THE PREVALENCE AND RISK FACTORS ASSOCIATED WITH TYPE 2 DIABETIC RETINOPATHY IN RWANDA

Objective: To determine the prevalence and risk factors associated with type 2 diabetic retinopathy in Rwanda. Methods: A case-control study was conducted from January to September 2019 in four hospitals within the Republic of Rwanda. Type 2 diabetic patients were screened for retinopathy. Patients with retinopathy were considered as cases and those without retinopathy as controls. A study sample of 592 participants were enrolled, 66 cases and 526 controls. Diabetic retinopathy was assessed by indirect ophthalmoscopy performed for each eye with a slit-lamp +90D lens. Plasma glucose and Glycated hemoglobin (HbA 1c ) were measured by colorimetric enzymatic tests. Albuminuria was measured by quantitative spectrophotometric method. Triglycerides, Total, HDL and LDL cholesterol, Urea and Creatinine were assayed by colorimetric methods. A questionnaire was used to assess medical history and demographic status. Data were analyzed by use of SPSS version 20. Statistical analyses were performed by Chi square to show association between nominal and ordinal data. Multivariate logistic regression analysis to select independent variables was performed. Odds ratio was used as a measure of association. Results : The prevalence of diabetic retinopathy in this study population was 11.2%. Independent risk factors associated with diabetic retinopathy were long duration of diabetes, hyperglycemia, LDL cholesterol, triglycerides, and albuminuria. Conclusion: The prevalence of diabetic retinopathy is high and is associated with traditional modifiable risk factors. Early detection and management of diabetes mellitus and these risk factors, combined with good adherence to scheduled eye examination would greatly improve the quality of life of affected patients.


403
patients and is the fifth leading cause of blindness globally (Glover et al. 2012). Whereas in developed countries the prevalence and incidence of DR is well documented, there is scanty published data on prevalence of diabetic retinopathy from sub-Saharan Africa (Glover et al. 2012). In the US, DR is documented as the main cause of vision loss in people aged 25-74 years (Abdhish et al. 2012), and in India its prevalence in 2014 was found to be about 22% (Gadkari et al. 2016).
In African populations, studies showed the prevalence of DR to be between 30.2 -31.6%, proliferative DR 0.9 -1.3%, and maculopathy 1.2 -4.5%. In clinical surveys, the reported prevalence range for DR was 7.0 -62.4%, proliferative DR 0 -6.9%, and maculopathy 1.2 -31.1% (Burgess et al. 2013). Recent studies conducted in Zimbabwe and Ethiopia reported a prevalence of 28.4% and 16% respectively (Pasipanodya et al. 2017, Tsegaw et al. 2021). In Rwanda the national prevalence of T2DM is about 3.16%, and mortality rate 2% per annum (WHO, 2016). In the Rwanda countryside there are no recent studies on ocular complications. Based on studies done in Kigali University Teaching Hospital, 20.4% of the patients with renal failure requiring hemodialysis suffered from T2DM (Igiraneza et al., 2018). The prevalence of diabetic retinopathy (diabetes mellitus type 1& 2) was 23% (Rudasingwa et al., 2012). The aim of this study was to determine the prevalence and risk factors associated with retinopathy in type 2 diabetic patients in Rwanda.

Study Design
A case-control study was conducted from January to September 2019 in four hospitals of Rwanda namely: University Teaching Hospital of Butare, Ruhengeri hospital, Kabutare and Kabgayi district hospitals. The patients were recruited from the routine diabetes clinics in these hospitals. They were consecutively consented to participate and the study information availed to them. Participants were screened for retinopathy and those with a positive result categorized as cases while those negative became controls. Of the 592 participants enrolled, 66 were cases and 526 controls.

Study Population
All type 2 diabetic patients receiving healthcare at the four mentioned hospitals were recruited. The inclusion criteria were age 25 years and above, with fasting plasma glucose above 7.0 mmol/l, be on regular oral antidiabetic drugs for at least 6 months, and willing to give informed consent and able to communicate orally.
Study tool: Type 2 diabetic patients were identified through hospital registries. The participants were interviewed guided by a questionnaire that collected information on medical history, demographic status, and anthropometric measurements performed.
Albuminuria was assayed by quantitative spectrophotometry. For positive albuminuria cases, full blood count, erythrocyte sedimentation rate and urinalysis were performed to rule out other causes of albuminuria. Specimens with urinary infection were not confirmed for albuminuria. Their albuminuria assays were run one month later after the infection cleared. The C -reactive protein was assayed by qualitative method using CRP reagent kit.
Diabetic retinopathy was assessed by indirect ophtalmoscopy performed for each eye with a slitlamp +90 D lens (Shin-Nippon, 2010, Japan).

404
Blood pressure was measured using a digital sphygmomanometer (AMRON, MI plus, Hem-4011C-E, China). Body Mass index (BMI) was calculated as follows: BMI= Body weight (Kg)/ Height (m 2 ). Waist and hip circumferences were determined using a tape measure (cm).

Ethical Consideration
The study was approved by the relevant Institutional Review Boards before commencement as follows: College of Medicine and Health Sciences, University of Rwanda (Permit no. 288/CMHS IRB/2018) and Kenyatta National Hospital/University of Nairobi (Ref.KNH/ERCA3).

Data analysis:
Data obtained were analyzed on SPSS version 20. Statistical analysis to determine interactions of risk factors was performed by Chi square (x 2 ). Multivariate logistic regression analysis was performed for independent variables. Odds ratio (OR) was used as a measure of association. Statistical significance was set at p < 0.05.

Results:-
Of the 592 type 2 diabetic patients, 526 were controls and 66 cases. The mean age for controls and cases was 56 years (SD ± 0.50464, CI 95%) and 58 years (SD ± 1.46768, CI 95%) respectively. The age range for the controls was 26-86 while in the cases it was 34-89 years. Twenty eight (42.42%) of cases and 169 (32.13%) of controls were males. Majority of the participants resided in the rural area: 37 (56.06%) cases and 322 (61.21%) controls. Forty eight (72.72%) out of the 66 cases and 264 (50.19%) of controls were hypertensive. Forty six (69.69%) out of 66 cases and 356 (67.68%) out of 526 controls had a history of alcohol consumption.
407 . The studies in Zimbabwe, Ethiopia and Uganda were conducted in a single hospital, while the current one was conducted in four hospitals and may therefore be said to be more representative of the Rwanda's population. The study conducted in Kenya reported higher prevalence compared to ours, this difference may be attributed to various factors such as the age of participants which was ≥ 50 years (Wanjiku et al., 2014) compared to the present study where the range was 26-89 years.
In the present study, the univariate analysis for hypertension was associated with diabetic retinopathy compared to multivariate logistic regression analyses which was not. The lack of statistical association between hypertension and diabetic retinopathy here is similar to the results of earlier studies (Pasipanodya et al., 2017, Tsegaw et al., 2021. This demonstrates the effect of better control of blood pressure on reduction of incidence and progression of diabetic retinopathy as reported in the previous studies (Glover et al., 2012, Stratton et al., 2001.
The duration of type 2 diabetes mellitus was an independent risk factor associated with diabetic retinopathy in the current study similar to previous studies (Tsegaw et al., 2021, Seyoum et al., 2001, Macky et al., 2011). In the current study there was increased chance of developing diabetic retinopathy after 11years of being diagnosed with type 2 diabetes. Also hyperglycemia was an independent risk factors associated with DR which is similar to previous studies (Pasipanodya et al., 2017, Gill et al., 2008.
Elevated LDL cholesterol and high triglycerides in univariate and multivariate analyses were independently associated with DR. This is similar to findings of other studies that high LDL cholesterol and high triglycerides were independent predictive factors for retinal complications in diabetic patients (Hadjadj et al., 2004, Wong et al., 2008. This study also found that significant albuminuria indicated by microalbuminuria and macroalbuminuria was strongly associated with DR. The study found that diabetic patients with microalbuminuria and macroalbuminuria had 7.78 and 10.83 times respectively, the risk of developing DR compared to those without albuminuria. A similar association between albuminuria and DR was reported in a previous study (Padmaja et al., 2011). The association between advanced degrees of DR and macroalbuminuria was reported by other authors (Wirta et al., 1999, Looker et al., 2003.

408
The present study showed that positive Creactive protein and high levels of plasma creatinine were associated with DR in univariate logistic analyses, but not in multivariate analyses. C-reactive protein is an acute-phase protein and is mainly synthesized by the liver or adipose tissue when microbial infection or tissue damage occurs (Genes, 2010). The measurement of CRP is useful in clinical settings for the diagnosis and treatment of some acute or chronic inflammatory diseases (Lim et al. 2010). Some studies reported higher levels of CRP in patients with proliferative diabetic retinopathy than in non proliferative diabetic retinopathy (Chen, 2010, Cai et al., 2006. In this study, Creactive protein was affected by duration of T2DM and the severity of diabetic retinopathy where the majority of DR, 8.6% had mild retinopathy. The severity of DR was associated with progressive renal function decline and albuminuria progression (Hayne et al., 2019). The strong association between DR and chronic kidney disease (CKD) was reported in an earlier study and the presence of CKD almost always accompanies DR (Parving et al., 1988). Elevated creatinine and microalbuminuria as markers of renal function increase with increasing levels of retinopathy (Kamran, 2018). In the current study plasma creatinine was also affected by duration of T2DM and severity of DR.
This study also found that obesity as indicated by BMI ≥ 30 kg/m 2 and waist:hip ratio >1, was not associated with DR. On meta-analysis Wei Zhu and colleagues found an association between obesity and DR (Zhu et al., 2018). A population-based cohort study found that BMI was associated with DR in risk, age and sex adjusted multivariate models (Grauslund et al. 2009), and increased waist:hip ratio was associated with DR in women (Raman et al., 2010). The lack of association between obesity and DR in the current study was attributed to general population fitness, where the majority of the patients came to hospital on foot and are in general of ectomorph body constitution as confirmed by the low prevalence of obesity in Rwandan population which was 5.8% in 2016 (Index Mundi, 2016).

Conclusion:-
The prevalence of diabetic retinopathy in this study was 11.2% (mild non proliferative 8.61%, moderate non proliferative 1.7%, severe non proliferative 0.34% and proliferative retinopathy 0.5%) in type 2 diabetic patients. The prevalence of cataract and maculopathy were 13.2% and 3.7% respectively. The major independent risk factors associated with diabetic retinopathy were long duration of T2DM, hyperglycemia, LDL cholesterol, triglycerides, and albuminuria (microalbuminuria and macroalbuminuria) It is expected that the management of these modifiable risk factors of diabetic retinopathy shall help significantly decrease its incidence or delay its onset. The Adherence to annual eye examination in diabetic patients for early diagnosis and management of DR and other causes of visual impairment such as cataracts and maculopathy is highly recommended to improve the quality of life of patients.