Authors: Drinkwater JJ, Davis TME, Davis W.A.

Geographical coverage: Brazil, Sweden, Japan, Spain, India, China, South Korea, Poland, South Africa and France

Sector: Biomedical

Sub-sector: Risk, association

Equity focus: Not reported

Study population: Patients with diabetes

Review type: Effectiveness review

Quantitative synthesis method: Not reported

Qualitative synthesis method: Narrative synthesis

Background: Microvascular (for example,  diabetic retinopathy, DR) and macrovascular (for instance,  carotid disease) complications of diabetes are interrelated due to their shared risk factors. However, specific association between the two is not well established.

Objectives: To assess the evidence linking carotid disease and DR in people with diabetes.

Main findings: The search identified 477 articles, of which 14 were included in this review. The included studies were from 10 different countries, including three from South Korea, two each from Spain and Brazil, and one each from China, Japan, Switzerland, Poland, France, South Africa and India.

Between studies, heterogeneity was noted due to differences used as markers of carotid disease and DR. Overall, five studies were of poor quality, one fair quality, and the remaining eight were of good quality. There were differences in the variables used as markers of carotid disease and DR across the included studies. Ten studies used carotid disease as the dependent variable, and the remaining four used DR. Ten of the 14 studies exhibited a statistically significant association between at least one ultrasound-assessed parameter of carotid disease and the presence or severity of DR, while the remaining four studies reported no significant association. The review authors concluded that the existing published evidence was inadequate to definitively determine the nature and direction of the relationship between DR and carotid disease. The authors proposed to conduct more well-designed longitudinal studies to determine the direction of association between DR and carotid disease, and to evaluate whether these two complications are independent coexisting conditions or if one condition contributes to the development of the other. In addition, the authors proposed that clinicians should perform a retinal examination if carotid atherosclerosis or stenosis is identified in an individual with diabetes. Conversely, if DR is detected, the authors proposed considering intensified cardiovascular risk management.

Methodology: Searches were conducted in PubMed, Medline and EMBASE to identify the articles that assessed the relationship between carotid disease and DR. The studies were included only if they were conducted on adult patients, had a minimum sample size of 50, and evaluated the effect of age and sex on the association between carotid disease and DR. The searches were conducted to the end of 2019 and limited to the English language only.

Two reviewers independently conducted searches, screening and data extraction. Reference lists of the included studies were checked to identify additional relevant articles. The risk of bias (RoB) was evaluated according to guidelines in the Cochrane Handbook, and the quality of the individual studies was assessed using the National Heart, Lung, and Blood Institute Quality Assessment Tool for Observational and Cross-Sectional Studies. The authors considered conducting a meta-analysis, but this was not feasible.

Applicability/external validity: The authors did not discuss the applicability or external validity of the results.

Geographic focus: The review focuses on studies conducted in Brazil, Sweden, Japan, Spain, India, China, South Korea, Poland, South Africa and France. The applicability of the findings of this review to low and middle income countries (LMICs) is not directly addressed by the authors.

Summary of quality assessment:

Authors used appropriate approaches to screen studies for inclusion, extract data of included studies, and overall analyse the findings of included studies. However, important limitations were identified in the review. The searches are not comprehensive enough to ensure that all potentially relevant studies were identified, therefore this study may be prone to publication and language bias. In addition, authors did not consider the risk of bias of included studies in the review findings. For these reasons, a low confidence was placed in this review.