Proliferative Diabetic Retinopathy Independently Associated With CVD in T1D

Diabetic retinopathy
Diabetic retinopathy
Study found that the absence of proliferative diabetic retinopathy in those with stage 3b chronic kidney disease is associated with a lower prevalence of cardiovascular disease in patients with long duration of type 1 diabetes.

Individuals with type 1 diabetes (T1D) with chronic kidney disease (CKD) and proliferative diabetic retinopathy (PDR) may have a higher risk for cardiovascular disease (CVD) compared with those without PDR, according to a study published in Diabetes Care.

Researchers performed a cross-sectional study that included data from US residents (n=762) diagnosed with T1D for 50 years or longer (medalist study) and compared it with data from a replication cohort (n=675) from the longitudinal Finnish Diabetic Nephropathy Study (FinnDiane) to determine the relationships between CVD status and the presence of CKD and PDR.

CKD and PDR were defined as estimated glomerular filtration rate <45 mL/min/1.73m2 (CKD stage 3b) and according to the Early Treatment Diabetic Retinopathy Study (ETDRS) protocol, respectively.  

Characteristics of individuals from the medalist study specifically note a significantly older age at time of diabetes diagnosis in individuals with CKD but no PDR. The majority of individuals were female (55%), with an average age of 65 years and 53 years of diabetes. The prevalence of CKD, PDR, and CVD in individuals was found to be 12.9%, 53.3%, and 38.9%, respectively. Individuals from the medalists study were placed into the following 4 categories with the prevalence of CVD noted as follows:

  1. CKD and no PDR (+CKD/-PDR) (n=30): CVD prevalence 34.5%
  2. No CKD and no PDR (-CKD/-PDR) (n=327): CVD prevalence 28.8%
  3. CKD and PDR (+CKD/+PDR) (n=66): CVD prevalence 68.2%
  4. No CKD and PDR (-CKD/+PDR) (n=339): CVD prevalence 42.8%

Individuals from the FinnDiane study comprised a cohort of 45.2% female with an average age of 44.9 years, and an average length of time diagnosed with diabetes at 32.7 years. The prevalence of CKD, PDR, and CVD in individuals in this study was found to be 28.3%, 63.1%, and 18.1%, respectively. Individuals in this study were also placed into the same 4 categories as medalists study with the prevalence of CVD noted as follows:

  1. +CKD/-PDR (n=21): CVD prevalence 19.1%
  2. -CKD/-PDR (n=228): CVD prevalence 6.6%
  3. +CKD/+PDR (n=170): CVD prevalence 37.1%
  4. -CKD/+PDR (n=256): CVD prevalence 15.1%

Study results in medalist patients show the CVD prevalence rate to be half in +CKD/-PDR group compared with +CKD/+PDR group, with a further examination of covariates finding +CKD/-PDR group to be older at the time of diabetes diagnosis than those in the +CKD/+PDR group (average age of diagnosis 15 and 10 years, respectively P =.003). Also interesting to note was a higher detectable random C-peptide in the +CKD/-PDR group compared to the +CKD/+PDR group (56.7% and 28.6%, respectively P =.006).

Related Articles

Study results in FinnDiane patients did not find significant differences in age at time of diagnosis between groups, however it did demonstrate an upward trend of detectable C-peptide in the +CKD/-PDR group when compared to the +CKD/+PDR group 23.5% and 13.9%, respectively (P =.29).

Overall, the study demonstrates after adjustment for multiple factors, an independent association between PDR status and CVD in patients diagnosed with T1D and CKD (odds ratio 0.21; 95% CI 0.08-0.58, P =.003)

Researchers concluded that individuals diagnosed with T1D and CKD without PDR were found to have a lower prevalence of CVD, as a consistent pattern of lower CVD prevalence in individuals with CKD without PDR was observed in the study. They suggested that the absence of PDR may play a role as a protective factor against CVD, but stated this requires further research to determine if the presence of PDR and CKD is just a sign of generalized vascular dysfunction, or whether a true protective factor is present.


Gordin D, Harjutsalo V, Tinsley L, et al. Differential association of microvascular attributions with cardiovascular disease in patients with long duration of type 1 diabetes [published online January 31, 2018]. Diabetes Care. doi: 10.2337/dc17-2250