Continuous monitoring of blood volume in double filtration plasmapheresis.

A continuous hematocrit (HCT) monitor, Crit-Line, was introduced to examine the change in patients' blood volume (BV) due to albumin loss during double filtration plasmapheresis (DFPP) treatments. Nine patients with autoimmune diseases or ABO incompatible renal transplantation received 15 DFPP treatments under Crit-Line monitoring. In these patients, plasma albumin concentration (C(P)) changed from 3.7 +/- 0.6 g/dl to 3.5 +/- 0.5 g/dl and HCT from 28.7% +/- 3.3% to 31.3% +/- 4.3% (change ratio [CR] of BV = -8.1%) during treatment with albumin concentrations (C(S)) of 9.5 +/- 1.0 g/dl and 500 ml volumes (V(S)) of supplementation fluid. Although the apparent CR value of C(P) was -5.3%, on average, the CR of albumin in the patients' plasma (M(P)) was -16.1%, which means a corrected CR value of C(P) by the HCT value to eliminate the influence of the patient's blood volume contraction during treatment. Albumin loss usually occurred in DFPP treatments. The decrease in BV was induced by an oncotic pressure drop due to albumin loss, and often resulted in a blood pressure drop. The amount of albumin loss during DFPP treatments strongly depends on sieving coefficients of the plasma separator (SC(PS)) and the plasma fractionator (SC(PF)), the filtration fraction of the plasma fractionator (FF(PF)), pretreatment C(P) value, and C(S) and V(S) values of the supplementation fluid. To determine the optimum C(S) and V(S) values for each patient, the authors introduced a variable blood volume model for albumin transport in DFPP. In this model, changes in C(P), HCT, and BV values could be estimated during treatment. For example, a patient with an HCT of 31.2%, body weight of 61.1 kg, and pretreatment C(P) of 4.4 g/dl received a DFPP treatment using a plasma separator, OP-05 (SC(PS) of 0.99), and a plasma fractionator, Evaflux 2A (SC(PF) of 0.40), under FF(PF) of 0.8 with a V(S) of 500 ml. A value for C(S) of about 10 g/dl is required for the patient to maintain a normal C(P) level during treatment by an estimation from the model. As a result of the treatment with a C(S) of 10 g/dl, the patient had no adverse reactions, such as a blood pressure decrease, during treatment under these conditions.