Comment on “Effects of Flaxseed Interventions on Circulating Inflammatory Biomarkers: A Systematic Review and Meta-Analysis of Randomized Controlled Trials”

Dear Editor:

We carefully read the article by Rahimlou et al. (1) recently published in Advances in Nutrition. Although respecting the authors’ efforts, we are concerned about different aspects of the current review: the inadequate search strategy, the methodological flaws, the interpretation of data, and the reporting of the results, that might lead to biased results.

Our initial search revealed that a considerable number of eligible publications have been ignored (2–6). Although a few of those studies have not provided data, all of them are eligible to be included in the systematic review. The authors used 2 sets of keywords to define their search strategy: 1) flaxseed and its related keywords, and 2) inflammatory markers. This is despite the fact that inflammatory markers might not be reported in the titles/abstracts of possibly related clinical trials. It would be better to write a search strategy including “flaxseed” and its related terms along with the keywords related to “clinical trials” to capture all relevant clinical trials in this field. Thus, a larger number of studies might have been missed.

Furthermore, we believe that the authors made some serious errors in including or excluding the studies that they did. One study by Ricklefs-Johnson et al. (7) should have been excluded from the meta-analysis, but not the systematic review. After careful review, we noticed that the aim of the aforementioned study was to investigate the effect of ground flaxseed fiber (28 g/d) and ground psyllium fiber (9 g/d), which were matched for fiber mass content, but not for energy intake. In fact, the intervention and control groups were not isocaloric and the total energy of the flaxseed group was 5 times greater than the ground psyllium group according to the nutrient composition table (7). Whereas, we know that to achieve accurate effect sizes, the control and intervention groups must be isocaloric to be included in a meta-analysis. On the other hand, the authors included several redundant effect sizes. It is not methodologically correct to include >1 effect size from the same population. According to the Cochrane handbook, entering several comparisons from 1 study into a meta-analysis must be avoided (8), because it leads to double-counting of participants and to unit-of-analysis error caused by an unaddressed correlation between the estimated intervention effects from multiple comparisons. In this systematic review and meta-analysis, the authors made this mistake in several ways. For instance, they included 2 effect sizes from the Machado et al. (9) study, which compared the effects of brown and golden flaxseed as intervention groups in comparison with wheat bran. In addition, including 2 effect sizes using 1 intervention group and 2 control groups from 2 included studies (3-arm clinical trials) in the meta-analysis led to double-counting of participants and unit-of-analysis error as well (10, 11). In another similar fault, 2 different doses (13 and 26 g ground flaxseed) from the same population were considered as 2 different effect sizes (12). Altogether, the authors included 4 redundant effect sizes in the overall analyses which may result in misleading findings and conclusions.

Finally, the authors did not discriminate between whole and ground flaxseed in their review. They have used the word “flaxseed” for the results of studies which used both types (for example, their Table 4). Nevertheless, they should not be considered the same because the bioavailability of their bioactive compounds is different (13). Ground flaxseed provides remarkably more α-linolenic acid as well as lignans to the blood than does whole flaxseed (13, 14). Although α-linolenic acid and lignans, as important anti-inflammatory components, are involved in the anti-inflammatory properties of flaxseed (15–17), referring to the 2 forms as the same is not correct. The same fault has occurred in a previous systematic review and meta-analysis regarding the effect of flaxseed consumption and blood pressure (18), which is addressed as a critical mistake in a similar letter as well (19).

Overall, the current systematic review and meta-analysis suffers from several serious methodological problems and the conclusions are likely to be erroneous. Hence, we believe that the results of the current study should be interpreted with caution because of the stated errors. The correction of the aforementioned faults may lead to meaningful differences in the conclusions that can be drawn from the present meta-analysis.

Notes

The authors’ responsibilities were as follows—MA and HR-D: drafted the manuscript; AS-A: critically reviewed and revised the manuscript; and all authors: read and approved the final manuscript.

The authors reported no funding received for this work.

Author disclosures: The authors report no conflicts of interest.

MA and HR-D contributed equally to this work.