Global Estimates of Diabetic Retinopathy Prevalence and Progression in Pregnant Women With Preexisting Diabetes

Abstract

IMPORTANCE Diabetic retinopathy (DR) may be worsened by pregnancy in pregnant women with preexisting type 1 diabetes (T1D) or type 2 diabetes (T2D). Conflicting findings from previous studies have resulted in inconsistencies in guidelines regarding DR management in pregnancy. Global estimates of DR prevalence and progression in pregnancy are therefore required to provide clearer information about the overall true burden of DR in this population.OBJECTIVE To estimate the prevalence of DR and its progression rate in pregnant women with preexisting T1D or T2D diagnosed before pregnancy.DATA SOURCES For this systematic review and meta-analysis, conducted from November 27, 2018, to June 29, 2021, a systematic literature search was conducted in MEDLINE/Ovid, Embase/Ovid, and Scopus databases to identify English-language articles that were published from inception through October 2020.STUDY SELECTION Observational studies that reported on DR and its changes in pregnant women with preexisting T1D and T2D.DATA EXTRACTION AND SYNTHESIS Two independent reviewers extracted relevant data from each included study. Data were pooled using a random-effects model with the Freeman-Tukey double arcsine transformation. This study followed the Meta-analysis of Observational Studies in Epidemiology (MOOSE) reporting guidelines. MAIN OUTCOMES AND MEASURES Prevalence of any DR, proliferative DR (PDR), and DR progression rates.RESULTS A total of 18 observational studies involving 1464 pregnant women with T1D and 262 pregnant women with T2D were included in the analysis. The pooled prevalence of any DR and PDR in early pregnancy was 52.3 (95% CI, 41.9-62.6) and 6.1 (95% CI, 3.1-9.8) per 100 pregnancies, respectively. The pooled progression rate per 100 pregnancies for new DR development was 15.0 (95% CI, 9.9-20.8), worsened nonproliferative DR was 31.0 (95% CI, 23.2-39.2), progression from nonproliferative DR to PDR was 6.3 (95% CI, 3.3-10.0), and worsened PDR was 37.0 (95% CI, 21.2-54.0). DR progression rates per 100 pregnancies were similar between the T1D and T2D groups, except for the development of new DR (T1D groups: 15.8; 95% CI, 10.5-21.9; T2D groups: 9.0; 95% CI, 4.9-14.8). A global trend toward a lower DR progression rate was observed after the 1989 St Vincent Declaration.CONCLUSIONS AND RELEVANCE Results of this systematic review and meta-analysis suggest that women with T1D and T2D had a similar risk of DR progression during pregnancy. Despite improvements in the management of diabetes and diabetes during pregnancy, DR prevalence and progression in pregnant women with diabetes remains higher than the nonpregnant population with diabetes, highlighting the need to improve DR management in pregnancy.

Statistical Considerations for Meta-analysis of Diabetic Retinopathy Prevalence

“…In all but 1 outcome (progression from nonproliferative diabetic retinopathy to proliferative diabetic retinopathy), the 2 methods were comparable (eTable 8 in the Supplement). 1 Although we note that a GLMM method would remove much of the concern around sampling error and bias, one downside to the GLMM is a lack of transparency regarding the weights given to each study. Thus, we decided to present the FTT results and forest plots so that our readers can focus on the clinical message and not be distracted by methodological arguments.…”

“…5 By taking into account the betweenstudy heterogeneity, which is substantial in Widyaputri et In Reply We thank Zhou and Shen for their interest in our article. 1 With the increasing prevalence of type 2 diabetes (T2D) in a younger population, it is unfortunate that there are few studies involving pregnant patients who have T2D that could provide relevant data for our review, and thus, limiting our ability to draw definite conclusions around the T2D population. As discussed in our article, diabetes type was one factor with high heterogeneity in our results; therefore, providing separate data from both diabetes types is important to elucidate the difference.…”

Learning by reading: A review of recent diabetes articles

“…In a metaanalysis of 18 studies of 1464 pregnant women with T1D and 262 with T2D, DR was present in 52.3% of pregnancies and 57.8% around delivery, but in T2D, prevalence was 9.4% in early pregnancy and 17.5% in late pregnancy. PDR was present in 6.1% during early pregnancy and 8.2% around delivery; new DR development occurred in 15%, worsening nonproliferative DR 31%, progression to PDR 6.3%, and worsened PDR occurred in 37% 19 . A commentary in JAMA ophthalmology highlighted the issue of continued vision loss in some patients with DR over long‐term follow‐up, despite optimal laser treatment and administration of antivascular endothelial growth factor therapy and suggested that many of these cases may reflect ganglion cell loss from the less well‐recognized entity of diabetic macular ischemia 20 …”

“…PDR was present in 6.1% during early pregnancy and 8.2% around delivery; new DR development occurred in 15%, worsening nonproliferative DR 31%, progression to PDR 6.3%, and worsened PDR occurred in 37%. 19 A commentary in JAMA ophthalmology highlighted the issue of continued vision loss in some patients with DR over long-term follow-up, despite optimal laser treatment and administration of antivascular endothelial growth factor therapy and suggested that many of these cases may reflect ganglion cell loss from the less well-recognized entity of diabetic macular ischemia. 20 It has been unclear whether nonalbuminuric CKD is associated with risk of poor kidney outcome and death comparable to than that seen in nonalbuminuric persons with diabetes without CKD.…”

Predictive model for diabetic retinopathy under limited medical resources: A multicenter diagnostic study

“…Several molecular and biochemical pathways are involved in the incidence and development of DR, but the interactions between various mechanisms remain to be fully elucidated ( 9 ). Clinical studies have identified a number of risk factors for DR, including demographic characteristics such as age ( 10 – 12 ), duration of diabetes ( 10 , 11 , 13 – 17 ), obesity ( 10 ), and pregnancy status in diabetic women of childbearing age ( 18 ), comorbidities or complications such as hypertension ( 10 , 12 , 16 ), dyslipidemia ( 10 , 12 , 13 ), cardiovascular disease(CVD) ( 11 , 12 ) and diabetic nephropathy ( 10 , 12 , 17 , 19 ), and other laboratory parameters such as glycated hemoglobin ( 10 , 11 , 13 , 15 – 17 , 19 ), glycemic variability ( 20 , 21 ), and susceptibility genes ( 22 ). However, the aforementioned risk factors derived from population-based studies can only account for 9% of DR progression ( 23 ).…”