Plasma exchange (PE) is considered the most effective nonsurgical treatment modality for the reduction of low-density lipoprotein (LDL) in patients with familial hypercholesterolemia (FH). However, the concomitant reduction of high-density lipoprotein (HDL) and the necessity and cost of using blood products are major drawbacks of PE. We studied the effects of selective LDL reduction using monoclonal anti-LDL antibodies in an investigational immunoadsorption (IA) system. Results were compared with the effects of PE. During the study period, two homozygous FH patients with baseline cholesterol levels greater than 10.34 mmol/L (400 mg/dL) were treated sequentially for a combined total of 37 IA treatments and the results were compared with a total of 19 sequential PE treatments. The IA system consisted of on-line plasma processing over two columns of monoclonal anti-LDL antibodies in alternating cycles of column adsorption and regeneration. No replacement solution was needed. PE was performed with a centrifugal plasma separator using 5% albumin as replacement solution. Results showed that the reduction of lipids with IA was 43% +/- 0.9% for cholesterol, 51% +/- 1.0% for LDL, and 19% +/- 1.3% for HDL, resulting in a reduction in the LDL to HDL ratio of 41% +/- 1.7%. Compared with IA, percent reduction by PE was significantly greater (P less than 0.001) for all lipids, but was nonselective (cholesterol, 74% +/- 1.0%; LDL, 77% +/- 1.2%; HDL, 73% +/- 2.7%), and therefore the reduction of the LDL to HDL ratio was only 6% +/- 3.6%, which was significantly less than for IA (P less than 0.001). Pretreatment HDL concentration appeared to increase with repetitive IA treatment, but decreased back to prestudy levels with repetitive PE.(ABSTRACT TRUNCATED AT 250 WORDS)