Guest Editor's Introduction: This paper was presented at the 26th ASAIO Congress held in April 1980. It was printed in Trans. Am. Soc. Artif. Intern Organs, Volume 26, page 406–411, 1980, and reprinted here with permission. This is the first clinical paper dealing with double filtration plasmapheresis. The plasma separator was made using Kuraray's polyvinyl alcohol hollow fiber, and Kuraray's plasma fractionator was made of ethylene vinyl alcohol hollow fibers. Seven patients afflicted with various diseases were treated by this method for 2 to 10 sessions. This procedure required supplemental infusion of albumin solution.

Several kinds of plasma fractionators have been introduced to actively separate protein fractions between albumin and globulins in double filtration plasmapheresis. However, relatively large molecular weight proteins are known to be partially trapped by the membrane in a plasma fractionator. In this paper, effects of membrane trapping on separation characteristics in plasma fractionators were examined during in vitro and in vivo studies. All in vitro experiments were done with a closed circuit under constant-flow rate filtration. Protein concentration in feed tank kept constant at no filtration in AS-14H, Evaflux 4A and 2A, while 20-40% of IgG and 40-60% of beta-lipoprotein in 2 liter plasma were removed by membrane trapping for 300 min when filtration fraction equaled 0.87. Protein plugging to the membrane seems to be a major factor in these proteins. And, Dead-end and Partially Discarded modalities with relatively high filtration fraction are effective for the separation between albumin and globulin. Twenty-seven DFPP treatments in 13 patients with autoimmune diseases were done to allow us to estimate the effects of membrane trapping during an in vivo study. All treatments using 6 types of plasma fractionator were performed under constant operating conditions with Partially Discarded modality. In any plasma fractionator, beta-lipoprotein was fairly trapped at 30 min after the start of treatment.

Until recently, the albumin concentration of supplementation fluid for double filtration plasmapheresis (DFPP) has been empirically determined. Inadequate albumin infusion often leads to hypoproteinemic symptoms such as edema. In the current study, an aimed condensation coefficient (CCaimed) was introduced in an attempt to estimate the appropriate plasma albumin level for each patient. This coefficient is theoretically derived from a one-compartment model for the patient's plasma albumin: CCaimed = CS/CD = 1 - (1 - CR)/[1 - exp(- CC.VR)] where CD and CS are albumin concentrations in discarded plasma and supplementation fluid. CR is the change ratio of albumin concentration in the patient's plasma during a DFPP treatment, and VR(= VS/VP) is the ratio of supplementation fluid volume (VS) to the patient's total plasma volume (VP). And CC denotes the albumin condensation coefficient in a DFPP line, which depends on the filtration fraction of the plasma fractionator (FFPF) and the sieving coefficients of both the plasma separator (SCPS) and the plasma fractionator (SCPF): CC = CD/CP = SCPS.(1 - FFPF.SCPF)/(1 - FFPF) where CP is the albumin concentration of the patient's plasma. From the above relations, CS can be determined as follows: CS = CC.CCaimed.CP Because many kinds of proteins are removed during a single DFPP treatment, a slightly higher albumin concentration in the supplementation fluid is needed to maintain an appropriate plasma level. Therefore, the CR value should be more than unity. For a patient with hematocrit (HCT) of 30%, body weight (BW) of 50 kg, and CP of 3.0 g/dl, who is receiving a DFPP treatment using AP-05H (SCPS of 0.970) and Evaflux 2A (SCPF of 0.526) under FFPF of 0.8 with VS of 500 ml, VP = BW(1- HCT/100)/13 = 50 x (1 - 30/100)/13 = 2.69 L, VR = 500/(2.69 x 1,000) = 0.186, CC = 2.81, and CCaimed = 1.25 assuming 1.1 for CR. Therefore, CS = 2.81 x 1.25 x 3.0 = 10.5 g/dl using the above equations.

No abstract

No abstract