Monounsaturated versus polyunsaturated dietary fat and serum lipids. A meta-analysis.

The objective of this study was to examine whether oils high in monounsaturated or polyunsaturated fats have a differential effect on serum lipid levels, using a meta-analytical approach. Fourteen studies (1983 through 1994) were identified that met six inclusion criteria, the primary criterion being that a study have at least two intervention diets that varied in monounsaturated and polyunsaturated fat content but were otherwise similar in total fat, saturated fat, fiber, and dietary cholesterol. Seven studies included a comparable high-saturated fat diet. Standardized effect sizes observed treatment difference in mean end-point lipid levels, divided by the pooled (SD) were calculated for individual studies, then individual effect sizes were pooled. The results indicated no significant differences in total, LDL, or HDL cholesterol levels when oils high in monounsaturated or polyunsaturated fats were compared directly. Triglyceride levels were modestly but consistently lower on the diets high in polyunsaturated fats (P = .05). Replacement of saturated fat with either monounsaturated or polyunsaturated fat led to significant decreases in total and LDL cholesterol (P < .001), and the pooled effect sizes were comparable for either type of unsaturate (effect sizes ranged from -0.64 to -0.68, ie, roughly a decrease of 0.65 mmol/L [25 mg/dL] relative to the high-saturated fat diets). Neither type of unsaturated fat significantly changed HDL cholesterol or triglyceride levels relative to the high-saturated fat diets. In conclusion, the evidence from this meta-analysis strongly indicates there is no significant difference in LDL or HDL cholesterol levels when oils high in either monounsaturated or polyunsaturated fats are exchanged in the diet. Any dietary recommendations for the use of one in preference to the other should be based on outcomes other than serum cholesterol levels.