The effect of plasmapheresis on the concentration of certain plasma proteins: a case identified by an inaccurate LDL-cholesterol estimation

Abstract

A 35-year-old man was found to have a negative LDL-cholesterol concentration (−0.05 mmol/L) when estimated on a fasting plasma sample using the Friedewald equation. Plasma urea, electrolytes and liver function tests (LFTs) were normal except for a raised total bilirubin of 74 μmol/L. Haematological results showed both a low haemoglobin and fibrinogen concentration. It transpired that the patient had undergone daily plasmapheresis treatments on the previous four days; plasma had been exchanged with a 5% albumin solution. He had been diagnosed with Evan's syndrome previously (characterized by autoimmune haemolytic anaemia) and had been admitted with severe anaemia, which had proved unresponsive to conventional treatments. The concentration of most plasma substances is reduced by 50–60% after one standard plasmapheresis treatment, with the rate of return to steady state concentrations varying among analytes. The finding of a negative LDL-cholesterol concentration (arising primarily as a result of normal triglyceride concentrations) may reflect the more efficient removal of LDL and HDL lipoproteins during the plasmapheresis procedure (PP) than lipoproteins containing proportionally more triglycerides. Plasma lipids, total protein, immunoglobulins and transferrin had recovered to steady state concentrations by eight days post-plasmapheresis, whereas caeruloplasmin concentrations had not. This case report illustrates the difficulties of obtaining accurate information on the steady state concentrations of plasma analytes, in particular protein bound substances, when analysis is carried out on a sample from a patient that has recently undergone plasmapheresis. The normal plasma albumin in this situation did not flag the possibility of the sample being artefactually diluted.