Blood And Vision Integration: Enhancing Diabetic Retinopathy Diagnosis with Haematology and Imaging

Sathishkumar S,  B.Optom

Research Fellow, Medical Research Foundation, Sankara Nethralaya, Chennai, India

 

Diabetes is a serious condition that poses the risk of developing complications not only amputations, strokes, heart attacks, and heart failure but can also lead to complications such as Diabetic Retinopathy (DR).⁽¹⁾ A microvascular condition which can lead to severe Vision Impairment even blindness shows the importance of screening and early detection. To enhance the diagnosis and to understand these complications better, integration of Haematology and Imaging seems like a better solution.

Diagnostic Dilemma:

Current screening programs, especially those relying on mydriatic digital fundus photos, which creates challenges. These include discomfort for patients, variation in image quality, and the prevalence of non-gradable images. This creates delay in early diagnosis and intervention, impacting the overall effectiveness of DR screening. ⁽²⁾

A comprehensive solution:

1. Narrowing the gap with multiple field:

Including peripheral fields through handheld retinal imaging devices has shown enhanced sensitivity and specificity in detecting referable DR while reducing the ungradable rate. Using a 1-field (1F) photography approach centred on the macula can serve as a proficient screening protocol for Diabetic Retinopathy (DR). However, the utilisation of multiple field protocols, offering various views of the retina, theoretically improves screening accuracy compared to 1-field protocols.⁽⁴⁾ The inclusion of peripheral fields in DR screening programs utilising handheld retinal imaging has been observed to decrease the ungradable rate, resulting in increased sensitivity and specificity for referable DR.

2. Blood and Vision integration:

While retinal images offer valuable information regarding diabetes-related changes in the retina, relying on clinical findings at this stage may be relatively late for diagnosis. Enhancing the quality of DR referrals can be achieved by incorporating additional information such as haematological parameters. Given that diabetes not only impacts the eyes but also leads to complications like amputations, strokes, heart attacks, and heart failure, exploring potential biomarkers for earlier detection becomes important. This can be accomplished by assessing risk factors available from screening programs, including haematology-based biomarkers like Neutrophil-to-Lymphocyte Ratio (NLR), Platelet-to-Lymphocyte Ratio (PLR), and renal markers (eGFR, Scr, UAER). ⁽⁵⁾

• Neutrophil-to-Lymphocyte Ratio (NLR): NLR is a measure of systemic inflammation and immune response balance. It is calculated by dividing the neutrophil count by the lymphocyte count in a blood sample. Studies have shown that a high NLR is associated with poor prognosis in patients with type 2 diabetes mellitus (T2DM), predicting complications such as cardiovascular disease. ⁽⁶⁾
  Moreover, individuals with diabetic peripheral neuropathy (DPN), a common complication of diabetes, have been found to have higher levels of NLR compared to those without DPN. ⁽⁷⁾
• Platelet-to-Lymphocyte Ratio (PLR): PLR is another inflammatory marker that has been associated with systemic inflammation and vascular complications in diabetes. It is calculated by dividing the platelet count by the lymphocyte count. Studies have suggested that a high PLR may predict diabetic retinopathy in T2DM patients. ⁽⁸⁾ Moreover, PLR has been found to be an independent risk factor for evaluating diabetic retinopathy in type 2 diabetes patients.

3. The power of Deep Learning for faster diagnosis:

Deep learning methods, particularly convolutional neural networks (CNNs), are employed to extract features from haematology reports, fundus, and OCT images, facilitating the classification of patients into different classes of DR. This creates efficient patient classification and risk assessment, and improves early detection of DR.

Conclusion:

The blood and vision integration approach would be a valuable method for DR screening. By integrating haematological data with advanced imaging techniques, we move closer to a future where early diagnosis is not just a possibility but a reality. As technology continues to evolve our ability to differentiate the complexities of DR, ensuring early diagnosis and effective intervention at the right time.

 

References:

1. JackW. (2023, April 13). Number of people living with diabetes in the UK tops 5 million for the first time. Diabetes UK. https://www.diabetes.org.uk/about-us/news-and-views/number-people-living-diabetes-uk-tops-5-million-first-time
2. Scanlon P. H. (2017). The English National Screening Programme for diabetic retinopathy 2003-2016. Acta diabetologica, 54(6), 515–525. https://doi.org/10.1007/s00592-017-0974-1
3. https://www.istockphoto.com/en/photo/retina-of-diabetic-diabetic-retinophaty-gm492630033-40228996
4. Aptel, F., Denis, P., Rouberol, F., & Thivolet, C. (2008). Screening of diabetic retinopathy: effect of field number and mydriasis on sensitivity and specificity of digital fundus photography. Diabetes & metabolism, 34(3), 290-293.
5. He, J., Bian, X., Song, C. et al. High neutrophil to lymphocyte ratio with type 2 diabetes mellitus predicts poor prognosis in patients undergoing percutaneous coronary intervention: a large-scale cohort study. Cardiovasc Diabetol 21, 156 (2022). https://doi.org/10.1186/s12933-022-01583-9
6. Rezaei Shahrabi, A., Arsenault, G., Nabipoorashrafi, S.A. et al. Relationship between neutrophil to lymphocyte ratio and diabetic peripheral neuropathy: a systematic review and meta-analysis. Eur J Med Res 28, 523 (2023). https://doi.org/10.1186/s40001-023-01479-8
7. Zeng J, Chen M, Feng Q, Wan H, Wang J, Yang F, Cao H. The Platelet-to-Lymphocyte Ratio Predicts Diabetic Retinopathy in Type 2 Diabetes Mellitus. Diabetes Metab Syndr Obes. 2022;15:3617-3626