Previous studies have indicated that consumption of boiled coffee raises total and low density lipoprotein (LDL) cholesterol, whereas drip-filtered coffee does not. We have tested the effect on serum lipids of consumed coffee that was first boiled and then filtered through commercial paper coffee filters. Sixty-four healthy volunteers consumed six cups per day of this boiled and filtered coffee for 17 days. Then, they were randomly divided into three groups, which, for the next 79 days, received either unfiltered boiled coffee (lipid content, 1.0 g/l), boiled and filtered coffee (0.02 g lipid/l), or no coffee. Serum total cholesterol levels rose by 0.42 mmol/l (16 mg/dl; 95% confidence interval [CI], 0.14-0.71), LDL cholesterol levels by 0.41 mmol/l (16 mg/dl; 95% CI, 0.16-0.66), and apolipoprotein B levels by 8.6 mg/dl (95% CI, 3.8-13.4) in those who consumed boiled coffee relative to those who consumed boiled and filtered coffee. Responses of triglycerides, high density lipoprotein cholesterol, and apolipoprotein A-I did not differ significantly among these groups. No significant effects on serum lipid levels were found in the boiled and filtered coffee-consuming group compared with those who drank no coffee. In subjects who drank boiled coffee, serum campesterol level, an indicator of cholesterol absorption, remained constant. The serum lathosterol level, an indicator of cholesterol synthesis, increased by 11% (p less than 0.05), but the lathosterol to cholesterol ratio did not change. We propose that paper filters of the type used for drip-filtered coffee retain the lipid present in boiled coffee and in that way remove the hypercholesterolemic factor.(ABSTRACT TRUNCATED AT 250 WORDS)