Burden and Predictors of Diabetic Nephropathy in an Adult With Diabetes Mellitus Patients on Follow up at Ambo

Background: Despite the growing evidence of Diabetic Nephropathy in adult patients with long-standing diabetes in sub-Saharan Africa, data on its burden and correlates in adult African patients with diabetes are limited. We, therefore, undertook this study to determine the burden and predictors of Diabetic Nephropathy among adult population with diabetes in Hospital. Objective: We, therefore, undertook this study to determine the burden and predictors of Diabetic Nephropathy among adult population with diabetes in Ambo University Referral Hospital Central, Ethiopia. Methods: A Hospital-based cross-sectional study was conducted from June to August, 2023 and a systematic random sampling method used to recruit participants. A total of 4,300 were interviewed using structured questionnaires to gather data from the study subjects. For data entry and analysis, Epi Data version 3.1.1 and SPSS version 25 were used. Multivariable logistic regression analysis was done after descriptive statistics. 95% confidence intervals (CI) and crude and adjusted odds ratios were presented. Results: Diabetic Nephropathy (DN) was documented in population was 24.9% from the 369 participants (95%; CI 23.37– 26.43) participants among the study). Male sex [AOR = 2.215; 95% CI: 1.34, 3.45, p = 0.002], physically inactive [AOR = 1.983; 95% CI: 1.05, 3.70), P = 0.034], dyslipidaemia [AOR: 1.98, CI: 1.009, 3.5), P = 0.024] and poor controlled glycemia [AOR= 2.70; 1.40, 5.2), P = 0.003] were significant determinants for the development of Diabetic


Introduction
Diabetes mellitus is one of the non-communicable diseases (NCDs) that account for 63% of mortality globally, which was originally considered a social illness.Globally, there is substantial evidence linking diabetes mellitus (DM) to the onset of a number of complications that raise the risk of premature death and place a significant financial burden on the health system [1] .
An American study found that the number of adults with diabetes would rise from 22.3 million (9.1%) in 2014 to 39.7 million (13.9%) in 2030 and 60.6 million (17.9%) in 2060 [2] .The prevalence of diabetes among people aged 20 to 79 has increased by more than threefold since 2021, according to the International Diabetic Federation (IDF), diabetes will affect 643 million people worldwide by 2030, or 11.3% of the entire population, and 783 million people, or 12.2% of the entire population, by 2045 if current trends continue [3] .
According to a meta-analysis, a study carried out in Iran between January 2000 and May 2020 showed that the prevalence of diabetic nephropathy in people with type 2 diabetes was 30.6% [4] .A cross-sectional study was carried out in Thailand between October 2016 and September 2017 to look at the prevalence and risk factors of chronic kidney disease (CKD) in 1,096 primary care type 2 diabetes (T2DM) patients.The study depicted the overall prevalence of CKD to be 24.4% (21.9-27.0) of < 60 mL/min/1.73m2 based on glomerular rate filtration measurements [5] .
An increasing burden of DN has been reported in adult patients with diabetes in sub-Saharan Africa.Given that the illness is a major underlying cause of cardiovascular disorders, renal failure, and mortality, this is a crucial component of the consequences of the disease's outcome [1] .According to a study conducted in Nigeria, out of 5078 patients, 1937 patients had DN (prevalence: 38.4%) [6] .
In Ethiopia, between 1990 and 2017, a study demonstrated an increase in the prevalence of DM and the microvascular and macrovascular issues it causes in diabetic people.Diabetic complications can be caused by a number of variables, including prolonged disease duration, lower socioeconomic status, the existence of additional issues, and advanced age [7] .Besides, study done in Gondar, Ethiopia showed that the incidence rate of diabetic nephropathy was 14 (95% CI 10.8-17.7)cases per 10,000 [8] .The incidence rate of diabetic nephropathy was also found to be 14 (95% CI 10.8-17.7)instances per 10,000 patient-month observations, according to a study conducted in Gondar, Ethiopia.The development of diabetic nephropathy occurred in 63 (13.6%)DM patients as well [9] .
Diabetic nephropathy may appear at the time of diabetes diagnosis.The connection between diabetes mellitus and problems in the microvascular and macrovascular tissues is one of the most significant clinical characteristics.The primary contributing element to the onset of organ damage is the degree and duration of long-term hyperglycaemia.
Nephromegaly and a modified Doppler are two early morphological indicators of renal impairment, although proteinuria and Glomerular Filtration Rate are the strongest indicators of how damaged the kidneys are (GFR) [10] .
The most important elements in the progression of DN are the level of blood pressure and glycemic control, dyslipidaemia, cardiovascular factors.However, recent studies have shown that, in addition to the classical albuminuric DKD phenotype, two new nonalbuminuric phenotypes of DKD exist, i.e., nonalbuminuric DKD and progressive renal decline, suggesting that progression of DKD can also occur through a non-albuminuric pathway and Patients with proteinuria >3 g/day reach the main end goal more quickly than those with baseline proteinuria <3 g/day [11] .
About 25-35% of people with type 1 or type 2 diabetes mellitus develop diabetic nephropathy.From hyper filtration to micro-albuminuria, macro-albuminuria, nephrotic proteinuria, and lastly to progressive chronic kidney disease, which eventually results in end-stage renal failure, the disease develops through numerous clinical phases.The glomerulus, the tubules, the vasculature, and the interstitial spaces of the kidney are all commonly affected by structural pathological alterations during these stages [12] .
Numerous epidemiological studies show that the main risk factors for the development of diabetic nephropathy are family history, high blood pressure, dyslipidaemia, obesity, and insulin resistance.Furthermore, glycated hemoglobin level (HbA1c), systolic blood pressure, proteinuria, and smoking are risk factors [13] .
According to research conducted in Tigray, Ethiopia, having hypertension, having poor glycemic control, having had diabetes for a long time before being diagnosed, and not adhering to diabetic medication, diet, and exercise were all significant predictors of diabetic nephropathy [14] .Although numerous studies have been conducted in Ethiopia, many important associated factors such as smoking, family history, high blood pressure, dyslipidaemia, glycosylated hemoglobin, and obesity were overlooked.Additionally, there is no recent data on diabetic nephropathy in the research area.Therefore, the study objective was to determine this study aimed to investigate the prevalence, and predictors of diabetic nephropathy in an adult population with diabetes.
In sub-Saharan Africa (SSA), the high costs of increased mortality and morbidity in this population as well as the cost of kidney replacement therapy provide a strong economic benefit for improving early detection of DN with diabetes a top priority in in sub-Saharan Africa (SSA) [15] .In order to undertake a program that intends to lower the incidence and mortality of the disease as part of targeted interventions, it is necessary to have accurate information about DN.

Eligibility criteria
The study included all adult with type I diabetes who had it for longer than five years, had proven type II diabetes, were age of 18 and above, were being followed up in the study area, and could provide their informed consent.Patients who were seriously ill and had a severe cognitive or hearing impairment were excluded from the study.

Sample size determination
A single population proportion formula with a 95% confidence interval, 35.3% of a systematic review and meta-analysis of diabetic nephropathy from sub-Saharan countries, and a margin of error of 5% were used to compute the sample size [16] i.e.

Sampling technique and procedure
The study subject in Ambo University Referral Hospital was selected by systematic random sampling.Six participants were excluded because of incomplete Laboratory investigations.
Variables used in the study Data collection tool and procedure Data were collected by using a structured interviewer administered questionnaire which was developed by researchers from relevant literatures.Behavioural variables were assessed based on WHO Step wise approach for chronic disease risk factor surveillance [17] .Clinical variables were taken from patient record review and physical measurements were conducted.
Following standardised study procedures, biophysical measurements which included resting blood pressure and relevant anthropometric measurements (weight, height) and body mass index [BMI] were then performed.

Biophysical Measurements
A fasting blood sample was then collected for the measurement of blood glucose (FBG), glycated haemoglobin (HbA1c), lipid profile, and serum creatinine (for estimation of the e-GFR) using electro-chemiluminescence immunoassays manufactured by Roche diagnostics Limited, Germany on a Cobas 311 C-model SN 14H3-15 machine (Hitachi High Technologies Corporation, Tokyo Japan).Glomerular filtration rate (GFR) was measured using Cockroft-Gault formula (140-age [yr]) X body wt [kg] X K/serum creatinine [|xmol/L]), K = 1.23 for men, 1.05 for women [18] .These values were then corrected for body surface area (1.73 m2).

Operational definition
Hypertension: if the SBP/DBP was >140/90mmHg and/or or patients on antihypertensive therapy.

Data entry, processing and Statistical Analysis
Data were categorized, cleared, compiled and coded, checked for completeness, accuracy then entered into Epi data version 3.1 and then exported to SPSS.Both bivariable and multivariable logistic regression analysis was done and variables that were significant in bivariable with a p-value of <0.25 were retained for further consideration with multivariable logistic regression to control confounders.Finally significance of statistical association was assured using 95% confidence interval and a p-value of (<0.05) was considered significant in multivariable regression.The necessary assumption of model fitness during logistic regression was checked using Hosmer-Lemeshow goodness-of-fit test statistics.

Data quality control
The principal investigator provided data collectors with a two-day intensive training on sampling techniques and the purpose of the study before they began collecting actual data.A questionnaire that was originally written in English was translated into the working regional language (Afaan Oromo), and then, in order to ensure accuracy, it was translated back into English by someone with strong two-language translation skills.On the basis of the pre-test, the study tool underwent all necessary revisions.The tool was pre-tested on 5% of the sample size at the nearby Holeta Hospital.For the data collection, experienced enumerators were hired, and a selected participants were given an introduction to the study.The two supervisors and the principal investigator conducted ongoing follow-up and supervision, and they also reviewed the collected data.
The analysis of all blood samples in the laboratory followed standard operating procedure.In accordance with the manufacturers' instructions, the tests were carried out.To guarantee a high-quality outcome, the pre-analytical, analytical, and post-analytical stages of quality assurance were all applied.To prevent cross contamination, visual checks of the lab's and working bench's cleanliness are made.The daily results were accurately recorded, and the principal investigator followed up every day.

Limitations of the Study
Cross-sectional research does not strongly suggest a cause-and-effect link.The most accurate nephropathy test is microalbuminuria, although it is not currently available.Second, for several variables, we relied on patient data that could have been affected by recall bias.We also had difficulties figuring out the precise age at which diabetes first manifested itself.
Another drawback was that kidney biopsy, the gold standard diagnostic test, was not carried out.

Socio-demographic characteristics of participants
A total of 369 diabetic patients participated in the study over the research period, with a response rate of 100%.The participants' average age was 46 years old.Males made up 192 (52.0%) of the participants, or nearly half.Urban residents made up the majority of participants (228,61.8%).(In Table 1).

Burden of diabetic nephropathy
In this study, the eGFR scores were used for defining burden of DN for each study subject and Diabetes Nephropathy was then defined as score of ≤60ML/Minutes.Accordingly, the overall Burden of DN among the study population was 24.9% from the 369 participants (95%; CI 23.37-26.43).

Independent predictors of Diabetic Nephropathy
In this study population, we report that DN was relatively common.In the multivariate logistic regression analysis, variables like sex, physical activity, dyslipidemia, and glycemic control were included since bivariate analysis revealed evidence of some association with the outcome variable at a p value of 0.25.The predictor variables that were found to be strongly linked to the development of DN at a p value of < 0.05 were; being male sex, poorly controlled glycemia, dyslipidemia and being physically inactive.
One of the independent predictor of DN was individuals' male sex.Participants of the male sex were two times more likely

Discussion
We found that Diabetes Nephropathy (DN) was highly widespread in this study population, that it is a public health burden, and that it is one of the main reasons of hospital admission and death in Ethiopia.We documented that Diabetes Nephropathy (DN), a serious public health concern that accounts for a considerable portion of hospital admissions and 96)20.353 (1− 0.353) (0.05) (0.05) = 351 and by adding 10% for nonresponse the final sample size was 369.
cross-sectional study conducted from June to August, 2023 among 369 diabetic patients attending their follow-up at chronic illness clinic of Ambo University Referral Hospital, Central Ethiopia.The West Shewa Zone is expected to have a total population of 2,058,676 people in 2018/2019, of which 1,028,501 are men and 1,030,175 are women, according to the Ethiopian census conducted in 2007.In this zone, there were 520 health posts, 92 health centers, and 9 hospitals.

Table 2 .
Clinical and behavioral characteristics of participants atAmbo University Referral Hospital Central Ethiopia, 2023 (n=369).

Table 3 .
than those of the female sex to develop DN (AOR = 2.215; 95% CI: 1.34, 3.45).After adjusting for other factors, Participants who were physically inactive had a twofold increased risk of developing DN compared to their counterparts [AOR = 1.983; 95% CI: 1.05, 3.70].Diabetic patients who were dyslipidemia two times more likely to develop DN than those who had not [AOR: 1.98, CI: 1.009, 3.5].Finally Patients with poor controlled glycemia were 2.70 times more likely to develop DN than the good controlled glycemia [AOR= 2.70; 1.40, 5.2] (In Table4).Bivariate and Multivariate analysis to identify predictors of DN, at Ambo University Referral Hospital Central Ethiopia, 2023 (n=369).