenance hemodialysis with polyacrylonitrile dialyzer membranes had postdialysis levels in serum that were 63% of predialysis levels; the intradialytic half-life was 5.7 h. Vancomycin concentrations in serum exiting the dialyzer were 68% of those simultaneously entering the dialyzer at the beginning of dialysis. When polyacrylonitrile and conventional cellulose membranes were perfused in vitro with a recirculating solution of vancomycin, vancomycin concentrations fell to 39 and 91%, respectively, of the original concentration. The vancomycin concentration in the ultrafiltrate collected from the polyacrylonitrile membranes was only 23% of the original perfusate concentration. A significant decrease in the serum vancomycin concentration may occur during hemodialysis with newer high-flux dialyzer membranes. It appears that vancomycin binds to polyacrylonitrile membranes; this binding does not require the presence of protein and is affected by the pH of the perfusate.