G6PD deciency is associated with renal impairment in T2DM

Background Diabetes and its complications including renal impairment are the consequences of increased oxidative stress. Glucose-6-phosphate dehydrogenase (G6PD) is responsible for protection from harmful oxidative metabolites. G6PD deciency hence leaves these individuals at high risk of oxidative stress and at high risk of diabetes and its complications including renal impairment. If any association between G6PD deciency and damage to kidney exists in diabetic population then screening for G6PD deciency in this population will help identify the susceptible patient who may be at higher risk of diabetic complications due to G6PD deciency and can prevent them with early interventions to help decrease patients’ morbidity in the long run. Methods Healthy control volunteers and diabetic with or without renal impairment were selected after counseling. Body Mass Index and blood pressure were recorded. Blood tests to measure the G6PD activity in fresh whole blood, hemoglobin, glycated hemoglobin, uric acid, and creatinine were done. Urine was tested for the presence of proteinuria. The activity of G6PD was measured. Glomerular ltration rate (eGFR) was estimated with the Modication of Diet in Renal Disease Study (MDRD) Equation. Calculations and statistical analysis were done with open software LibreOce 6.3.3 and SPSS version 22.0. Results The mean G6PD level of healthy control was 84.27 ± 9 and that of diabetic patient without renal impairment diabetic patients renal impairment were 65.05 ± 5.35 and 46.64 ± 6.56 respectively. There only that the prevalences of G6PD deciency were 4.55%, 22.73%, and 59.09% in diabetics, and diabetics with renal in our The healthy control participants showed prevalence of G6PD deciency similar to previous studies. The proportions of G6PD decient among the diabetic participants and diabetic with renal impairment were numerically much higher than even the ethnic group Rajbanshi who were reported to the highest prevalence of G6PD deciency in The difference statistically relevant in participants with T2DM < 0.01) in with only T2DM = 0.09). T2DM with and G6PD deciency


Abstract
Background Diabetes and its complications including renal impairment are the consequences of increased oxidative stress. Glucose-6-phosphate dehydrogenase (G6PD) is responsible for protection from harmful oxidative metabolites. G6PD de ciency hence leaves these individuals at high risk of oxidative stress and at high risk of diabetes and its complications including renal impairment. If any association between G6PD de ciency and damage to kidney exists in diabetic population then screening for G6PD de ciency in this population will help identify the susceptible patient who may be at higher risk of diabetic complications due to G6PD de ciency and can prevent them with early interventions to help decrease patients' morbidity in the long run.
Methods Healthy control volunteers and diabetic with or without renal impairment were selected after counseling. Body Mass Index and blood pressure were recorded. Blood tests to measure the G6PD activity in fresh whole blood, hemoglobin, glycated hemoglobin, uric acid, and creatinine were done. Urine was tested for the presence of proteinuria. The activity of G6PD was measured. Glomerular ltration rate Results The mean G6PD level of healthy control was 84.27 ± 9 and that of diabetic patient without renal impairment and diabetic patients with renal impairment were 65.05 ± 5.35 and 46.64 ± 6.56 respectively.
There was only one participant (4.55%) with G6PD de ciency among healthy controls whereas the proportion of participants having normal, mild to moderate, and severe degree of G6PD de ciency were 77.27%, 13.64%, and 9.09%, respectively for diabetic patient without renal impairment while 40.91%, 22.73%, and 36.36% for those with renal impairment. The proportion of participants with renal impairment had more severe degree of G6PD de ciencies than those without renal impairment (p=0.03).
Conclusion The prevalence of G6PD de ciency was higher in diabetics with renal impairment than those with diabetes alone suggesting the relation between G6PD activity and diabetes with renal impairment. Background Type 2 Diabetes Mellitus (T2DM) was previously seen as a disease of "Western" countries but is also a growing problem in developing countries. Numerous studies have identi ed various risk factors related to T2DM 1 . Understanding the pathophysiology of T2DM has helped identify the population at risk so that primary preventive programs can be successfully implemented to reduce the risk. It is, however, not understood why some are vulnerable to develop this disease and its complications while others are spared. There are evidences that point to important roles of oxidative stress in the pathogenesis of T2DM 2 and its role further in development of diabetic complications 3 . Hence, weakness of antioxidant defense system of the body could be one of the reasons for vulnerability to T2DM and its complications.
The Glucose-6-Phosphate Dehydrogenase (G6PD) enzyme is one of the major components of intracellular antioxidant system -the pentose monophosphate pathway 4 . A decrease in G6PD activity makes cells very sensitive to damage from oxidation 4 and has also been linked to risk of developing T2DM 5,6 and its complications 7 . Role of G6PD de ciency and its relation to pathogenesis of kidney disease 8 has been studied and shown to play a part in injury to kidneys in animal models [9][10][11] . None of the previous studies have examined the association between G6PD de ciency and renal injury in T2DM though there is evidence that there is of some degree of occult acute kidney injury in G6PD de cient children during acute hemolytic episodes 12 .
G6PD de ciency is a common enzyme de ciency whose prevalence is 0.23% 13 in general population of Nepal and up to 6.3% in eastern Nepal 14 and 11.7% in certain ethnic group of Nepal 15 . The abnormalities in G6PD genotype is different than other types in South Asia and described as G6PD Mediterranean-Middle East subtype. It is yet not known if G6PD de cient population of Nepal is prone to diabetes and its complication. It will also be interesting to know whether any relation between G6PD de ciency and damage to kidney exists in diabetic population. If any relevant association is seen, it will further pave a way to recognize the population at risk and study interventions to reduce this risk. This study aims to know the prevalence and degree of G6PD de ciency in healthy, diabetic, and diabetics with renal impairment to ascertain whether G6PD is a risk factor for T2DM and renal impairment in Nepali population.

Methods
This is a cross-sectional comparative study done in out-patient setting in Kathmandu. Consecutive diabetic individuals above 40 years of age who had come for their regular blood tests in two centers were approached for their willingness to participate in the study. Participants were classi ed as diabetic if they were already under treatment for T2DM and renal impairment if the glomerular ltration rate (eGFR) estimated by Modi cation in Diet in Renal Disease (MDRD) formula was below 60 ml/min/1.73 m 2 . They were asked about presence of any other active disease and ongoing medical treatment. Those subjects with any symptoms of active disease, recent blood transfusions, dialysis, and consumption of herbal or antioxidant medications were excluded from the study. For the study, 22 consecutive diabetics with no renal impairment and 22 consecutive diabetics with renal impairment were recruited. For the controls, 22 adults above 40 years of age and without any current medical illness or taking any medication and who had come for regular health checkup were recruited. The sample size was calculated with a 95% con dence level assuming a prevalence of 0.23% of G6PD de ciency in Nepalese adults which was previously reported 13 .
The selected participants were interviewed about their history of T2DM, renal impairment, and smoking habits. Height, weight, and blood pressure were recorded. Blood samples were obtained by standard phlebotomy procedure and tested for hemoglobin (Hb), HbA1c, Uric Acid, Creatinine, and G6PD levels. Venous blood was collected in plain and EDTA tubes for serum and whole blood, respectively. For serum, samples in plain tubes were processed within 2 hours of collection. After clotting at room temperature, serum was separated by centrifugation at 3500 rpm for 10 min. Gross hemolytic or icteric serum samples were rejected. Hb level was measured by Sysmex XN330 analyser. Biorad D10, based on High Performance Liquid Chromatography (HPLC) was used to determine HbA1c level. Creatinine and uric acid were measured by Randox Imola analyser. Creatinine assay was based on alkaline picrate without deproteinisation method whereas uric acid was based on uricase perioxidase with ascorbate oxidase method. G6PD assay was performed on CareStart Biosensor analyser based on electrochemical principle which measures G6PD enzyme activity from the electricity generated by reduction of NADPH. Urine was tested for presence of albumin in urine.
The participants were then classi ed as non-diabetic controls with normal renal function, diabetic with normal kidney function and without proteinuria, and diabetics with impaired renal function. The participants were classi ed as having normal, mild to moderate, and severe G6PD de ciency when they had G6PD levels of > 50 U/dL, 30 to 50 U/dL, and < 30 U/dL, respectively.
Ethical approval for the study was obtained from Nepal Health Research Council, Nepal. The participants were given detailed information about use of the anonymous data obtained from them and the use of blood and urine samples collected from them. The participants were given opportunity to decline from participating or withdrawing from the study at any time during the period of study.

Results
There were 66 participants with an average age of 56.26 ± 1.03 years. There were 22 individuals in each group of control, diabetic with normal renal function, and diabetic with renal impairment. The characteristics of participants are shown in Table 1.
The three groups were similar in age and gender composition and also in smoking habits ( Table 1). Prevalence of high blood pressure at the time of study was seen in 4.55%, 18.18%, and 22.72% among the control, diabetics without renal impairment, and diabetics with renal impairment, respectively.

Relationship between diabetes and G6PD de ciency
The control participants and participants with diabetes without renal impairment were compared ( Table  1). The two groups were similar in age (p=0.07) and sex (p=0.5) composition, smoking habits (p=0.24), body mass index (p=0.1), and prevalence of high blood pressure (p=0.34). The two groups were also similar in levels of Hb (p=0.22), serum creatinine (p=0.05), and uric acid (p=0.14). The participants with diabetes without renal impairment had a history of diabetes for 8.55 ± 1.35 years. The participants with diabetes had a lower average value of G6PD level compared to the healthy controls (p=0.02). There was only one participant with G6PD de ciency among healthy controls (4.55%) while 5 participants had G6PD de ciency among diabetic participants (22.73%) ( Figure 1); however, this did not have statistical signi cance (p=0.09).

Figure 1. Comparison between prevalence of different degrees of G6PD de ciency between three groups
A -control participants, B -diabetics with normal renal function, and C -diabetics with impaired renal function. The proportion of participants with different degrees of G6PD de ciency were signi cantly different in three populations with more severe degree of G6PD de ciency in C.

Relationship between renal impairment and G6PD de ciency
To examine the relation of renal impairment and G6PD de ciency, diabetic participants without renal impairment and diabetic participants with renal impairment were compared (Table 1). Each group had 22 participants with similar age (p=0.05) and sex (p=0.5) distribution and prevalence of high blood pressure (p=0.5). The participants with renal impairment had lower body mass index (BMI) (p<0.01) and longer history of diabetes (p=0.04) than those without renal impairment. However, the degree of glycemic control as measured by HbA1c was similar in two groups (p=0.48). The participants with renal impairment also had signi cantly lower Hb level (p<0.01) and higher serum creatinine (p<0.01) values.
The average value of G6PD was lower in participants with renal impairment than those without renal impairment (p=0.02). The proportion of participants having normal, mild to moderate, and severe degree of G6PD de ciency were 77.27%, 13.64%, and 9.09%, respectively for those without renal impairment while 40.91%, 22.73%, and 36.36% for those with renal impairment (Figure 1). The proportion of participants with renal impairment had more severe degree of G6PD de ciencies than those without renal impairment (p=0.03).

Discussion
G6PD de ciency is known to be the most common enzyme de ciencies in human. It is an essential component of antioxidant system 4 . Various diseases have been linked to oxidative stress and de ciency of G6PD [16][17][18][19][20][21] . This comparative study addresses association of G6PD de ciency, T2DM, and renal impairment.
T2DM is a global disease and is a growing public health problem 1 . A systematic review by Gyawali et. al. in 2015 had shown that the pooled prevalence of T2DM in Nepal was 8.4% (95% CI: 6.2-10.5%) which was in agreement with prevalence in different parts of the world. The review further reported variation of prevalence between 1.4% and 19% in different study populations in Nepal 22 . Similar variation has also been reported from other countries of South Asia 23 . CKD is one of the most common complications diabetics are prone to develop and is seen in nearly 50% of the patients 24 . Various biologic, socioeconomic, and lifestyle factors play important roles in pathogenesis of diabetes and its renal complication affecting some and sparing others even in seemingly homogeneous population. This study examines G6PD de cient condition as one of the factors that needs to be considered.
G6PD de ciency in Nepal was studied previously in relation to malaria. The molecular abnormalities in G6PD genotype found in Nepal are unique in South Asia and belong to the G6PD Mediterranean-Middle East subtype 25 . The prevalence of G6PD de ciency was found to be 0.23% 13 in general Nepali population and up to 11.7% 15 in certain ethnic group of Nepal.
Our study has shown that the prevalences of G6PD de ciency were 4.55%, 22.73%, and 59.09% in healthy adults, diabetics, and diabetics with renal impairment in our study population. The healthy control participants showed prevalence of G6PD de ciency similar to previous studies. The proportions of G6PD de cient among the diabetic participants and diabetic with renal impairment were numerically much higher than even the ethnic group Rajbanshi who were reported to have the highest prevalence of G6PD de ciency in Nepal 14 . The difference was statistically relevant in participants with T2DM with renal impairment (p < 0.01) but not in participants with only T2DM (p = 0.09). So, this study clearly shows an association between T2DM with renal impairment and G6PD de ciency while suggests possibility of relation between T2DM and G6PD de ciency.
This study recruited random participants visiting the out-patient department. A population-based study would have given better understanding of distribution of G6PD de ciency state and its correlation with T2DM and chronic kidney disease (CKD). This study shows that G6PD de ciency is more common among cohort with T2DM and T2DM with renal impairment but does not tell if T2DM and CKD is more prevalent in people with G6PD de ciency than those with normal activity of G6PD. Such study will require recruitment of a large population as the prevalence of G6PD is as low as 0.23% in general population.
Study in ethnic group with higher prevalence of G6PD de ciency 14 like Rajbansi and Tharu is desired in such situation.
Lower activity of G6PD in our participants is assumed to be due to genetic mutation. There are reports of transient lowering of G6PD activity in different conditions 26 and also when blood glucose level is high 27 .
One of the limitations of our study is that we cannot verify that the G6PD de ciency seen in our participants is due to genetic mutation and not due to high level of blood glucose. The patients with only T2DM and with T2DM and renal impairment had similar control of blood sugar as determined by HbA1c levels (p = 0.48), but the difference in G6PD activity between these two groups ( Table 1) were statistically signi cant (p = 0.02). This shows that the difference of G6PD activity between the two groups is not due to difference in blood sugar levels. Further con rmation of genetic mutation in the participant will need genetic workup of the participants which will also tell if our population has unique mutation as shown before 25 .
The implication of G6PD in primary and secondary prevention of diabetes and complication also needs to be understood. Prospective follow-up of healthy volunteers without T2DM but with G6PD de ciency and diabetic volunteers without CKD but with G6PD de ciency can tell us if G6PD de ciency state is of health concern. Study of factors affecting variation in G6PD activity in human and interventions that can alter the G6PD activity will be other areas of interest if G6PD de ciency is found to affect the outcome in prospective studies.

Conclusion
In conclusion, this study compared the prevalence of G6PD de ciency among healthy population, diabetics, and diabetics with renal impairment. The level of activity of G6PD was lower in participants with T2DM and T2DM with CKD. The prevalence of G6PD de ciency state was also only numerically higher in patients with T2DM than healthy participants while the difference in prevalence was statistically higher in diabetics with renal impairment than those with T2DM alone. This shows relation of G6PD Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
The study was funded by Samyak Diagnostic Pvt. Ltd. for laboratory analysis and collection of data.
Authors' contributions NBB conceptualized the study, prepare the protocol, collected and analyzed the data and prepared the manuscript. SP prepared the proposal, obtained the ethical approval and collected and analyzed the data. RKK revised the manuscript. KG conceptualized the study. All authors read and approved the nal manuscript.

Figure 1
Comparison between prevalence of different degrees of G6PD de ciency between three groups Acontrol participants, B -diabetics with normal renal function, and C -diabetics with impaired renal function. The proportion of participants with different degrees of G6PD de ciency were signi cantly different in three populations with more severe degree of G6PD de ciency in C.