Methodological quality of the review: Medium confidence

Author: Wong TY, Mwamburi M, Klein M, Larsen M, Flynn H, Hernandez-Medina M, Ranganathan G, Wirostko G, Pleil A, Mitchell P.

 Region: Details not provided

Sector: Diabetic retinopathy

Sub-sector: Proliferative DR, severe visual loss

Equity focus: None specified

Review type: Effectiveness review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis methods: Not applicable

Background

In the last three decades, a relative decline in rates of diabetic retinopathy (DR) has been suggested by some studies; however, studies were conducted over 30 years ago. Contemporary estimates for DR are clearly needed, some of which may be provided by recent studies.

Research objectives

To summarize the best available evidence providing contemporary data on the clinical course of diabetic retinopathy and to examine potential differences in rates of diabetic retinopathy progression over time.

Main findings

In total, 28 studies were included in the review, of these 14 were conducted over the period 1975 – 1985 and the remaining 14 were conducted between 1985 and 2008.

After four years, pooled incidence rates for proliferative diabetic retinopathy (PDR) and severe visual loss (SVL) were 11.0% and 7.2%, respectively. Rates were lower among participants in 1986 – 2008 than in 1975 – 1985. After 10 years, similar patterns were observed. Participants in 1986 – 2008 had lower proportions of PDR and non-PDR at all time points than participants in 1975 – 1985 studies.

Authors noted that since 1985, diabetic patients had lower rates of progressions to proliferative diabetic retinopathy (PDR) and SVL. These findings may have reflected an increased awareness of retinopathy risk factors; earlier identification and initiation of care for patients with retinopathy; and improved medical management of glucose, blood pressure, and serum lipids. Based on baseline diabetic severity, the authors suggested researchers should consider contemporary rates of progression to PDR and/or SVL in estimating sample sizes for clinical trials and that future studies should report common nomenclature for DR and VA outcomes. There was a need for the publication of data on treatment-naïve patients with DR from larger population-based studies.

Methodology

Authors included prospective interventional or observational studies reporting the progression of diabetic retinopathy to proliferative diabetic retinopathy and/or severe visual loss at four-, five-, and 10-year time periods.

The authors searched MEDLINE, Current Contents, and the Cochrane Library for published studies from January 1975 to February 2008. PUBMED was also searched at six months prior to the search date (20 August 2007 to 20 February 2008) with no restrictions applied and current Contents for the year prior to the search date. Additionally, manual reference checks were also performed of bibliographies of articles and reviews published within the last five years (2004-2008).  All studies published in English, French, German, Spanish and Portuguese assessing the progression of diabetic retinopathy among patients with diabetes were included. Where necessary, authors of the included articles were contacted for specific data and analyses.

Inclusion criteria included (1) patients not yet treated for diabetic retinopathy; (2) followed at least for one year; (3) diabetic retinopathy assessed using retinal photography and/or fluorescein angiography; and (4) categorized using modified ETDSDR severity grades. Outcome measures included progression to proliferative diabetic retinopathy and progression to severe visual loss.

In the initial screening, abstracts were reviewed by a single reviewer for exclusion criteria. The complete studies were then rescreened and reviewed by two investigators based on prospective protocol. Studies were assigned a level of evidence using criteria from Centre for Evidence-based Medicine in Oxford, UK.

Studies were stratified into two time periods 1975 – 1985 and 1986 – 2008 for the meta-analysis.

Applicability/external validity

The authors did not review the applicability/external validity of the results.

Geographic focus

Geographical focus of included studies was not reported by the authors.

Quality assessment

Medium confidence in the conclusions about the effects of this study was attributed to the review. This review was based on a comprehensive search of the literature restricted to articles written in English, French, German, Spanish and Portuguese. Authors did not report assessing the quality and risk of bias of included studies. Nevertheless, authors acknowledged limitations of included studies appropriately and noted that included prospective interventional and prospective studies were very diverse in terms of study design, sample sizes, treatment settings and study inclusion criteria.