Table 4.
Qualitative aspects of the included systematic reviews and meta-analyses.
| Reference | Aim | Methods (Inclusion/Exclusion criteria) | Heterogeneity | Period | Quality Assessment |
|---|---|---|---|---|---|
| Hegsted et al. 1993 [45] | Overall evaluation of the rather extensive literature on the effects of dietary fatty acid composition and cholesterol on serum lipid concentration | Design: metabolic studies (appear to have been done under rather careful control in which food was prepared and fed to the subjects); field trials (diet was modified by instructions or by a combination of instructions and provision of some foods) | not analyzed | until 1991 | 8 |
| Mensink et al. 1992 [46] | Combining results to derive equations that relate changes in the dietary fatty acid intake to changes in serum HDL-C, LDL-C, TC and TG | Design: parallel design, crossover or Latin-square; “before and after” designs that lacked a control group were excluded. Diets enriched with very-long-chain ( n-3) PUFA were also excluded | not analyzed | 1970–1991 | 10 |
| Gardner et al. 1995 [47] | The purpose of this investigation was to address the controversy regarding a differential effect of MUFA vs. PUFA on serum lipids | Design: randomized trials comparing a high-mono and high-poly fat diet; similar in all respects (isoenergetic, total fat content, SFA) except for levels of monounsaturated and polyunsaturated fat intake; minimum 10 subjects on each diet arm | analyzed | 1966–1994 | 12 |
| Yu et al. 1995 [48] | Conducted to more comprehensively examine the effects of steraic acid, MUFAs, and other fatty acids on total and lipoprotein cholesterol concentrations in both men and women | Studies reported the quantity of individual SFA or steraic acid, sum of lauric, myristic and palmitic acids, and sum of MUFA and PUFA of the experimental diets. | not analyzed | 1970–1993 | 8 |
| Exclusion. Liquid formula diets; diets that were specifically enriched with in trans isomers; diets enriched with very-long-chain PUFA; subject with familiar hypercholesterolemia | |||||
| Clarke et al. 1997 [49] | The aim of this meta-analysis of metabolic ward studies is to provide reliable quantitative estimates of the relevance of dietary intake of fatty acids and dietary cholesterol to blood concentrations of total cholesterol and cholesterol fraction | Design: dietary intervention studies conducted under controlled conditions that ensured compliance | not analyzed | / | 9 |
| Garg 1998 [50] | Examining the effects of high carbohydrate low fat diets vs. high MUFA diets on metabolic indexes in T2D subjects | Design: randomized, crossover trials using isoenergetic, weight maintaining diets | not analyzed | / | 9 |
| Mensink et al. 2003 [30] | Combining results to derive equations that relate changes in the dietary fatty acid intake to changes in serum HDL-C, LDL-C, TC and TG, Apo-B and Apo A-I, TC:HDL-C | Design: parallel design, crossover or Latin-square; “before and after” designs that lacked a control group were excluded. Diets enriched with very-long-chain ( n-3) PUFA were also excluded | not analyzed | 1970–1998 | 13 |
| Shah et al. 2007 [46] | Comparing high carbohydrate and high- cis-MUFA interventions trials conducted to increase understanding of the effect of carbohydrate and cis-MUFA on blood pressure | Design: randomized and non-randomized intervention studies comparing the effects of high-carbohydrate diets with those of high- cis-MUFA diets on blood pressure (crossover or parallel design), comparison of diets isoenergetic, body weight had to remain stable | analyzed | until 2006 | 12 |
| Cao et al. 2009 [51] | Objective was to quantify the magnitude of the changes in lipids and lipoproteins in response to a MF blood cholesterol-lowering diet rich in unsaturated fat vs. LF in subjects with and without diabetes | Design: controlled feeding with a crossover or parallel design comparing MF vs. LF diets; designed to lower blood lipids; comparisons were isoenergetic; participants maintained constant weight during study; dietary protein and cholesterol were kept constant between diets | not analyzed | 1987–2007 | 14 |
| Jakobsen et al. 2009 [52] | Associations between energy intake from MUFA, PUFA, and carbohydrates and risk of CHD while assessing the potential effect-modifying role of sex and age | Design: cohort studies; published follow-up study with ≥150 incident coronary events; availability of usual dietary intake; a validation or repeatability study of the diet-assessment method used | analyzed | / | 10 |
| Kodama et al. 2009 [53] | To elucidate the effect of replacing dietary fat with carbohydrate on glucose and lipid parameters | Design: randomized controlled trials (crossover and parallel-group design); isoenergetic; only T2D | analyzed | 1966–2007 | 16 |
| Exclusion: T1D, diets with change in in the content or quality of carbohydrates; heterogeneity analyzed | |||||
| Mente et al. 2009 [54] | Examining the association between nutrient intake, dietary components, and dietary patterns and CHD and its related clinical outcomes | Design: cohort studies; dietary pattern: higher intake level is compared with lowest intake level; p-values for trend, where available, were used to evaluate dose-response relationship. FFQ, food records, 24 h recalls; Bradford Hill criteria | analyzed | 1950–2007 | 15 |
| Mozaffarianand Clarke2009 [55] | Examining the effects on CHD risk of replacing partially hydrogenated formulations on other specific fats on the basis of the content of TFA, SFA, MUFA and PUFA | Design: randomized controlled trials (consumption of fatty acids on risk factors), cohort studies (association of habitual intake of fatty acids with incidence of CHD events); isocaloric replacement | not analyzed | until 2008 | 10 |
| Skeaffand Miller2009 [56] | The purpose of this article was to summarize the evidence from cohort studies and randomized controlled trials of the relation between dietary fat and risk of CHD | Design: cohort studies; quintiles intake of PUFA, MUFA, SFA, TFA; The dietary assessment methods used in the cohort studies included single 24 h recall, diet records, diet histories and food frequency questionnaires; For MUFA only studies included in which exposure was determined by dietary assessment because blood fatty acids are not good biomarkers of MUFA intake | analyzed | / | 10 |
| Schwingshackl et al. 2011 [57] | Comparing high MUFA (>12% of TEC) vs. low MUFA (≤12% MUFA of TEC) on cardiovascular risk factors | Design: randomized controlled trials, ≥6 months, isocaloric and hypocaloric diets; subgroup analysis MUFA vs. LF, PUFA, LGI, HGI, Controls | analyzed | 1966–2011 | 13 |
| Schwingshackl et al. 2011 [58] | Comparing high MUFA (>12% of TEC) vs. low MUFA (≤12% MUFA of TEC) on glycemic control in subjects with abnormal glucose metabolism | Design: randomized controlled trials, ≥6 months, isocaloric and hypocaloric diets, subgroup analysis MUFA vs. LF, PUFA, LGI, HGI, Controls | analyzed | 1966–2011 | 13 |
Apo A I: Apolipoprotein A-I; Apo B: Apo lipoprotein B; CHD: coronary heart disease; FFQ: food frequency questionnaire; HDL-C: high-density lipoprotein cholesterol; HGI: high glycemic index; LDL-C: low-density lipoprotein cholesterol; LF: low fat; LGI: low glycemic index; MF: moderate fat; MUFA: monounsaturated fat; PUFA: polyunsaturated fat; SFA: saturated fat; T2D: type 2 diabetes subjects; TC: total cholesterol; TEC: total energy content; TFA: trans fat; TG: triacylglycerols.