Utility of albumin to creatinine ratio in screening for microalbuminuria among newly diagnosed diabetic patients in Uganda: a cross sectional study

Background The aim of this study was to determine the prevalence and factors associated with microalbuminuria among newly diagnosed diabetic patients in Mulago National Referral Hospital, Uganda. Methods In this cross-sectional study conducted between June 2014 and January 2015, we collected information on patients' socio-demographics, biophysical profile, blood pressure, biochemical testing and echocardiographic findings using a pre-tested questionnaire. Bivariate and multivariate logistic regression analyses were used to investigate the association of several factors with microalbuminuria. Results Of the 175 patients recruited, males were 90(51.4%) and the mean age was 46±15 years. Majority of patients had type 2 DM 140 (80.0%) and the rest had type 1 DM 35 (20.0%). Mean glycated hemoglobin (HbA1C) was 13.9±5.3%. Mean duration of diabetes was 2 months. Prevalence of microalbuminuria was 47.4 % (95% CI: 40.0%-54.9%) overall. Pregnancy was associated with microalbuminuria (OR7.74[95%CI.1.01–76.47] P=0.050) while mild and moderate physical activity at work were inversely associated with microalbuminuria respectively (OR0.08[95%CI0.01–0.95] P=0.046) and (OR0.07[95%CI0.01–0.77] P=0.030). Conclusion Prevalence of microalbuminuria was high in this group. Physical activity at work may be protective against microalbuminuria and this calls for longitudinal studies. Early detection and management of microalbuminuria in diabetics may slow progression to overt diabetic nephropathy (DN).

the next decade 6,13 . Therefore, early detection and appropriate interventions in asymptomatic individuals may help in preventing deterioration in renal function, progression to diabetic nephropathy and ESRD 5 . Therapeutic interventions which reverse microalbuminuria include intensified glycemic control, use of Angiotensin Converting Enzyme (ACE) inhibitors and these should be initiated in diabetics with microalbuminuria to prevent progress to overt diabetic nephropathy 12 . DN, the end result of microalbuminuria, is a major cause of morbidity, premature mortality, end stage renal disease, need for renal replacement therapy, cardiovascular diseases, and escalating health-care costs in diabetic patients 6,12,[14][15][16][17][18] . The prevalence of DN is increasing steeply along with the diabetes epidemic 15 . Approximately one third to half of patients with diabetes develops renal manifestations 14,15,19 . DN may be more frequent among patients in Africa as compared to those in the developed world due to delayed diagnosis, limited screening and diagnostic resources, poor glycemic control and inadequate treatment of microalbuminuria 14,19,20 . From studies in the western world, the factors associated with microalbuminuria have been elucidated. 5,6,12 . In Africa, there is paucity of data on the prevalence and factors associated with of microalbuminuria among diabetic patients 5,14 . Therefore this study sought to determine the prevalence and factors associated with microalbuminuria among newly diagnosed diabetic patients at Mulago National Referral Hospital in Uganda.

Study design and participants
This was a cross-sectional study among 175 newly diagnosed diabetic patients at Mulago National referral hospital in Uganda conducted between June 2014 and January 2015. DM was diagnosed using the standard criteria of fasting blood sugar of ≥6.9 mmol/L. All newly diagnosed diabetic patients aged 18 years and above attending the diabetic clinic or admitted to the medical wards of Mulago hospital during the study period who met the inclusion criteria and provided informed consent were recruited consecutively. We excluded patients with urinary tract infection and patients who were unable to provide information.

Operational definitions
Microalbuminuria was defined as Albumin to Creatinine Ratio (ACR) between 30 and 299 mg/g Type1 DM: This referred to patients who required insulin and were depending on insulin for glucose control since diagnosis. Type 2 DM: Patients were classified as having type 2 diabetes mellitus if they required oral hypoglycemic agents or usage of combination of insulin and the oral hypoglycemic agents for glucose control.

Study setting
The study was carried out in the diabetic out-patient clinic, the medical endocrine ward and the medical emergency ward of Mulago National referral Hospital. Mulago is a National Referral Hospital in Uganda and Teaching Hospital for Makerere University with a bed capacity of 1500. Mulago Hospital receives referrals from all parts of the country including referrals from neighboring countries like Southern Sudan, the Democratic Republic of Congo, Rwanda among others. The study population is representative of the Ugandan diabetic population.

Sample size estimation
Using prevalence of 17% for microalbuminuria among African Health Sciences Vol 19 Issue 1, March, 2019 diabetic patients as determined by Son MK et al 1 , a sample size of 180 was estimated using the Kish Leslie (1965) formula. N= Z2P(1-P) D2 N = Sample size 180 Z = 1.96, the normal value corresponding to the 95% confidence interval P = 0.17, prevalence from the above study, D = 0.05, Acceptance error. However data from 5 participants was not included because it was incomplete. Therefore we analyzed data from 175 participants.

Clinical assessment
We took a focused history and performed a specific physical exam to determine the biophysical measurements. Information gathered was entered into a pre-tested questionnaire. We assessed the following factors: patients' demographic data, history of hypertension, age, physical exercise, marital status, date of diagnosis of DM, drug history, occupation, education level, and Last Normal Menstrual Period (LNMP). Weight was measured using a Secco weighing scale to the nearest Kg, height was measured in meters using a non-stretchable tape and these were used to compute Body Mass Index (BMI). Waist and hip circumferences were measured and waist to hip ratios determined for all patients. Glycated haemoglobin (HbA1C) was measured by automated high performance liquid chromatography. Other investigations included urinalysis and Microalbuminuria using ACR. Echocardiography parameters were acquired using a commercially available machine, Phillips HD11XE (Eindhoven, The Netherlands) with 2-D, MMode and Doppler capabilities was used according to the American Society of Echocardiography 21 . All equipment used were calibrated to avoid measurement bias.

Assessment of microalbuminuria
We explained to participants the procedure for collecting a mid-stream urine sample. Each participant was given two urine containers and instructed to provide 2 separate samples of mid-stream urine each measuring 10 ml. One of the containers was sterile and the sterile sample was used for urinalysis including urine microscopy. We excluded participants who were found to have urinary tract infection and we started them on antibiotics. The second container collected a spot urine sample for measurement of urine albumin to creatinine ratio (ACR). Microalbuminuria was defined as ACR between 30 and 299 mg/g.

Blood pressure measurement
Blood pressure was measured using a mercury sphygmomanometer and an average of the 2 readings was used in the analysis. We instructed each participant to sit on a chair with both feet resting on the floor uncrossed and rest for 5 minutes before taking the 2 blood pressure readings. The two blood pressure readings were recorded 5 minutes apart using a standard manual dekametaccoson MK3 mercury sphygmomanometer. The patient exposed the arm from which blood pressure was to be measured and rested it on a table at the level of the heart. A blood pressure cuff (with a bladder length > 80% of the arm circumference) was placed 2cm above the antecubital fossa. We inflated the cuff while feeling for the patient's radial pulse. The estimate of the systolic blood pressure was obtained by inflating the cuff 30 mmHg above the point of disappearance of the radial pulse. We then placed the diaphragm of the stethoscope on the antecubital fossa and the cuff deflated at 2 mmHg per second while listening for korotkoff sounds.
Korotkoff sounds 1 and 2 represented systolic and diastolic blood pressures respectively. For cases where Korotkoff sounds remained audible despite complete deflation of the cuff, abrupt muffling of the sounds was used to denote diastolic blood pressure. We recorded the average of the 2 blood pressure readings. Participants were declared hypertensive if they; were on anti-hypertensive medication, had history of hypertension and/or evidence of hypertension (blood pressure ≥140/90 mmHg). Committee (SOMREC) of Makerere University College of Health Sciences. All study participants provided written informed consent for involvement in the study. Enrolment was totally free and voluntary and participants were free to withdraw at any time without any consequences to them. The patients' records/information was anonymized and de-identified prior to analysis.

Data management and analysis
Data was double entered in a database developed with EP-IDATA version 3.1, validated and inconsistences cleared then exported to STATA 13 for analysis. Continuous data was summarized using measures of central tendency while categorical data was summarized as frequencies and percentages and presented in tables. Prevalence was presented as percentages with their confidence intervals. Comparisons were made using the student t-test for the continuous data and xhi-square or fisher's exact test for the categorical data. The outcome was dichotomized as having microalbuminuria or not then logistic regression was used to determine the association between the predictors and microalbuminuria. This was presented as Odds ratio (OR) and their 95% confidence interval (CI). Only factors with P-value <0.2 at bivariate analysis were considered for multivariate analysis. Multivariate logistic regression was performed; interaction was assessed for using the Chunk test. Confounding was assessed for using a 10% difference between the crude and adjusted models. Significance was at p value of 0.05 and less.

Results
Characteristics of newly diagnosed diabetic patients at Mulago National referral hospital who participated in the study. This study recruited 175 newly diagnosed diabetic patients between June 2014 and January 2015. Of these, 90(51.4%) were males. The mean age of all the participants was 46±15 years. Majority of patients had type 2 DM, 140 (80.0%) and the rest had type 1 DM, 35 (20.0%). The mean HbA1C was 13.9±5.3%. Mean duration of diabetes was 2 months. (Table 1 and table 2).   Table 3).

Discussion
In this study, we found a high prevalence of microalbuminuria among diabetic patients who were newly diagnosed with the disease.  al, they had a prevalence of microalbuminuria of 44.4% among type 2 diabetic patients 1 . This high prevalence may be due to a period of latency for patients with type 2 DM before diagnosis. Chowta et al in a cross-sectional study reports that in type-2 diabetics, microalbuminuria ranges from 8-47% 12,24,25 . These findings are in agreement with our findings for patients with type 2 DM. Three other cross-sectional studies among type 2 diabetic patients in Nepal and Bangladesh consistently reported a high prevalence of microalbuminuria ranging from 30-47% [26][27][28] . The high prevalence of microalbuminuria among type 2 diabetic patients cannot be over emphasized in Uganda, a country with majority of diabetic patients having type 2 DM. Jean Jacques added more evidence to the fact that approximately one third to half of patients with DM develops renal complications 14,29 .
In our patient population, age had no effect on the degree of MA and Lampropoulou reached a similar conclusion in their study where age showed just a weak correlation with MA 16,30 . Although microalbuminuria was more common in participants 40 years and older in our study, this is predictable because age and diabetes duration for more than 10 years are well known risk factors for the development of diabetic nephropathy 16 . We found the prevalence of microalbuminuria to be more common among females and Okpere et al came to a similar conclusion in their study of Nigerian adolescents. Similarly patients who had central obesity in our study had a higher prevalence of MA compared to those without central obesity, Okpere reports similar findings 5 . The prevalence of microalbuminuria we found of 47.4% is 3 times higher than that of the general population which is reported to be 10 -15% 5 . However most of this evidence is from the developed world and in Africa, there is paucity of data on the prevalence of microalbuminuria both in the general population and among diabetics 5 .
Among diabetic patients, microalbuminuria predicts the development of overt diabetic nephropathy 5 . Therefore a high prevalence we found predicts a great burden of renal disease for our patient population if no interventions are done. In our study, microalbuminuria had no significant association with BMI and this is in keeping with finding by Chowta et al who found that there is no effect of BMI and sex on the prevalence of microalbuminuria 12 . Likewise sex had no significant association with microalbuminuria in this study. Patients who exercised mildly and moderately during work were less likely to develop microalbuminuria. This makes exercise a possible protective factor against microalbuminuria among diabetics. Pregnant patients in our study were seven times more likely to have significant microalbuminuria compared with those who were not pregnant. The possible reason could be the pregnancy associated protein loss and likely hypertension in pregnancy. These are potential confounders in our study.
Physical activity at work was inversely associated with microalbuminuria in this study. In SSA, this is one of the first studies to describe a relationship between microalbuminuria and physical activity at work. To better understand the long term effect of physical activity at work on kidney function among diabetic patients in Africa, there is need for longitudinal studies. We found no association of microalbuminuria with glycemic control as measured by HbA1C. This is different from findings by Alleyn. In their study, microalbuminuria occurred in association with poor glycemic control and elevated blood pressure 6 . In addition hypertension had no significant association with microalbuminuria in this study which is in contrast with findings by Alleyn. The differences could be due to the differences in age for the two populations. Alleyn assessed a predominantly young population of type 1 diabetics while we assessed a population with mean age of 46±15 years. In addition Alleyn followed up their cohort for 2 years yet ours was a cross sectional study 6 . Gender was not associated with microalbuminuria in this study although the prevalence of microalbuminuria was higher among females than males; this difference was not statistically significant. Chowta et al reported similar findings: There was no correlation between gender and microalbuminuria in type-2 diabetes mellitus 12,31 .
Studies in the western world show a positive correlation between degree of microalbuminuria, BMI and blood pressure 12 ; these findings are different from evidence generated in our study where there is no association of microalbuminuria with hypertension and BMI. Asadujjaman et al reached a similar conclusion is a cross-sectional African Health Sciences Vol 19 Issue 1, March, 2019 study in Bangladesh which found no association between microalbuminuria and BMI among diabetic patients 28 . However, Khadka et al in a cross sectional study among type 2 diabetic patients in Nepal found hypertension to be associated with microalbuminuria 26 . Evidence shows that DN is more frequent among patients in Africa as compared to those in the developed world. Possible reasons include delay in diagnosis, limited screening/diagnostic resources, poor control of blood sugar and blood pressure and inadequate treatment at an early stage 14,19,20 . However, evidence on the burden of kidney diseases in people with diabetes in Africa remains very patchy 14 . A meta-analysis in sub-Saharan Africa has found the prevalence of DN to be 36% 32 , while in China DN is only present in 8% of diabetic patients 33 . We recognize that the lack of a non-diabetic control population is a limitation of our study; however, this study was not designed to determine the effect of diabetes on the kidney, but rather to explore the magnitude of subclinical microalbuminuria among newly diagnosed diabetic patients in the SSA context, including the factors associated with microalbuminuria. The short duration of the study could have obscured seasonal variability. This was due to limitations in logistics: however newly diagnosed diabetic patients were recruited consecutively over a seven months period. We did not analyze data from 5 participants because it was incomplete. This limitation could have affected the power of the study.

Conclusion
Prevalence of microalbuminuria was high in this patient population of newly diagnosed diabetes mellitus. Pregnancy was positively associated with significant microalbuminuria while physical activity at work was inversely associated with microalbuminuria. Early detection and management of microalbuminuria may help in preventing deterioration in renal function and development of overt diabetic nephropathy and progression to ESRD. Physical activity at work may have a protective effect against microalbuminuria and DN; longitudinal studies on this subject are needed.

Source of funding
Research reported in this manuscript was supported by the Fogarty International Center of the National Institutes of Health under award number R24TW008861. Dr Martin Muddu was also supported by the Fogarty International Center and the National Heart, Lung, and Blood Institute (NHLBI) at the National Institutes of Health under the Global Health Equity Scholars Consortium at Yale University (D43TW010540). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.