Prevalence and Determinants of Diabetic Retinopathy Among Type 2 Diabetic Patients in Saudi Arabia: A Systematic Review

Diabetic retinopathy (DR) is a preventable complication of diabetes; however, it is a serious one if not early recognized and properly managed as it can lead to visual impairment. This review aimed to summarize the prevalence and determinants of DR among type 2 diabetic patients in Saudi Arabia. Eligible articles for this systematic review were quantitative observational studies that were English-published between 2015 and 2021, peer-reviewed, and conducted on patients with type 2 diabetes mellitus (DM). The studies were obtained by searching PubMed/Medline, ScienceDirect, Cochrane Library Database, and Google Scholar. The risk of bias was assessed using the National Institutes of Health quality assessment tool for observational cohort and cross-sectional studies. Out of 16 preliminary studies, 12 were eligible for inclusion in this systematic review. One study was a chart review, one was a prospective study, and the remaining were cross-sectional studies. Regarding the study tool, retinopathy was diagnosed by an ophthalmologist using fundus photography and/or slit-lamp examination in most of the studies (n=7). However, some studies reported obtaining data from patient interviews and medical files (n=4). Concerning the outcome, an overall high prevalence of DR (ranging between 6.25% and 88.1%) and some significant associated risk factors were determined, including longer duration of diabetes, older age, poor blood pressure control, poor glycemic control, and physical inactivity. Most studies showed moderate overall quality. In conclusion, DR is a common complication of type 2 diabetes in Saudi Arabia. Some avoidable risk factors are identified, through which the doctors can identify patients at high risk of DR through early screening and can, thus, initiate prompt treatment to reduce the risk of vision deterioration.


Introduction And Background
According to the World Health Organization, diabetes mellitus (DM) is defined as "a metabolic disorder of multiple etiologies characterized by chronic hyperglycemia with disturbances of carbohydrate, fat, and protein metabolism resulting from defects in insulin secretion, insulin action, or both" [1].
Diabetes mellitus is a worldwide chronic health problem affecting more than 366 million persons, and the prevalence will expectedly reach 552 million by the year 2030 [2]. The Kingdom of Saudi Arabia occupies the seventh rank globally regarding the prevalence of diabetes and the second rank in the Middle East. Current statistics indicate that approximately seven million Saudi people are diagnosed with diabetes, and another three million are in the prediabetes stage [3].
Diabetes mellitus causes several complications, including diabetic retinopathy (DR), which affects almost 18.5% of diabetic patients all over the world [4] and is considered a leading cause of blindness in diabetics in the age group 20-74 years [5].
Commonly reported risk factors for DR were older age, higher levels of glycated hemoglobin (HbA1c), dyslipidemia, obesity, longer duration of diabetes, smoking, nephropathy, and high blood pressure [6,7].
In Saudi Arabia, the prevalence of DR ranged from 28.1% to 45.7%, and vision-threatening DR affects 4.5% to 17.5% of patients with diabetes [8]. The global prevalence of DR was estimated to be 22.27% (95% confidence interval (CI), 19.73%-25.03%), according to a meta-analysis that included 59 population-based studies [9]. In the United Kingdom, a study reported a DR prevalence of 38.9% in 2012 which decreased to 36.6% in 2016 [10]. In Denmark, results of the DR screening program showed that the prevalence and fiveyear incidence rate of DR among type 2 DM patients were 8.8% and 3.8%, respectively [11]. Based on the rates from 41 studies (48,995 patients), a recent meta-analysis [12] found that the prevalence rates of DR, proliferative DR, and non-proliferative DR were, in order, 28%, 6%, and 27% in Asian type 2 DM patients. The meta-analysis found a variation in the prevalence rates of DR among Asian countries, with the highest rates being in India (42%), Malaysia (35%), and Singapore (33%) [12].
Diabetic retinopathy is characterized by signs of retinal ischemia (retinal microvascular abnormalities, intravenous caliber abnormalities, hemorrhages, microaneurysms, neovascularization, and cotton-wool spots) and/or signs of increased retinal vascular permeability. Depending on signs, retinopathy is grouped into two main categories: non-proliferative diabetic retinopathy (NPDR "mild, moderate, and severe") and proliferative diabetic retinopathy (PDR). Loss of vision in DR can result from many mechanisms, including neovascularization resulting in vitreous hemorrhage and/or macular edema, retinal detachment, and retinal capillary nonperfusion [13].
The American Diabetes Association recommends screening type 2 diabetic patients for DR at the time of diagnosis of diabetes and then routinely every one or two years if there is no evidence of retinopathy. However, patients showing signs of retinopathy should undergo retinal examinations more frequently, with the time intervals decided according to their stage of DR [14].
Diabetic retinopathy is a preventable complication of diabetes; however, it is a serious one, if not early recognized and properly managed, as it can lead to visual impairment. Unfortunately, most patients are asymptomatic until the very late stages. Consequently, the disease can rapidly progress before the initiation of therapy leading to poor outcomes. Therapy can produce favorable results in the early stages of DR, hence comes the necessity of screening for DR for early identification of at-risk patients who will then benefit from prompt treatment.
There is a need to define the magnitude of the problem of DR in Saudi Arabia to draw the attention of doctors providing care to type 2 diabetic patients so that they become alert for identifying such patients. Also, the risk factors of DM should be defined, both to identify high-risk patients and to change any modifiable risk factors to improve visual outcomes in diabetic patients. Moreover, Saudi health authorities can benefit from the evidence provided by primary and secondary research to assess the feasibility of initiating screening programs for DR among high-risk patients and training doctors in diabetic clinics. Therefore, this systematic review aimed to summarize the prevalence and determinants of DR among type 2 diabetic patients in Saudi Arabia. The objectives of this systematic review included: (a) estimating the prevalence of DR among type 2 diabetic patients in Saudi Arabia and (b) identifying the significant risk/preventive factors that increase/reduce the risk of developing DR among type 2 diabetic patients in Saudi Arabia.

Review Methods
A thorough search was conducted using relevant keywords such as ("Diabetic retinopathy" OR "diabetic complications") AND ("prevalence" OR "rate" OR "risk factors") AND "Saudi Arabia" AND ("type 2 diabetes"). Studies were extracted from the online databases of PubMed/Medline, ScienceDirect, Cochrane Library Database, and Google Scholar.
Articles were first screened by abstract, and then the screening of the full-text articles was based on eligibility. Criteria for article selection included: peer-reviewed publications, studies conducted on Saudi patients with type 2 DM, articles that investigated the prevalence and associated factors of DR, studies carried out recently between 2015 and 2021, full articles should be in English, and quantitative observational studies. Other studies were excluded from this systematic review, including ongoing studies, reviews, conference proceedings, and abstracts without full-text studies.
According to the specified inclusion and exclusion criteria, articles were screened, and the data were then extracted and included in this review. The preliminary search yielded 16 articles. Four articles were excluded (one review article, one article on type 1 diabetic patients, and two abstracts without full-text studies).
Primary outcome measures were the prevalence of DR and its risk factors (demographic and diabetes/clinical-related factors).
The risk of bias (ROB) of the included studies was assessed using the National Institutes of Health quality assessment tool for observational cohort and cross-sectional studies [15].

Results
Out of 16 preliminary studies, 12 were found eligible for inclusion in this systematic review. Four articles were excluded due to different reasons, including review articles or including only type-1 diabetic patients ( Figure 1). Only one study was a chart review, and one was a prospective study, whereas the remaining were observational cross-sectional studies. Regarding the study tool, retinopathy was diagnosed by an ophthalmologist using fundus photography and/or slit-lamp examination in most of the studies (n=7). However, some studies reported obtaining data from patient interviews and medical files (n=4), and in one study, the nature of the study tool was not clear. Concerning the outcome, the prevalence of DR and its associated factors, either patients' demographics, diabetes-related factors, or general health-related factors, were assessed in these studies. The main findings revealed an overall high prevalence of DR (ranging between 6.25% and 88.1%), and some significant associated risk factors were determined ( Table 1).  The results of the ROB assessment revealed that two studies had a higher ROB (lower methodological quality) [19,23], while the remaining studies had moderate quality with less ROB ( Table 2).

Discussion
The prevalence of DM has increased dramatically in Saudi Arabia over the last few decades. Consequently, the rate of DM-related complications, including DR, is expected to increase [28]. This systematic review summarizes the results concerning the prevalence and determinants of DR in Saudi Arabia between 2015 and 2021.
The variations in the rates of DR varied widely among the studies, which may be attributed to differences in the study designs as most studies were cross-sectional in the current systematic review [16][17][18][19][20][21]23,[25][26][27], while two studies were longitudinal [22,24]. Also, heterogeneity among studies regarding the patients' sociodemographic and clinical characteristics could have a role in explaining this variation in the prevalence rates, particularly the age of the included patients and the duration of DM. The duration of DM varied among the studies and would impact the development of DR; thus, the rate of DR will be lower if patients with short durations of DM were included, as was noticed in the study by Al-Zamil [24]. Another potential contributing factor for this variation is the utilization of different diagnostic tools for diagnosing DR, either clinically through the assessment with a slit-lamp and colored fundus photography or through data obtained from medical records. In addition, the study setting is an important factor for determining the prevalence of DR, as carrying out a study in a specialized diabetic center usually yields higher rates [18,25] compared to primary care settings [17,22,23] and general hospitals.
The current systematic review summarized the identified risk factors for DR among the Saudi population as stated by the included studies. Identification of potential risk factors is of paramount importance in reducing the rate of DR by helping identify the patients at risk to design frequent screening schedules and rigorous management of modifiable risk factors to hinder disease development or progression.
The most notable risk factor was poor glycemic control in univariate [8,17,18,20,[22][23][24][25]27] and multivariate analyses [18,20,22,27]. Elevated HbA1C levels are known to correlate with a higher risk of DM-related complications such as nephropathy, neuropathy, and retinopathy [41,42]. Good control of glycemic status can reduce the risk of developing vascular complications and/or reduce their severity if they develop [6,43,44]. The effect of strict glycemic control can last for several years after treatment as shown by the extension to the Diabetes Control and Complications Trial [45].
The older age of patients was significantly associated with a higher probability of DR in most studies in univariate analysis [17][18][19][20][21][22][23]25,27] and was confirmed as an independent risk factor on multivariate analysis in two studies [22,27]. However, it was found insignificant in multivariate analysis in the study by Bajaber and Alshareef [20]. The age did not differ significantly in two of the studies [16,24], but one study included only newly diagnosed cases with DM [24] and the other assessed the age within risk groups based on the HbA1C level [16]; thus, the effect of age may not become apparent in these two studies. Longer duration of DM increased the risk of DR in univariate [17,18,20,22,25,27] and multivariate analyses [18,20,22,25,27].
Only one study reported a lack of significant correlation between the duration of DM and DR [16], which may be explained by the assessment of risk factors within subgroups of patients. A study in the United States found that each year following the development of DM increases the risk of DR by 6% [13]. The effect of the older patient's age and longer duration of DM could be explained by the longstanding hyperglycemia, which induces macro-and micro-vascular effects and the aggravation of hyperglycemia when the pancreatic function further deteriorates with age. This explanation may also apply to the effect of insulin treatment found in the current review, as insulin is usually required for the treatment of advanced cases in which oral hypoglycemics do not produce good glycemic control, because of pancreatic deterioration, which is common with older age and after several years of suffering from DM [46]. However, insulin might increase the progression of DR by increasing the permeability of the microvasculature of the retina [47]. Treatment with insulin was found to significantly increase the risk of DR both on univariate [18,20,22,23,25,27] and multivariate analyses [18,20,22,25,27]. All these findings emphasize the need for early diagnosis and strict management of DM to reduce the risk of developing DR or at least delay its onset and severity. Screening programs for DM and raising the Saudi public's awareness about the importance of regular health check-ups, healthy lifestyle practices, and treatment adherence can play a pivotal role in reducing the incidence of DMrelated complications, including retinopathy.
Male sex was a risk factor for developing DR in three studies on univariate analysis [18,21,27], but there was no significant sex difference in six studies [16,19,20,[22][23][24]. The presence of sex differences in the risk of DR is a controversial issue, and there is no clear explanation for the observed male predominance in some studies [48].
Hypertension was a significant risk factor of DR in univariate [17][18][19][20]22,23,25] and multivariate analyses [20,22]. This finding could be explained partially by the common coexistence of hypertension and DM. In addition, hypertension may contribute to the retinal pathophysiological changes with the formation of retinal hard exudates and hemorrhages [49]. However, two previous clinical trials did not find a significant effect of tight control of blood pressure on the progression of DR [50,51].
The presence of other complications of DM seemed to increase the risk of developing DR, particularly nephropathy [18,22,23,25,27] or neuropathy [25,27]. Their effect was noted in multivariate analysis in one study only [27]. This association is well documented in previous research and reflects the common pathophysiology of these complications, which involve macro-and micro-vascular involvement [52].
Some factors showed conflicting results, such as smoking, dyslipidemia, and obesity, which significantly reduced the risk in the study by Al-Rubeaan et al. [27] on both univariate and multivariate analyses. The association of obesity and body mass index with the development and/or progression of DR is contradictory in the literature [48]. The protective effect of overweight was hypothesized to reflect the observation that patients with poorer glycemic control are usually underweight, while strict glycemic control is commonly associated with increased body weight [48].
The identified risk factors for developing DR among the Saudi population are in accordance with those reported in other populations. A longer duration of DM and a higher body mass index increased the likelihood of developing DR, as shown in studies conducted in the United Arab Emirates [29] and Iran [32]. Noncompliance with the intake of anti-diabetic medications as well as regular annual eye examinations were shown as risk factors for DR [53]. Furthermore, diabetic patients treated with insulin were more likely to develop DR [53,54]. The association between other complications of diabetes such as neuropathy, nephropathy, and cardiovascular diseases on one side and DR on the other side has been documented by earlier studies [55][56][57]. The association between uncontrolled diabetes and DR has been documented by other researchers [32,57].
The results of this systematic review should be interpreted cautiously in light of the limitations and potential biases observed in the included studies. Limitations of these studies included a lack of generalizability of findings as being single-facility studies or limited to certain patients' age [16,18,19,21,22], small sample sizes [19,[22][23][24], being cross-sectional studies that doesn't prove causation [17,18,20,[25][26][27], and depended on obtaining information from medical records rather than ophthalmic examination [16,17,27]. Another important limitation was the lack of sample size calculation in several studies [19,[22][23][24]. In addition, some studies did not state the number from which the sample was chosen and whether any eligible subjects were excluded for some reason [16,19,23,24]. Moreover, several studies did not assess different levels of independent factors (e.g., duration of DM, HbA1C levels) on DR [17,19,21,23,24,26]. Another important source of bias is whether the outcome assessors were blinded to the patients' history and medical condition, which was not stated in any of the studies. Also, half the studies did not perform a multivariate analysis to adjust for confounding factors and find the independent factors increasing the risk of DR [17,19,21,23,24,26]. Future studies should avoid these limitations to achieve highquality evidence regarding the prevalence of DR in patients with type 2 DM in Saudi Arabia and the associated independent risk factors.

Conclusions
Overall, the findings from this systematic review indicate that DR is a common complication of type 2 diabetes in Saudi Arabia with a prevalence rate ranging from 6.25 to 54.6%, which is comparable to the reported worldwide rates, despite the variation across the Saudi studies. Several risk factors were identified, including unavoidable factors such as age, gender, and duration of diabetes, while some others are avoidable, such as poor glycemic control, poorly controlled hypertension, obesity, non-compliance with anti-diabetic medications, and annual eye check-ups. The findings of the current systematic review helped to realize the magnitude of the problem in Saudi Arabia and can guide the health authorities in implementing future preventive and management projects. The preventive projects should initiate screening for DR in all high-risk patients and increase the awareness of primary care physicians, family physicians, and specialists in DM to encourage their patients and refer them for routine screening. Also, the identified preventable risk factors should be addressed to hinder the development and/or progression of DR.
However, the quality of the available studies is impaired by several limitations in their design and methodology, necessitating the conduction of future longitudinal Saudi research to determine the potential association between various risk factors and the development of DR. Future studies should preferably be prospective cohorts in design, ensure adequate sample size, and sufficient follow-up. Studies should endeavor to assess the less commonly evaluated preventable risk factors, such as smoking and physical activity. In addition, adjustment for confounding factors is of paramount importance to identify the independent risk factors of DR.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.