This review concluded that the consumption of partially hydrogenated vegetable oils should be replaced to reduce the coronary heart disease risk. Replacement with vegetable oils was preferred, but tropical oils or animal fats were also beneficial. There were problems with the review and the conclusions were based on work beyond the review, which means that they might not be reliable.

To assess the effects of trans-fatty acid consumption on blood lipids and lipoproteins.

MEDLINE was searched for randomised controlled trials (RCTs), and the reference lists of previous reviews were checked for trials published to January 2008. For a second analysis of prospective cohort studies, no information was given about the search other than that only published studies were eligible.

Parallel or crossover RCTs that assessed controlled dietary interventions of trans-fatty acids were eligible for inclusion if food intake was controlled and described, the intervention lasted at least two weeks, and blood lipids were measured. The primary outcome appeared to be changes in the ratio between total and high-density lipoprotein (HDL) cholesterol. Other outcomes were changes in total cholesterol, low-density lipoprotein (LDL) cholesterol, HDL cholesterol, triglycerides, apolipoproteins, and other non-lipid risk factors (inflammation, endothelial dysfunction, insulin resistance, or weight gain).

The included trials were generally of healthy men and women, without diabetes, obesity, or dyslipidaemia. The mean age ranged from 25 to 63 years and the mean weight was 71kg (SD 11). The mean values of blood lipids reflected the population mean values for each age group (details reported). The amount of trans-fatty acids in tested diets ranged from zero to 11% of total energy. Dietary assessments included trans-fatty acids, saturated fatty acids, cis-polyunsaturated fatty acids, and cis-monounsaturated fatty acids. Mean duration of diet ranged from 17 to 70 days.

In a secondary analysis, prospective cohort studies were eligible for inclusion if they evaluated the association between trans-fatty acid consumption and coronary heart disease (CHD), including death due to CHD and nonfatal myocardial infarction. In included studies dietary habits were assessed either at baseline or using serial questionnaires.

The authors did not state how many reviewers selected the studies.

The authors did not state that they assessed validity.

RCTs: Outcome data were extracted to permit the calculation of differences in blood lipid measures between intervention groups.

Cohort studies: Data were extracted on the adjusted risk ratio for CHD per 2% more energy from trans-fatty acids. Adjustments were made for other cardiovascular risk factors and lifestyle habits.

The authors did not state how many authors extracted the data.

RCTs: Meta-variable regression analyses, stratified by gender, were used to assess the changes in risk factors in relation to age, weight, duration of intervention, and intake of trans-fatty acids, saturated fatty acids, cis-monounsaturated fatty acids, cis-polyunsaturated fatty acids, protein, cholesterol, and total energy.

Cohort studies: Pooled adjusted risk ratios were calculated using a random-effects model.

RCTs: Thirteen RCTs (554 participants) were included, with 66 interventions. Each 1% energy replacement of trans-fatty acids with saturated, cis-monounsaturated, or cis-polyunsaturated fatty acids decreased the total to HDL cholesterol ratio by 0.31 (saturated), 0.54 (monounsaturated), or 0.67 (polyunsaturated). There were similar results for apolipoprotein levels. There was insufficient data to assess the effects on other non-lipid risk factors.

Cohort studies: Four prospective cohort studies (139,836 participants, 4,965 events) were included. A 2% higher intake from trans-fatty acids was associated with an increased risk of CHD death or non-fatal myocardial infarction (RR 1.23, 95% CI 1.11 to 1.37).

The replacement of partially hydrogenated vegetable oils (containing trans-fatty acids), with alternative fats and oils, substantially reduced the CHD risk. Indications were that the benefits were greatest with replacement with vegetable oils, but even tropical oils or animal fats were beneficial, particularly in replacing partially hydrogenated vegetable oils that contained higher levels of trans-fatty acids. [Note: These conclusions were derived from work beyond the systematic review.]

The inclusion criteria were partially defined; those for participants were not described, while those for study design were clearly described for RCTs, but not for cohort studies. The search for RCTs was limited to one database and the checking of references. No search terms were reported. It is possible that trials were missed and the results of the review may have been affected by publication bias. There was no mention of any language restrictions and it is not possible to say whether language bias could have affected the results. The method used to identify the cohort studies was not clear and it is not possible to assess its reliability. The methods of the review were not described and it is not possible to decide whether they aimed to reduce reviewer error and bias. No validity assessment was reported, so it is not possible to assess the reliability of the results, or of the included studies. The details of the analyses of RCTs, including assessment of heterogeneity, were not presented and it is not possible to comment on them. The participants in the cohort studies were not described. The RCTs appear to have included healthy individuals and it might not be appropriate to generalise the results to those with impaired blood lipids or associated diseases. The data in the review came from small short-term RCTs and observational studies. The results of observational studies are generally considered to be of lower quality than those of RCTs. There were discrepancies between the text and tables in the number of participants in RCTs and within the figure for cohort studies.

The limitations or lack of reporting of the review mean that any conclusions drawn from its results might be unreliable. The authors' conclusions were based on work that was beyond the scope of the systematic review and it is not possible, within this abstract, to comment on these methods.

Practice: The authors stated that food manufacturers and suppliers should reduce the trans-fatty acid content in their food products and take advantage of this reformulation to maximise the benefit by replacing them with higher content cis-unsaturated fats.

Research: The authors did not state any implications for research.

National Heart, Lung, and Blood Institute, National Institutes of Health, grant number K08-HL-075628.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.