The association of smoking and risk of diabetic retinopathy in patients with Type 1 and Type 2 diabetes: a meta-analysis

Methodological quality of the review: Medium confidence

Author: Cai X, Chen Y, Yang W, Gao X, Han X, Ji L.

Region: Not reported

Sector: Diabetes Type 1 and 2

Sub-sector: risk, smoking

Equity focus:  None specified

Review type: Other review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background:

Diabetic retinopathy (DR), as a specific microvascular complication of diabetes, is the leading cause of blindness among adults in developed countries. A few of the potential risks of DR may include smoking and alcohol, however these have not been clearly demonstrated in the literature.

Objectives:

To clarify the relevance between smoking and diabetic retinopathy in patients with Type 1 and Type 2 diabetes mellitus.

Main findings:

In total, authors identified 73 studies, among which 19 studies included Type 1 diabetes patients and 56 studies included Type 2 diabetes patients. Out of the 44 included cross-sectional studies, 33 publications were high quality and 11 publications met the medium quality criteria. Out of the 42 included cohort studies, 27 studies were in high quality and 15 studies were attributed medium quality. The four included randomized control trials (RCTs) were attributed low risk of bias.

Authors reported that in Type 1 diabetes, compare with non-smokers, the risk of diabetic retinopathy significantly increased in smokers (risk ratio (RR) =1.23, 95% confidence interval (CI) 1.14, 1.33, P < 0.001), and the risk of proliferative diabetic retinopathy also significantly increased in smokers (RR= 1.48, 95% CI 1.20, 1.81, P < 0.001).

In Type 2 diabetes, compare with non-smokers, authors report that the risk of diabetic retinopathy significantly decreased in smokers (RR= 0.92, 95% CI 0.86, 0.98, P = 0.02) and the risk of proliferative diabetic retinopathy also significantly decreased in smokers (RR= 0.68, 95% CI 0.61, 0.74, P < 0.001).

In subgroup analysis, compared with non-smokers, authors report that the risk of DR decreased non-significantly in smokers in 21 cohort studies (RR = 0.96, 95% CI, 0.91–1.01, P = 0.12), but decreased significantly in smokers in 39 cross sectional studies (OR = 0.91, 95% CI, 0.84–0.98, P = 0.02). Authors report that the risk of DR decreased significantly in smokers than non-smokers in studies with standard fundus photo examination (OR =0.90, 95% CI, 0.87–0.92, P < 0.001), while decreased without significance in smokers in studies with fundoscopy examination (odds ratio (OR) = 0.91, 95% CI, 0.74–1.12, P = 036). Meta-regression analysis results showed that in Type 1 diabetes, covariates as age, gender, BMI, HbA1c, duration of diabetes and blood pressure levels were not associated with the risk of DR in smokers when compared with nonsmokers. However, authors found that in Type 2 diabetes, the duration of diabetes was negatively associated with the risk of DR in smokers when compared with non-smokers (β = −0.020,95% CI, −0.037, −0.002; P = 0.027), but not the other covariates.

According to authors, no publication bias was identified in the review.

Methodology: 

Comprehensive searches for published studies from Feb 2017 were performed using two electronic databases, including MEDLINE and EMBASE. The search strategy included the following search terms: diabetic retinopathy, diabetic complication, microvascular complication, risk factor, smoking, smoke, nicotine, cigarette, cigar, tobacco, diabetes mellitus, Type 1 diabetes, and Type 2 diabetes. The primary outcome of this meta-analysis was to clarify the association between smoking and diabetic retinopathy in Type 1 and Type 2 diabetes patients.

Studies were considered eligible for inclusion if they fulfilled the following inclusion criteria: (i) presented primary outcome as evaluating the association between smoking and diabetic retinopathy, or the risk factors of diabetic retinopathy including smoking in observational studies (cross-sectional studies, cohort studies, or case–control studies) or randomized controlled trials (RCTs); (ii) diagnosed study participants clearly with Type 1 diabetes (or insulin dependent diabetes mellitus) or Type 2 diabetes (or non-insulin dependent diabetes mellitus); (iii) evaluated the presence of diabetic retinopathy in smoking and non-smoking groups separately.

Two investigators independently screened the titles, abstracts, and full articles according to eligibility criteria. Full-text articles were assessed and data were extracted into standardized data tables and independently verified by two investigators. Qualities of cross-sectional studies were evaluated by using the Agency for Healthcare Research and Quality (AHRQ) tool. For each study, the quality was assessed as follows: low quality = 0–3; moderate quality = 4–7; high quality = 8–11. Qualities of cohort studies were evaluated by using the Newcastle-Ottawa Scale (NOS). The evaluation included selection, exposure/outcome and comparability. High quality studies were scored 6–9 stars. By using the Cochrane Collaboration tool, RCTs were rated as having a low, high, or unclear risk of bias from six aspects.

Authors pooled data by calculating OR or RR with 95% confidence interval (CIs) for comparisons of dichotomous outcomes between smokers and non-smokers by using the Mantel-Haenszel statistical method. Higgins I² statistic was used to quantify the percentage of total variance in the summary estimate due to between-study heterogeneity. Random-effects models were used when an I² value was more than 50% which represented substantial high levels of heterogeneity, while fixed-effects models were used when an I² value was no more than 50% which represented low levels of heterogeneity. Authors also conducted a sensitivity analysis, assessed publication bias, and conducted subgroup analysis of observational studies and meta-regression.

Applicability/external validity:

Authors note that compared with non-smokers, the risk of diabetic retinopathy significantly increased in smokers with Type 1 diabetes while significantly decreased in smokers with Type 2 diabetes.

Geographic focus:

Authors did not discuss the applicability of the results to low- and middle-income settings.

Summary of quality assessment:

Overall, medium confidence was attributed in the conclusions about the effects of this study as important limitations were identified. Although authors used appropriate methods to screen and extract data of included studies, authors did not conduct a thorough search to ensure that relevant studies were included in the review. In addition, it is not clear from the review if authors avoided language bias within the review.