Retinal vascular calibre - a potential biomarker for early renal and retinal diseases?

Chronic kidney disease (CKD) is common in older people sharing overlapping risk factors with some diseases of the eye, including a complication of diabetes mellitus which affects the back of the eye (the retina) called diabetic retinopathy. CKD is more common in people with diabetes. Unfortunately some individuals with diabetic kidney disease develop progressive kidney failure and may eventually need dialysis. Previous research has suggested measurement of the blood vessels within the retina using a non-invasive procedure may provide early indications of some of the vascular complications associated with diabetes, such as those affecting the kidney and the eye. With recent advances in imaging and analysis technology, this non-invasive examination is now carried out routinely within the National Health Service as part of a regular screening programme for diabetic retinopathy. This application evaluates participant information collected from several large independent population-based CKD and diabetes cohort studies to determine whether measuring retinal blood vessels would help identify older or diabetic individuals with increased risk of developing kidney or eye complications. This may provide opportunities for earlier treatment to prevent or slow the onset of these complications.

The prevalence of diabetes is rising significantly in the United Kingdom and globally resulting in increased numbers of people being treated for diabetic complications. In the UK and elsewhere, kidney failure associated with diabetes is the commonest cause of end-stage renal disease (requiring dialysis or kidney transplant). Up to 40% of individuals with diabetes will develop diabetic kidney disease. Diabetic retinopathy is the primary cause of blindness in the UK working population and annually more than 2,000 people suffer permanent sight loss secondary to this complication. If our research demonstrated that retinal vascular measurements were predictive of CKD, diabetic kidney disease or diabetic retinopathy, then it could become a further component of the UK diabetic retinopathy screening programme, refining this process in terms of target age, disease duration and interval screening to maximise efficiency. Persons predicted to have highest risk of complications could be targeted for additional treatments to reduce their individual risk of diabetic kidney disease or diabetic retinopathy.

Although previous population-based studies suggest that retinal vessel measurements may provide an early prognostic indication of subsequent renal and visual dysfunction, the findings have not always been consistent. The reasons for these inconsistencies are primarily due to small sample sizes and loss of study participants during long-term follow-up. As such, this project will address these difficulties through collaborations with the principal investigators for all the major studies within this research field (to provide data relevant to the proposed outcomes) and limit the deficiencies affecting independent studies. This project will examine the relationships between measures of kidney function and retinal vessel diameter. Measurements will be assessed in a large number of individuals from population-based studies at baseline entry and again after 5 years follow-up. Variation in renal and visual function will be determined and assessed against retinal blood vessel measurements. Follow-up analysis of the findings will be undertaken in several large cohorts of participants with CKD or diabetes to establish if retinal blood vessel measurements are predictive of subsequent decline in visual or renal function. We anticipate our findings could be independently replicated and validated shortly after completion of the proposed study. If confirmed, this novel data could lead to additional clinical benefit being derived from routine retinal screening programmes i.e. identification of persons at increased risk of diabetic retinopathy, diabetic kidney disease or CKD.

This project will establish if retinal image analysis can provide novel biomarkers for chronic kidney disease (CKD), diabetic kidney disease or diabetic retinopathy using data from well-characterised prospective population-based and diabetes cohort studies using a two-stage meta-analysis approach on individual participant data.
Individuals will be stratified into two groups using measures of renal function (estimated glomerular filtration [eGFR]). The stable kidney function group are individuals in whom eGFR measurements remained >60mL/min/1.73m2 from baseline to 5 year follow-up. Retinal vascular calibre measurements in this "stable" kidney function group will be compared to retinal vascular calibre in individuals with progressive decline in kidney function. This "declining kidney function" group will contain individuals in whom eGFR decreased over 5 years to <60mL/min/1.73m2. Retinal vascular calibre will be correlated with risk of developing progressive CKD or retinal complications (including non-proliferative or proliferative diabetic retinopathy). Estimates of association will be undertaken using logistic regression and meta-analysis approaches independently in each study initially and then combined. This strategy will facilitate potential study-specific design bias and covariates not uniformly defined across studies, which can make evaluation of a combined model difficult. Adjustment for appropriate confounders will be made.
This approach will also be evaluated in diabetes cohorts with multiple measurements of kidney function (eGFR; urinary albumin/creatinine ratio) to determine if retinal vascular calibre can identify individuals at increased risk of renal and retinal complications. This research will help determine whether measuring retinal vascular calibre, as part of the NHS diabetic retinopathy screening programme, would offer additional clinical value in identifying patients at highest risk of developing renal failure or sight threatening retinopathy.

Who will benefit from this research?
Chronic kidney disease (CKD) is common, harmful and treatable and an important public health issue. As prevalence increases with age, the disease burden will increase with our ageing population. CKD is an independent risk factor for other diseases, particularly cardiovascular disease, commonly coexisting with other cardiovascular conditions necessitating co-management alongside other diseases and risk factors such as diabetes and hypertension, as well as the social needs that come with frailty and multiple conditions. In a minority of cases, CKD progresses to end stage renal disease (ESRD), which may require renal replacement therapy. This progression and the risks of other vascular events, such as stroke and heart failure can be reduced if CKD is identified and managed. Early diagnosis is therefore essential. Therefore ageing members of our society and individuals with chronic vascular disease will benefit from the outputs of this study.

The societal and economic costs attributed to CKD are immense. Treatment of ESRD, the most severe form of CKD, is resource intensive for the NHS. The current cost of treatment has been estimated at ~2% of the total NHS budget, yet individuals with ESRD comprise only 0.05% of the total population. As such society will benefit from reduced financial requirements for treatment of ESRD attributed to early intervention.

Social deprivation has been associated with a higher incidence and prevalence of CKD in developed countries. Furthermore, CKD progresses more rapidly in socially deprived patients. The effects on both incidence and progression are mediated through many intermediate factors at an individual-level including low birth weight, smoking, obesity, diabetes and hypertension and poor compliance with treatment, or regional variation, for example; variation in quality of primary care services and poorer access to secondary care. While measures of socioeconomic status (SES) are difficult to compare between studies, the effect of SES will be assessed within the context of the proposed primary outcome measures.

How will they benefit from this research?
CKD has historically been a significantly under-diagnosed condition. Historically many patients with CKD were diagnosed late with many requiring emergency dialysis. Early detection and evidence based management is cost effective, in both financial and human terms. The expected benefits of early diagnosis include:
1. Prevention of cardiovascular disease may be improved because people found to have early CKD are also at increased risk of premature cardiovascular disease and so can be identified earlier.
2. Reduction in hospitalisation through earlier diagnosis.
3. Progression to ESRD is delayed or prevented.
4. A reduction in lost working lives for people with ESRD who are too ill to work, and the social care cost of supporting people with ESRD.

Although it is currently not possible to repair kidney damage, it does not mean decline will definitely ensue. Assuming a healthier lifestyle with treatment for blood pressure and other conditions, it is possible to live without symptoms and prevent further decline of kidney function. In addition, reductions in preventable deaths due to kidney disease may be attributed to healthier lifestyles and diets through life-style modification. This will be of immense benefit to the patient, the NHS, public health policy makers and society as a whole. Earlier diagnosis of CKD is central to leveraging these benefits.

Practitioners and policy makers will benefit from new evidence that we will generate on the value (if proven) of a new prognostic marker.

Those commissioning research will be able to identify the future priorities for research in the area.