Global Impact of Obesity and Diabetes on the Increase Incidence and Prevalence of Chronic Kidney disease (CKD) and End Stage Renal Disease (ESRD): A Systematic Review

Background: Obesity and diabetes are the main causes of chronic kidney disease (CKD) and end stage renal disease (ESRD). The objective of this study was to analyze the impact of obesity and diabetes on CKD and ESRD incidence and prevalence. Methods: A comprehensive literature search was conducted from 2001 to 2018. 494 articles were retrieved via PubMed and 125 articles through Google scholar and reference list of the selected articles. Among which thirty (30) studies met our inclusion criteria consisting of 17 cohorts, 11 cross-sectional, and 2 case-control studies. Results: Majority of the studies indicated direct relationship between body mass index (BMI) and ESRD risk. Notably, the association of obesity and diabetes potentially increases the incidence and prevalence of CKD and ESRD. Results from the cohort, case-control and cross-sectional studies pointed out a positive association between obesity, diabetes and risks for renal disease outcomes. Even though many complications may occur, renal transplantation (RT) is still the preferred renal replacement therapy (RRT) advised in multiple studies for diabetic ESRD patients. Renal transplantation was associated with better quality of life and survival advantage than dialysis. Interestingly, overweight and obese ESRD patients on dialysis had a signicant survival advantage in comparison to lean body weight patients. Conclusion: Taken together, obesity and diabetes are signicantly associated with the increasing incidence and prevalence of CKD and ESRD. Regulation of Weight and diabetes are highly recommended in obese and diabetic patients to prevent the subsequent renal disease. Previous reviews have discussed the relationship between obesity and ESRD or diabetes and ESRD separately. However, importantly, this review gives an insight on the association between obesity, diabetes and CKD/ ESRD.

The search terms included BMI or obesity, diabetes or diabetes type-2 or diabetic nephropathy, end stage renal disease or ESRD, chronic kidney disease or CKD, renal replacement therapy or RRT, dialysis or peritoneal dialysis or hemodialysis or PD or HD and kidney transplantation or renal transplantation or RT.

Selection criteria
The manuscript published in English language as full text articles were included in the study. The searched articles by PubMed and Google scholar were analyzed initially by titles closely related to de ne the obesity, diabetes and their impact on renal disease speci cally the ESRDs. Articles with not well-de ned titles were reviewed only on the abstract level. Furthermore, we made an additional search to see potential eligible studies through reference list of review articles that might have been missed in the initial searching. All population-based studies addressing the impact of obesity and diabetes on CKD and ESRD incidence and prevalence were included in this review. We excluded the studies published in language other than English and those who addressed only type-1 diabetes (Insulin Dependent Diabetes). The risk of bias across the studies might exist, but to our best, we tried to include all those studies related to our topic, irrespective of their statistically signi cant data or journal of publication to avoid the selection bias and publication bias. All the authors were quali ed and experienced enough to follow the protocol for the selection of the studies.

Data collection and extraction
In this study, (AK) independently screened all the retrieved titles and abstracts as part of the search strategies to identify potentially eligible articles and subsequently veri ed by (SG). The quality of methodology and the risk of biasness in the included studies were assessed by two authors and the disagreements were discussed and resolved in the weekly meetings. Data extraction was carried out for different variables, including information about the author, type of study, country, study period, sample size, mean age, study objectives, BMI, diabetes, type of renal disease, main ndings, ORs (odd ratios) or RRs (relative risks), statistical analysis, registry/data source. Remarks in the tables corresponds to conclusion. ORs and RRs were used as measures of the association between obesity, diabetes and CKD or ESRD.

Results
Literature search and inclusion Figure 1 presents Preferred Reporting Items for Systematic review and Meta-Analysis (PRISMA) ow diagram for the article selection [35]. We retrieved 494 articles concerning our topic through PubMed search, and 125 additional articles were identi ed through Google scholar and through the reference list of the selected articles. After the selection process, 30 eligible studies were included in this review.

General characteristics of the included studies
General characteristics of the chosen studies addressing obesity, diabetes and ESRD are shown in Table 1. Maximum number of the included studies were based on adult population consisting of 18 to 98 years, while, male ratio was higher than female. Importantly, by distributing the retrieved studies geographically according to the continents (Fig. 2), 13 (43.3%) studies were performed in North America, 9 (30%) in Europe, 7 (23.3%) in Asia and only one study was carried out in Australia.
Majority of the studies were carried out in North America (43.3%) followed by Europe (30%), Asia (23.3%) and the lowest studies were performed in Australia (3.3%).
Importantly, thirteen (13) studies identi ed that the increase in BMI (mainly the obesity) have an impact on the increase incidence of ESRD. In addition, 11 studies revealed that diabetes is the most prominent risk factor for the increase incidence of ESRD while in 06 studies the ESRD patients were both: obese and diabetic. Two studies identi ed that obesity, diabetes and additional risk factors such as hypertension, glumerulonephritis, smoking and proteinuria are the major causative agents of increasing ESRD incidence [28,36]. A case control study [6] in two ethnic groups (black and white) revealed the differences between black and white population in the incidence of ESRD in relation to the association between obesity and ESRD. Importantly, signi cant differences were observed in the study between the two groups regarding the increase odds of ESRD. Risk factors for CKD/ ESRD Several risk factors were identi ed in the selected studies which had an impact on the increase incidence and prevalence of CKD and ESRD (Fig. 3). However, obesity and diabetes were the most prominent risk factors point out by majority of the authors.
Majority of the studies observed that obesity and DM are the major risk factors for CKD and ESRD. Abbreviations: DM, Diabetes Mellitus; HT, Hypertension; CVD, Cardiovascular disease; PU, Proteinuria; MS, Metabolic syndrome; GN, Glumerulonephritis; Fam hist of ESRD, Family history of ESRD Obesity is associated with profound increased incidence and prevalence of CKD and ESRD Previously, the association between obesity and kidney disease has been described in several studies. Table 2 brie y summarizes the main characteristics and ndings of 13 eligible studies which describes the impact of obesity on end stage renal disease. Most of the studies (n = 7) were conducted in USA, 2 in Japan and one each in Korea, Norway, UK and Spain. The sample size included in the studies were varied signi cantly ranging from 125 to 100753 individuals [14,44]. The analysis indicates that obesity is the prominent risk factor for kidney disease including CKD [38,40,43] and ESRD [37,42,44,58] leading to the ultimate or instant need for renal replacement therapy such as HD and PD or renal transplantation (RT). In a community-based cohort of 2585 men and women, it was analyzed that each unit increase in BMI was associated with 23% (OR 1.23 95% CI, 1.08-1.41) increased risk of new onset renal disease [41]. The prevalence of CKD in the US in 1999-2004 was higher than 1988-1994. The cross-sectional analysis stated that prevalence of CKD increased from 10% in 1988-1994 to 13.1% in 1999-2004 with a prevalence ratio of 1.3 (95% CI, 1.2-1.4). This increase was due to the increase prevalence of diabetes and hypertension. This further raises concerns about future elevated incidence of kidney disease [38].
In a population-based study carried out by Evangelistq and colleague in South Korea determined that obesity was more prevalent in CKD patients than those without CKD. Importantly, prevalence rate of general obesity was 37.8% in stage 4 and 5 CKD patients. The study summarizes that weight loss is a good potential intervention to prevent the disease progression [40]. In a cohort of 11104 initially healthy men and 14 years of follow up, it was analyzed that higher baseline BMI was associated signi cantly with increased risk for CKD. The increase BMI (> 10%) was associated with signi cant increased risk for CKD (OR 1.27, 95%CI, 1.06-1.53) [12].
A case control study detected a signi cant difference between two ethnic groups: black and white in relation to the association between obesity and ESRD.
BMI of overweight and obese persons at the age 21 was associated with increased ESRD incidence in both black and whites but more prominent in whites than in blacks, while BMI (overweight and obese) at enrollment was associated with non-signi cant odds of ESRD in blacks. However, signi cantly, obese whites had 2-fold increase odds of ESRD (OR 2.17, 95% CI, 0.94-4.98) [37].
Obesity is linked directly or indirectly in the development of chronic kidney disease. The results from the study with 18.5 years follow up, where 36% of the participants were overweight and 12% obese, revealed that 7.9% developed stage 3 CKD and 14.4% proteinuria. One unit increase in BMI was associated with 5% increase in the odds of stage 3 CKD (OR 1.05 (1.02-1.09) P = 0.005 [39].
Obesity, smoking and lack of physical activity are the signi cant risk factors for CKD [43], and the relative risk (RR) for BMI ≥ 30 kg/m 2 was 1.77 (95% CI,1.47-2.14). Using data from a community-based screening of 100735 participants in Okinawa, Japan, it was noted that a higher BMI is associated with the increased ESRD risk in men (OR 1.273, 95% CI, 1.121-1.446) P = 0.0002 but not in women in the general population [44]. In summary, the obesity enhances the risk for developing the kidney disease particularly the ESRD in the general population. Furthermore, due to the large expenditures on the treatment of ESRD patients it is also an economic problem as well.
Diabetes is a prominent risk factor for the elevated burden of CKD and ESRD A wide range of studies has been carried out to nd the effect of diabetes on increasing kidney disease including ESRD. Table 3 [47,52]. By reviewing the summary of these studies, we concluded that diabetes is continue to be the prominent risk factor for kidney disease, including CKD [53,55] and ESRD [49][50][51][52] leading to an urgent need for renal replacement therapy that is HD and PD or RT. 2) however, type-2 diabetes is still a major cause of ESRD incidence [29].
Hochman and colleagues estimated the prevalence and incidence of ESRD in native American adults living on the Navajo nation using USRDS data. Higher prevalence and incidence were observed in native American adults living on the Navajo nation. Age adjusted prevalence and incidence of ESRD was 0.63% and 0.11% respectively. Majority of the ESRD patients were diabetic [52]. To assess the prevalence of CKD in a Mexican urban population, a population based cross sectional survey was conducted. Prevalence rate of CKD in Mexico was like those in the developed countries. Increase prevalence was partially due to DM, however other factors such as genetic and socioeconomic may also play a role. OR for DM as a risk factor for CKD in siblings was 1. In a study in Germany determined that incidences of ESRD in patients with and without diabetes were 157.9 and 25.6 per 100,000 person/year respectively (6.2-fold increased risk for those with diabetes). [61] Diabetic nephropathy was the most frequent reason for RRT (29.7%) and the relative risk of RRT in the estimated adult population was increased by 8-fold comparing the non-diabetic population [48]. Khan et al 2016, revealed a signi cant relation between ESRD and three major risk factors namely diabetes, hypertension and glumerulonephritis. Importantly, the ESRD in diabetic patients was 11.04 times more than nondiabetic patients [28].
The longitudinal study in Cyprus 2004-2011 found that 84.4 pmp (36.0%) with ESRD were due to diabetic nephropathy suggesting that diabetes is a major cause of ESRD and specially in population under 65 years of age [62]. Furthermore, in a retrospective study, it was analyzed that the use of RAAS blockers has a signi cant impact on the delay onset of ESRD in diabetic patients. In this study it was described that some DN patients not receiving RAAS blockers developed ESRD in two years, while those receiving RAAS blockers took an average of 7 ± 1.91Y to progress into ESRD. Mean duration for the onset of ESRD was 4.59 ± 1.50 Y for those who were not prescribed RAAS blockers. The statistically signi cant difference was observed between the two groups, P = 0.001 (95% CI -3.69 to-1.13) [47].
The results of the different studies have con rmed that the ESRD incidence and prevalence is much higher in the diabetic than the non-diabetic population, demanding serious efforts to combat diabetes in order to stop or slow-down the ESRD progression.

The coupling of obesity and diabetes, and their ultimate impact in overwhelming growth of CKD and ESRD
The impact of obesity on the increase in the ESRD events have been carried out in several studies. However, obesity in combination with diabetes leads to ESRD risk much quickly than alone. Table 4 brie y describes the main characteristics and ndings of 6 eligible studies in which the ESRD patients were obese and diabetic too. Four studies (2 each) were carried out in USA and Spain and one each in Sweden and Portugal. The sample size was varied among studies ranging from 237 to 615192 individuals [58,59]. among normal weight to 108/100000 persons/year among extreme obese (≥ 40 kg/m 2 ). Remarkably, baseline BMI remained a strong risk factor for ESRD even after adjustment for diabetes and blood pressure [15].
Highest prevalence of obesity stage 2 (BMI ≥ 35 kg/ m²) was observed in ESRD population with DM at dialysis initiation between the age of 45-64 years. It positively in uenced the ESRD population on dialysis due to the survival advantage with obesity [58].
Chen et al determined that obesity was associated with increased proteinuria in the early stage while it was bene cial in terms of improved renal survival in the later stages con rming the reverse epidemiology. 28.8% of the patients developed ESRD by the end of the study period [63].
A Cross sectional study observed that prevalence of CKD in Spain was high especially in the elderly population. Two modi able risk factors namely diabetes and hypertension were responsible for the increased prevalence of CKD. Association between clinical characteristics and the presence of CKD for obesity verses normal was OR 3.5 (95% CI, 2.0-6.0), hypertension verses absence OR 6.2 (95% CI, 4.0-9.6) and DM verses absence OR 2.0 (95% CI, 1.4-2.8) [60].
In another study, it was found that prevalence of CKD and cardiovascular risk factors was high in the randomly selected sample of the general population.  [59].
Interestingly, higher prevalence of CKD was not responsible for the high incidence of ESRD in Portuguese population. Infact the high prevalence of risk factors may account for the high incidence of CKD. The obesity prevalence was 33.7%, diabetes 11.7% and metabolic syndrome 41.5%. ESRD incidence was higher than other European countries but lower than the US. Adjusted OR (95% CI) for CKD: Diabetes = 1.20 (0.96-1.50) and Obesity = 1.14 (0.94-1.39) [57].
Even though some studies have shown that obesity is the major risk factor for developing ESRD, independent of diabetes. However, this review reveals that obesity enhances the risk for developing diabetes and they in combination give rise to the kidney disease particularly the ESRD in the general population.

Discussion
ESRD in patients with obesity and diabetes is a life-threatening disease with a poor survival rate and is associated with high healthcare costs. In this review, we speci ed a clear eligibility-criteria and conducted a comprehensive research to achieve the objectives.

Obesity and its contribution in ESRD development
The obesity epidemics and diabetes are growing worldwide. It has a strong affect across the globe and have far reaching social and health consequences.
Several studies carried out on this topic realized a strong co-relation between obesity, diabetes and kidney disease resulting that obesity and diabetes increases the risk of CKD and ESRD. Our systemic review based on retrospective, prospective, case control and cross-sectional studies, gives strong enough evidence regarding the unavoidable impact of obesity and diabetes on the increased growth of ESRD.
Fox et al evaluated that baseline BMI predicts subsequent kidney disease after a mean follow up of 18.5 years. In this cohort of 2285 men and women, increase of each unit in BMI was associated with 1.23-fold elevated risk for new onset kidney disease. Importantly in a cohort study (n = 100753), the relationship between obesity and the risk of kidney disease in men and women was analyzed. A strong dose-response relationship between BMI and risk of ESRD was found in men, but not in women [41].
In several other studies it was found that overweight and obesity is a common and strong risk factor for the development of ESRD in the general population. Furthermore, the increase in the BMI increased the rate of CKD, ESRD [15,32,39] and risk of chronic renal failure [56]. A case control study between two ethnic groups that is black and white showed a signi cant difference in the association between obesity and ESRD. Overweight and obesity at the age of 21 was associated with increased ESRD incidence in whites than in blacks. Strikingly, a 3-fold increase was observed in obese whites compared to normal weight person [37].
The risks for the adverse outcomes of obesity were progressive with increasing BMI. Furthermore, the obesity in the presence of DM increased the risk of graft failure. However, the study showed that obesity alone may also be a risk factor for a shorter time to graft failure [64]. Notably the prevalence of CKD in US in the year 1999-2004 was higher than 1988-1994 [38]. This increase in the prevalence was observed in the total sample regardless of their BMI state. As the data shows that 28% of the participants were obese, in our opinion, this would have been much better if the study had carried out on different BMI categories.
On the other hand, the cross-sectional study by Evangelista et al provides enough information about obesity prevalence in CKD patients. Importantly, obesity was higher in prevalent CKD patients than non-CKD. This support the idea that weight loss might be a good potential intervention for the avoidance of disease progression [40].
Obese patients with family history of ESRD were at higher risk of developing ESRD than non-family history of ESRD. Obesity and the start of dialysis therapy were independently associated with patients having family history of ESRD and genetic factor may also contribute to the familial risk of ESRD [42]. It was concluded after 14 years of follow-up that higher baseline BMI was linked with enhanced risk for CKD [12]. Similar ndings were carried out in another study where the baseline BMI was strongly and independently associated with rapid CKD progression [14].
The review study on "elevated BMI as a risk factor for CKD" summarized that the impact of obesity in the pathogenesis of CKD seemed to be independent of hypertension and DM [34]. Interestingly, the review study by Wang et al observed that obesity in women was associated with high risk than in man and a positive linkage was observed between BMI and risk for kidney disease [65].

Diabetes and its role in ESRD progression
Diabetes in ESRD patients is life threatening disease with low survival rate and high healthcare costs. The prevalence of diabetic ESRD is still on the rise while due to better management of healthcare system, the incidence rate has declined in the developed and some developing countries.
In the retrospective study, a high burden of CKD was observed among persons with undiagnosed diabetes and prediabetes. The prediabetes individuals need earlier detection and management strategies for the prevention of development, progression and complications of diabetes and CKD associated with DM [53].
Furthermore, early detection and treatment of DM can prevent the DM related ESRD, as DM has been the leading risk factor of incident dialysis in Japan since 1988 [49].
It was noted that diabetes related ESRD incidence was increased in the early 1990s in the US, however it decreased in the later years in all age groups due to reduction in prevalence of ESRD risk factors, better treatment and care. The ESRD patients with diabetes are better treated now than in the late 1990s. Similar ndings were observed during the examined period (1978)(1979)(1980)(1981)(1982)(1983)(1984)(1985)(1986)(1987)(1988)(1989)(1990)(1991), where the growth in ESRD incidence was higher than the growth in prevalent chronic renal insu ciency in the US [51].
In a cross-sectional study, it was analyzed that the prevalence of proteinuria was 4-fold higher in those with DM compared to those without DM, indicating that proteinuria is a good indicator of kidney damage. Furthermore, early investigation of proteinuria, hematuria and GFR in the initial stage of kidney disease may provide a mean to reduce the ESRD burden [54]. The ndings of Iseki et al strengthens the results of Chadban et al where a strong relationship between ESRD and proteinuria was found. It was concluded that proteinuria is a strong and independent risk factor for ESRD [50,54].
A cross sectional study was conducted by Hochman and colleagues to analyse the incidence and prevalence of ESRD in the native American adults living on the Navajo nation. Higher prevalence and incidence of ESRD were observed in native American adults living on the Navajo nation. Majority of the ESRD patients were diabetic and higher ESRD prevalence was noted than the incidence [52].
Moreover, to assess the prevalence of CKD in the Mexican urban population, a cross sectional study revealed that prevalence rate of CKD in Mexico was similar as in developed countries. The higher prevalence rate of kidney disease may be due to DM but other factors such as genetic and socioeconomic may also play a role. Diabetes was a prominent risk factor for CKD in the siblings [55].
A high proportion of RRT risk was due to diabetes [48]. Strong connection was found between ESRD and three major risk factors namely diabetes, hypertension and glomerulonephritis, and ESRF population was largely in uenced by age, gender and diabetes [28]. Type-2 diabetes was found to be the major cause of ESRD incidence and preventive strategies were strongly recommended to reduce the burden of ESRD incidences [29]. The review study by Ghaderian and colleagues concluded that renal transplantation, particularly preemptive transplantation is the best renal replacement therapy in diabetic ESRD patients. Although many complications may be associated with renal transplantation, but several studies recommended that it is associated with survival bene t and better quality of life [24].
Our review has some limitations which needs to be mentioned. Firstly, our search strategy only included PubMed and Google scholar, which might have resulted in the loss of some important articles related to our topic. Secondly, we did not include the studies that have discussed only type 1 diabetes, and also those studies other than the English language, which also might have resulted in the loss of some important studies.

Conclusion
Our systematic literature review describes the signi cant effects of obesity and diabetes on the increasing incidence and prevalence of CKD and ESRD. It was analyzed that overweight and obesity in younger age is markedly and positively associated with future treated ESRD incidence. Obese individuals having family history of ESRD are at much higher risk than the general population. Furthermore, diabetes, particularly type-2 diabetes is the major cause of CKD and ESRD incidence leading to RRT. In conclusion, the incidence and prevalence of CKD and ESRD in diabetic and obese population is more than the non-diabetic and non-obese population. We strongly recommend regular provision of health education and awareness trainings by the healthcare professionals on the prevention of CKD and ESRD, to control the CKD and ESRD incidence and prevalence in the future. Competing interests