Protocol for the meta-analysis on the association of olive oil consumption and risk of coronary heart disease and/or stroke

Miguel A. Martínez-González1, Ligia J. Dominguez 2, Miguel Delgado-Rodríguez3

1 Department of Preventive Medicine and Public Health, University of Navarra, Pamplona, Spain and CIBER Fisiopatologia de la Obesidad y Nutricion (CIBERobn), Instituto de Salud Carlos III, Madrid, Spain

2 Geriatric Unit, Department of Internal Medicine and Geriatrics, University of Palermo, Italy

3 Division of Medicine and Health Sciences, University of Jaen, Spain and CIBER Epidemiologia y Salud Pública (CIBERESP) Spain

1.Objective

This meta-analysis aims to quantify the association between olive oil consumption and risk of coronary heart disease and stroke and to assess the consistency of previous observational findings with those of a primary prevention randomised controlled trial (RCT).

2.Introduction

One of the most representative characteristic of the Mediterranean diet (Med-Diet) is the abundant consumption of olive oil, which is the typical culinary fat used in Mediterranean countries. Beyond ecological studies showing inverse associations between average country consumption of olive oil and rates of cardiovascular disease, there is also evidence to support that olive oil intake may have a profound beneficial influence on cardiovascular risk factors(1-5). Some minor components of olive oil other than oleic acid are likely to exert anti-inflammatory, antioxidant and other bioactive properties which can benefit the endothelium, favourably modulate haemostatic factors, and improve the stability of the arteriosclerotic plaque(6). Therefore, there are plausible reasons to believe that olive oil could contribute to protection against coronary heart disease and/or stroke.

Several case-control and prospective cohort studies have examined the association between consumption of olive oil and the risk of coronary heart disease and stroke. However, there are controversies on these relationships(7-9). The “Prevención con Dieta Mediterranea” (PREDIMED) RCT has recently shown that a Mediterranean diet supplemented with either virgin olive oil or tree nuts was able to reduce the incidence of cardiovascular disease in primary prevention among high-risk subjects(10). But, in spite of a significant protection for the primary combined end-point, in separate analyses, the protection by the diet supplemented with virgin olive oil was only significant for stroke, but not for myocardial infarction.

3.Search strategy and study selection

We will conduct electronic searches in PubMed, Embase, the Cochrane Library, Web of Science, and Ovid.

Search terms (with no restriction of language of publication or publication period): “olive oil” in combination with keywords relating to cardiovascular events (“cardiovascular disease”, or “cardiovascular event”, or “myocardial infarction”, or “coronary heart disease”, “coronary artery disease”, or “ischemic heart disease”, or “angina”, or “stroke”, or “cerebrovascular disease”); and in combination with keywords relevant to the study methods (“incidence”, or “cohort”, or “follow-up”, or “case-control” or “hazard ratio”, or “odds-ratio”, or “relative risk”, “rate ratio”). We also will review the bibliographies of the extracted articles and reviews to locate additional publications.

Two independent reviewers (Miguel A. Martínez-González, Ligia J. Dominguez) initially will search primary titles and abstracts to select potential articles (full text) for further scrutiny. When the title and abstract cannot be rejected by any of the reviewers, the full article will be obtained and carefully reviewed for inclusion or exclusion by the two reviewers after discussion and consensus.

4.Inclusion and exclusion criteria

Inclusion criteria

  • The study design was case-control, or a cohort with prospective follow-up
  • The exposure of interest was olive oil consumption
  • The outcome was coronary heart disease event or stroke
  • The investigators reported relative risks with 95% confidence intervals for at least three quantitative categories of olive oil intake

Exclusion criteria

  • Reviews, editorials, comments, letters without sufficient data
  • Abstracts of meeting presentations
  • Non-human studies
  • Studies that did not specifically considered olive oil consumption on coronary heart disease events or stroke incidence, or for which estimates for olive oil associations were not available
  • Studies of other exposures (such as, “oil” or “vegetable oil” without specification, or the whole Mediterranean dietary pattern without specification of olive oil consumption)
  • Studies reporting outcomes of other diseases

5.Data extraction

Study details

  • First author
  • City, country, community or hospital where participants were recruited
  • Year (or period) of participant recruitment
  • Year of publication
  • Study design
  • Number of participants
  • Number of cases (and controls if applicable)
  • Data collection (retrospective, prospective)
  • Consecutive or random assignment of participants
  • Characteristic of patient population (mean or median age, sex distribution, country of residence, percentage of smokers)
  • Number of centres (if applicable)
  • Dietary assessment method
  • Number of FFQ items
  • Olive oil intake categorization (intervention in RCT)
  • Validation of nutrients
  • Validation in study population
  • Single or multiple assessment of diet
  • Average duration of follow-up
  • Disease outcome
  • Method of diagnosis
  • Number of non-fatal and fatal events
  • Median or mean olive oil intake
  • Covariates in fully adjusted model
  • Original reported RR (95% CI)
  • RR (95% CI) for 25 g/day increase of olive oil consumption
  • Study quality will be assessed with Newcastle-Ottawa Quality Assessment Scale

Analyses plan

  • Relative risks and 95% confidence intervals will be considered as the magnitude of associationfor allstudies, and the odds ratios or hazard ratios will be considered equivalent to relative risks.
  • Those articles reporting both coronary heart disease and stroke will be treated as two separate reports.
  • Because it is possible that different cut-off points for olive oil intake categories will be present in different articles, we will compute a relative risk with 95% confidence interval for an increased intake of 25 g per day for each report.
  • We will use the method described by Greenland and O’Rourke(11) to estimate the pooled relative risk of an increment in olive oil consumption.
  • The median or mean olive oil consumption in each category will be used as the corresponding dose of consumption.The midpoint of the upper and lower boundaries will be considered the dose of each category if the median or mean intake per category is not available. If the highest category is open ended, the midpoint of the category will be set at 1.5 times the lower boundary.
  • Between-studies heterogeneity will be assessed with the Cochran's Q chi squared test and with the I2 statistic(12-14). Forest plots will be used to examine the overall effect. Funnel plots and Egger’s tests will be used to assess publication bias. Trim and fill method will be performed to identify and correct for funnel plot asymmetry arising from eventual publication bias(15).
  • Papers published by the same research group and studying similar factors will be checked for potential duplicate data. When it occurs, the largest and most recent set will be used for meta-analysis.

6.Management and Coordination

  • MAM-G and LJD will develop the initial plan, study design, and will be responsible for data collection, data extraction, and data initial interpretation.
  • MAM-G and MD-R will be responsible for statistical analysis.

7.Publication policy

  • Individual authors: Miguel A. Martínez-González, Ligia J. Dominguez, Miguel Delgado-Rodríguez.
  • MAM-G, LJD, and MD-R will be responsible for data interpretation, manuscript drafting, critical revision of intellectual content, and approval of the version to be published.

References

1.Lopez-Miranda J, Perez-Jimenez F, Ros E et al. (2010) Olive oil and health: summary of the II international conference on olive oil and health consensus report, Jaen and Cordoba (Spain) 2008. Nutr Metab Cardiovasc Dis 20, 284–294.

2.Hu FB (2003) The Mediterranean diet and mortality--olive oil and beyond. N Engl J Med 348, 2595–2596.

3.Huang CL & Sumpio BE (2008) Olive oil, the mediterranean diet, and cardiovascular health. J Am Coll Surg 207, 407–416.

4.Covas MI, Konstantinidou V, Fito M (2009) Olive oil and cardiovascular health. J Cardiovasc Pharmacol 54, 477–482.

5.Quiles JL, Ramírez-Tortosa MC, Yaqoob P (2006) Olive Oil and Health. Wallingford, UK: CAB International.

6.Ruiz-Canela M & Martinez-Gonzalez MA (2011) Olive oil in the primary prevention of cardiovascular disease. Maturitas 68, 245–250.

7.Bertuzzi M, Tavani A, Negri E et al. (2002) Olive oil consumption and risk of non-fatal myocardial infarction in Italy. Int J Epidemiol 31, 1274–1277; author reply 76–77.

8.Ferro-Luzzi A, James WP, Kafatos A (2002) The high-fat Greek diet: a recipe for all? Eur J Clin Nutr 56, 796–809.

9.Alemany M (2011) The Mediterranean diet: a group of healthy foods, a type of diet, or an advertising panacea? (Spanish). Med Clin (Barc) 136, 594–599.

10.Estruch R, Ros E, Salas-Salvado J et al. for the PREDIMED investigators (2013) Primary prevention of cardiovascular disease with a Mediterranean diet. N Engl J Med368, 1279–1290.

11.Greenland S & O’Rourke K (2008) Meta-analysis. In Modern Epidemiology, 3rd ed., pp. 658–659 [Rothman KJ, Greenland S and Lash TL, editors]. Philadelphia, PA, USA: Lippincott Williams and Wilkins.

12.Higgins JPT & Thompson SG (2002) Quantifying heterogeneity in a meta-analysis. Stat Med 21, 1539–1558.

13.Higgins JPT, Thompson SG, Deeks JJ et al. (2003) Measuring inconsistency in meta-analyses. BMJ 327, 557–560.

14.Higgins JP (2008) Commentary: heterogeneity in meta-analysis should be expected and appropriately quantified. Int J Epidemiol 37, 1158–1160.

15.Duval S, Tweedie R (2000) A non-parametric “trim and fill” method of assessing publication bias in meta-analysis. J Am Stat Ass 95, 89–98.

1