The hypothesis that increased cholesterol synthesis provides a mechanism that contributes to nephrotic syndrome-associated hyperlipidemia is mainly based on experimental evidence. The serum level of the cholesterol precursor, lathosterol (expressed per millimole cholesterol), is a reliable marker of whole-body cholesterol synthesis in normocholesterolemia and primary hypercholesterolemia. Serum lathosterol and lipoprotein levels were measured in 11 moderately hyperlipidemic patients with nephrotic-range proteinuria and 22 matched controls. The proteinuric patients were evaluated before and during three antiproteinuric treatment periods with angiotensin-converting enzyme (ACE) inhibition therapy (n = 6) or a low-protein diet (n = 5) alone, in combination, and again as a single treatment. In untreated patients, serum total cholesterol, very-low-density (VLDL) and low-density (LDL) lipoprotein cholesterol, apolipoprotein B (apo B), and lipoprotein (a) [Lp(a)] levels were higher than in controls (P < .01 to P < .001), but the lathosterol to cholesterol ratio tended to be lower in patients (0.99 +/- 0.43 micromol/mmol) as compared with controls (1.29 +/- 0.41 micromol/mmol, P < .10). During combined antiproteinuric treatment, total and VLDL + LDL cholesterol, apo B, and Lp(a) decreased (P < .02 to P < .01), but remained higher than levels in controls. Yet the serum lathosterol to cholesterol ratio changed little and was even lower (P < .05) in treated patients than in controls. Serum total cholesterol (r = -.82, P < .01) and apo B (r = -.84, P < .01) were inversely correlated with serum albumin in untreated patients, whereas the serum lathosterol to cholesterol ratio was not (r = -.01, NS). In the patient group, multiple regression analysis showed that changes in the lathosterol to cholesterol ratio during the study were only related to changes in the dietary polyunsaturated to saturated fatty acids ratio (P:S) coinciding with the low-protein diet (P < .01). In contrast, the decrease of VLDL + LDL cholesterol, apo B, and Lp(a) was independently related to reduction of proteinuria (P < .02 to P < .001), but not to changes in the lathosterol to cholesterol ratio. In conclusion, the present data, based on the serum lathosterol to cholesterol ratio, do not support the concept that increased cholesterol synthesis plays an important role in the maintenance of human nephrotic syndrome-associated hypercholesterolemia. Moreover, it appears unlikely that the decrease of apo B-containing lipoproteins with antiproteinuric treatment is attributable to inhibition of cholesterogenesis. These findings warrant further documentation of cholesterol synthesis in human nephrotic syndrome by direct methods.