|
|
Non-modifiable risk factors
Duration of diabetes
It is of note that it is the time since diagnosis independently of diabetes control which is the most important risk factor for developing some form of retinopathy,2 2,366 diabetic patients participated in the famous Wisconsin Epidemiologic Study of Diabetic Retinopathy which found a strong link between duration of diabetes and presentation of retinopathy (Figure 1): 'the prevalence of retinopathy varied from 2 per cent in persons with less than two years of diabetes to 98 per cent in persons with 15 or more years of disease the severity of diabetic retinopathy was also related to duration of diabetes.'3
Bek et al published a reputable set of data, gathered from 5,365 diabetic patients over a period of more than 14 years at Aarhus University Hospital, who were involved in a standardised screening programme there.4 From their data, it can be seen that as the years go by from first examination of diabetes, progressively more diabetics go on to develop retinopathy that clinically requires treatment (regardless of type). Type 2 patients seemingly need treatment earlier in the disease than type 1 patients; however it is questionable whether type 2 diabetes is always diagnosed the same year as its onset.
Age of patient
Given the importance of disease duration, it follows that by the end of their lives, diabetic patients are more likely to have severe retinopathy. Certainly, older people are more likely to have diabetes. However, many have questioned whether age alone can be used to predict the stage of retinopathy. Klein5 is doubtful: 'the effect of disease duration is not uniform at all age...following puberty, age is not a significant predictor of retinopathy status in either insulin-dependent or non-insulin diabetic persons.'
While a review by Pinhas-Hamiel and Zeitler6 demonstrates a considerable rise in diabetes in younger individuals over the last two decades, it is possible that their findings simply reflect an increased awareness, screening and therefore recording of diabetes (particularly in young people following associated media interest).
However, a thorough statistical analysis by Wong et al7 of 14,074 type 2 patients provides a strong argument that 'age of diagnosis is an independent predictor of long-term retinopathy'. They looked at 624 patients with disease duration of 20-30 years, and 852 with disease duration 10-12 years, all referred to hospital in Sydney (Australia), and made statistical allowances for both hyperglycaemia and duration of diabetes. In the longer duration group, patients under the age of 45 (at age of diagnosis) had approximately three-times more retinopathy, and more sight-threatening retinopathy. Wong et al put forward an interesting hypothesis, that young people are more predisposed to hyperglycaemia and an aggressive response to VEGF, and thereby young (under age 45) people are more predisposed to retinopathy; this has not yet been proven.
Gender
Klein et al3 have stated that males are more at risk of retinopathy than females: 'In the WESDR, younger-onset male patients with 10 or more years of diabetes were one and a third times more likely to have PDR [proliferative diabetic retinopathy] than younger-onset female patients with similar duration of disease.' Another large study4 strongly supports these findings; they show that men on average tend to require treatment earlier than women. There was less than a 5 per cent statistical chance that this is not the case for male Type I patients. Furthermore, the UK government figures show that male type 2 patients had a higher incidence of retinopathy than female type 2 patients.
The reason why males on the whole are more susceptible to retinopathy has not yet been established - one possibility is lifestyle differences.8 Although the evidence suggests men require treatment earlier on average, the disparity between the sexes is not especially large; this risk factor is less important than the age of the patient, and definitely less than disease duration.
Family history
Diabetes UK9 has published statistics on diabetes, which confirm that having two parents with diabetes, significantly increases the risk of developing the condition. In type 1 diabetes the risk goes from 2 per cent (mother with condition) or 8 per cent (father with condition) to 30 per cent if both parents are affected. In type 2 diabetes the risk with one parent affected is 15 per cent, rising to 75 per cent if both parents are affected. Genetics seems likely to play a role in which offspring will become diabetic. However, not all children with diabetic parents will get this condition. It is possible that a diabetic 'lifestyle' (non-genetic factors such as lack of exercise and poor diet) may be confounding factors.
Ethnicity
According to Diabetes UK9 type 2 diabetes is up to six times more common in people of South Asian descent and up to three times more common among people of African and African-Caribbean origin. Their published data on self-reported cases in England give a fuller picture (Table 1).
The WECI (Welsh Eye-Care Initiative) gives free eye tests to individuals considered at risk of certain eye diseases, including people who are of a particular race considered more at risk of diabetic retinopathy: Black-African, Black-Caribbean, Indian and Pakistani. From Table 1, this makes sense, especially for the Indian and Black-Caribbean populations, which have over twice the prevalence of diabetes as the general population.
Nephropathy
Kidney disease (nephropathy) is more common in diabetes, though until recently it was unclear whether nephropathy or diabetes was the cause or effect. Klein's analysis of the WESDR showed that in people with 10 or more years of diabetes, PDR was found two or three times as frequently in the presence of proteinuria [excess protein in urine due to bad filtering by kidneys] as in its absence.3
Glynn et al10 measured blood samples from the 1,441 participants of the Diabetes Control and Complications Trial (DCCT) and, based on their results, proposed that nephropathy may represent an early risk marker of long term complications in type 1 diabetics. These results hold merit, as the authors had adjusted results for several confounding variables including age, sex, duration of diabetes, and had randomised treatment assignment.
In 2010, findings were published based on the 15-year Beaver Dam Eye Study, in which over 3,000 people were examined. They showed that nephropathy was a risk factor for new cases of retinopathy, as well as its severity. The authors point out that renal disease is associated with hypertension, inflammation, and endothelial dysfunction, all hypothesised as pathogenic factors for the development of retinopathy.11
Hypothyroidism
The American Diabetes Association12 performed a cross-sectional study, the underlying hypothesis of which was that there would be a link between poor thyroid function and diabetic retinopathy, because patients with subclinical hypothyroidism have an increase in cardiovascular events, as is often the case with diabetic patients. The results of the study showed a trend towards higher rates of sight-threatening diabetic retinopathy in a group of subclinical hypothyroid patients when compared with the 'euthyroid' (normally functioning thyroid gland) group. Given the 95 per cent confidence interval found, and the 1,170 type 2 patients looked at in the study, it is tempting to make a definitive link between poor thyroid function and level of diabetic retinopathy.
This trial has merit due to its randomised nature and acknowledgement of other factors which may affect retinopathy, including age, sex and duration of diabetes. however, there is no description given as to what adjustment was made. Furthermore although the article claims to investigate 1,170 patients, they were actually only screened; 327 patients were further investigated by qualified retinal photographers. It is also likely that patients with poor thyroid health would be more susceptible to poor health and lifestyle in general. It is this author's opinion that more needs to be known about this relationship. Hypothyroidism has not yet been established as a definitive risk factor.
Pregnancy
Gestational diabetes (diabetes occurring in pregnancy) should be highlighted when giving advice to patients who are, or wish to become pregnant. If patients are considered at risk they should be closely monitored. Gestational diabetes usually occurs in the second or third trimester, and often ceases after the birth. The strain on the body to produce extra insulin in this period may cause the mother-to-be to become prematurely diabetic; poor control of existing diabetes can exacerbate problems during pregnancy.
If a diabetic woman is pregnant, this poses a threat to her unborn baby, and every effort should be made to tighten the mother's diabetic control. Babies of diabetic women have a three-fold risk of death in the first few months of life9. Furthermore, while the condition is related to the period of pregnancy, those who develop gestational diabetes are more at risk of long-term complications: the lifetime risk of developing Type 2 diabetes after gestational diabetes is 30 per cent.
References
1 American Diabetes Association (2003c) Implications of the United Kingdom Prospective Diabetes Study, Diabetes Care, 26, 28-32.
2 Huang K, Starita C and Sultan M. Epidemiology, risk factors and management of paediatric diabetic retinopathy?, British Journal of Ophthalmology, 2012; 96(3), 312-317.
3 Klein R. The Epidemiology of Diabetic Retinopathy: Findings from the Wisconsin Epidemiologic Study of Diabetic Retinopathy, International Ophthalmology Clinics, 1987; 27(4), 230-238.
4 Bek T, Erlandsen M, Jesper, Mehlsen, Mogens, Per L and Poulsen. Identification of independent risk factors for the development of diabetic retinopathy requiring treatment, Acta Ophthalmologica, 2011; 89(6), 515-521.
5 Klein B, Klein R, Lee K and Myers C. 15-Year Cumulative Incidence and Associated Risk Factors for Retinopathy in Non-diabetic Persons, Archives of Ophthalmology, 2010; 128(12), 1568-1575
6 Pinhas-Hamiel O and Zeitler P. The global spread of type 2 diabetes mellitus in children and adolescents, The Journal of Paediatrics, 2005; 146 (5), 693-700.
7 American Diabetes Association. Timing Is Everything: Age of Onset Influences Long-Term Retinopathy Risk in Type 2 Diabetes, Independent of Traditional Risk Factors, Diabetes Care, 2008; 31(10), 1985-1990.
8 Bell R and Hardy L. An Epidemiological Perspective on Type 2 Diabetes Among Adult Men, Diabetes Spectrum, 2004; 17(4), 208-214
9 Diabetes UK. (2010) Key statistics on diabetes. [Online] Available from: http://www.diabetes.org.uk/Professionals/Publications-reports-and-resources/Reports-statistics-and-case-studies/Reports/ (Accessed 01/11/11).
10 Glynn R, Lin J, Manson J, Nathan D, Ridker P, Rifai N and Schaumberg D. Inflammation and Progressive Nephropathy in Type 1 Diabetes in the Diabetes Control and Complications Trial, Diabetes Care, 2008; 31, 2338-2343.
11 Klein B, Klein R, Lee K and MyersC. ?15-Year Cumulative Incidence and Associated Risk Factors for Retinopathy in Non-diabetic Persons?, Archives of Ophthalmology, 2010; 128(12), 1568-1575
12 American Diabetes Association. An Association Between Subclinical Hypothyroidism and Sight-Threatening Diabetic Retinopathy in Type 2 Diabetic Patients?, Diabetes Care, 2010; 33(5), 1018-1020.