All currently used measurements of recirculation in blood access by chemical methods are flawed due to intradialytic disequilibrium or recirculation at low flow

Blood flows and recirculations with standard and reversed direction of line s were measured by chemical (urea and creatinine) and ultrasound dilution ( saline) methods in 47 chronic hemodialysis patients. Thirty-seven patients had 47 dual-lumen, central vein (CV) catheters: 32 were PermCath (Quinton I nstruments Company, Seattle, WA), 6 were Access Cath (MEDCOMP, Harleysville , PA), 3 were Soft Cell PC (Vas Cath, Mississauga, Ontario, Canada) and 6 w ere SNIJ (experimental catheters). Three of these last catheters had the ti p staggered 7 mm, and three had flush tips; PermCath, Access Cath, and Soft Cell PC catheters have the tips staggered 23 to 25 mm, Forty-six catheters were implanted into the superior vena cava/right atrium, and one catheter was implanted through the left saphenous vein into the left iliac vein. The catheters were studied 1 to 31 months after implantation (median, 3.0 mont hs). Ten patients with arteriovenous (AV) graft access were also studied. T he stop-flow method was used in catheter dialysis, and the slow-flow method was used to calculate recirculations in AV access dialysis with samples fo r systemic blood concentrations taken from arterial line both before and af ter samples from the arterial and venous lines. At 500 mL/min pump speed, a ctual blood flow was 436 +/- 18 mL/min (mean rt SD; range, 407 to 464 mL/mi n) with standard direction of catheter lines. At 500 mL/min pump speed, the arterial chamber pressure was -330 +/- 48 mm Hg (mean +/- SD; range, -380 to -225 mm Hg, and the venous chamber pressure was 259 +/- 48 mm Hg (mean /- SD; range, 140 to 310 mm Hg). Arterial chamber pressure was less negativ e, and venous chamber pressure was less positive with SNIJ catheters, which had larger internal diameter (2.1 mm) compared with the other catheters (2 .0 mm), Recirculation varied with the catheter design and the location of t he catheter tip. In the catheters with tip staggered more than 20 mm and wi th standard line connection at pump speeds of 50 mL/min and 500 mL/min, rec irculations were approximately 1% and 5%, respectively, when measured by th e chemical method. In the same catheters with reversed lines, the recircula tions were approximately 5% and 27%, respectively. Inflow failure catheters with reversed lines had similar recirculation values to those of well-func tioning catheters with reversed lines. In catheters with tips staggered 7 m m, and with standard connection of lines, recirculations were approximately 3% and 8%, respectively, at pump speeds of 50 and 500 mL/min, With reverse d lines, at the same pump speeds, the values were 7% and 12%, respectively. In flush-tip catheters, the recirculation was higher at a 50 mL/min pump s peed (approximately 17%) than at a pump speed of 500 mL/min (approximately 13%). The ultrasound dilution method usually gave lower values than the che mical methods, most likely because of overestimation of recirculation by ch emical methods. At least triplicate measurements are needed because single measurements by the ultrasound dilution method are associated with substant ial variation. We conclude that both currently used methods (stop flow and slow flow) of taking systemic samples for measurements of recirculation by chemical methods are flawed because of disequilibrium and recirculation at low flow. (C) 1498 by the National Kidney Foundation, Inc.