Home / Resources / Articles / Relationship between Metabolites and Lipids in People with Diabetic Retinopathy 

Relationship between Metabolites and Lipids in People with Diabetic Retinopathy 

May 25, 2021
Editor: Steve Freed, R.PH., CDE

Author: Tarshay Boyd, PharmD. Candidate, LECOM School of Pharmacy

Can specific risk markers be identified for onset and early progression of diabetic retinopathy in people with type 1 diabetes? 

Diabetic retinopathy (DR) in type 1 diabetes is one of the leading disorders that can lead to blindness. Literature states that diagnosis and treatments have advanced. However, there are still limits to specific risk markers to determine the onset and early progression of DR. The purpose of this study was to investigate the omics panel of metabolites and lipids related to DR in type 1 diabetes. Previous studies reported similar metabolites associated with DR in patients with type 2 diabetes with DR, with one study finding that 3,4-DHBA and ribose were higher in the DR groups. Other studies found that 2-deoxyribonic acid and other sugars were significant in DR.  Another study found that branched-chain, aromatic amino acids were associated with diabetic microangiopathy in type 2 diabetes.  In a cross-sectional study, they evaluated metabolites that were found to be associated with DR in type 2 diabetes, hydroxyl fatty acids, and sugar derivatives.  


This study’s objective was to investigate a broad panel of metabolites and lipids found in plasma related to the presence, onset, and progression of DR in patients with type 1 diabetes. The cohort study evaluated 648 participants, mean age 54.4 ± 12.8 years, 56% men, who were followed 5.1 to 5.5 years. These patients had an extensive range of albuminuria; investigators subdivided the study population into stages of regular, micro, and macroalbuminuria. The participants were divided into no DR (stage 0), mild nonproliferative DR (NPDR) (stage 1), moderate NPDR (stage 2), proliferative DR (stage 3), and proliferative DR with fibrosis (stage 4). The primary endpoints were to report any DR progression, including mild to severe DR using Cox models. These endpoints were well-defined as a progression from any stage to any other set of DR, the onset of DR, and progression from stage 1-2 to stage 3-4. 

ANOVA and Chi-Square are used to analyze the baseline characteristics for continuous and categorical variables.  Multivariate linear regression models were used to assess the metabolites or lipids and baseline DR stages.  Benjamini-Hochberg method corrects for multiple testing for p-values. The metabolomics samples were measured using two-dimensional gas chromatography with time-of-flight mass spectrometry. The prepared lipidomics samples were analyzed by ultra-high-performance liquid chromatography combined with quadrupole time-of-flight mass spectrometry.   

The study reported 141 participants with no DR and 186 with moderate DR at baseline.  Patients with PDR with fibrosis resulted in a higher average standard deviation BMI (24.7 VS. 26.8 kg/m2), UAER (11.5 [IQR 7.6-22.4] vs. 49.0 [IQR 14.9-231.8] mg/24 h), systolic blood pressure (127 vs. 135 mmHg), and lower eGFR (94.8 vs. 64.5 ml/min/1.73 m2).  Patients with higher stage DR had a much higher rate of treatments with antihypertensive drugs and statins. 4 out of 75 metabolites identified were positively related to the baseline DR stage. 2,4-DHBA identified with p-value < 0.001, ribonic acid p-value 0.017, ribitol p-value 0.032 and 3,4-DHBA p-value 0.036.  As seen in the higher DR stages, there was an increase in the four metabolite levels; ribonic acid p-value < 0.001, 2,4-DHBA p-value <0.001, ribitol p-value 0.013, and 3,4-DHBA p-value 0.041. The levels of the four metabolites increased significantly in the higher DR stages. The lipids levels of LPCs (16:1), TG (49:3), TG (50:1), and TG (50:2) were all related to an association with DR grade. The LPC (16:0) p-value of <0.1 was detected.  The number of DR events was 133, 47, and 29, as seen for any progression, the onset of DR and progression from mild to severe DR. 3,4-DHBA showed higher levels with HR 1.55, 95% CI (1.12-2.15), p-value 0.033.  2,4-DHBA was not statistically significant but did show an increase in HR for progression from mild to severe DR with HR 1.92, 95% CI (0.94-3.93), p-value 0.290.   

To conclude, diabetic retinopathy is associated with metabolites and lipids, as reported in this cohort study showing a high percentage of more severe DR patients with type 1 diabetes. Four metabolites are related to DR’s presence. They showed a higher relationship with 3,4-DHBA as an independent risk marker for DR’s progression, identifying ribonic acid, ribitol, and two DHBAs associated with the DR stage. The study identified triglycerides that were not related to the DR stage. The study’s limitations were the following: laser surgery or vascular endothelial growth factor injections were not accounted for, no data for concomitant medication available, and no data on how the therapy with statins, antihypertensive medications changed during follow-up. There were no lifestyle parameters documented. The authors suggest that further studies are required to evaluate further and confirm their findings to use as markers for clinical application for DR progression.  

Practice Pearls: 

  • Advancements in omics-based methods might be the future of identifying new risk markers associated with diabetic retinopathy. 
  • 3,4-dihydroxybutyric acid shows as a risk marker for the progression of diabetic retinopathy in patients with type 1 diabetes. 
  • Identification of specific risk markers for onset and early progression of diabetic retinopathy is critical to control DR. 


Circulating Metabolites and Lipids Are Associated with Diabetic Retinopathy in Individuals With Type 1 Diabetes. Diabetes Oct 2020, 69 (10) 2217-2226; DOI: 10.2337/db20-0104 https://doi.org/10.2337/db20-0104 


Tarshay Boyd, Pharm D. Candidate, LECOM School of Pharmacy