Hemoglobin A 1 c levels in maintenance hemodialysis patients in India

Introduction Glycated hemoglobin (hemoglobin A1c; HbA1c) levels in uremic patients may overestimate glycemic status as they are influenced by a variety of factors including anemia, frequent blood transfusions, decreased lifespan of erythrocytes, and susceptibility of haemoglobin to glycosylation (1-3). While in the general population, several trials have demonstrated the long-term benefits of intensive glycemic control, the optimal HbA1c target in diabetic patients on haemodialysis is not established. Dialysis patients may be more prone to hypoglycemia due to various factors including malnutrition, decreased insulin clearance, and decreased renal gluconeogenesis (4-7). Thus, intensive glycemic control may be associated with increased mortality as shown by several studies (810). Most guidelines suggest a HbA1c target of 7%-8% (11,12).


Introduction
Glycated hemoglobin (hemoglobin A1c; HbA1c) levels in uremic patients may overestimate glycemic status as they are influenced by a variety of factors including anemia, frequent blood transfusions, decreased lifespan of erythrocytes, and susceptibility of haemoglobin to glycosylation (1)(2)(3).While in the general population, several trials have demonstrated the long-term benefits of intensive glycemic control, the optimal HbA1c target in diabetic patients on haemodialysis is not established.Dialysis patients may be more prone to hypoglycemia due to various factors including malnutrition, decreased insulin clearance, and decreased renal gluconeogenesis (4)(5)(6)(7).Thus, intensive glycemic control may be associated with increased mortality as shown by several studies (8)(9)(10).Most guidelines suggest a HbA1c target of 7%-8% (11,12).

Objectives
There is paucity of Indian data on the optimal HbA1c values to be targeted in dialysis patients, and hence this prospective study was undertaken to address this gap.

Study design
This prospective study was conducted in 61 prevalent maintenance hemodialysis (HD) patients with type 2 diabetes mellitus at the Madras Medical Mission Vijayan M et al Hospital, Chennai and Pondicherry Institute of Medical Science, Pondicherry between July 2015 and June 2016.All patients currently undergoing dialysis who were known diabetics were included in this study.We collected the following clinical details and laboratory measurements: demographic details, dietary patterns, vintage of haemodialysis, dialysis prescription, urine output, haemoglobin A1c (using high performance liquid chromatography, D10, Bio-Rad), blood hemoglobin levels, blood pressure, treatment history and echocardiographic details.Five patients had previously received percutaneous coronary intervention (4 M/1 F) and three patients (3 M/0 F) had received coronary artery bypass graft.We followed up these patients after one year.The number of HD per week varied from 2 to 4 sessions.Each HD session lasted for 4 hours.We looked at the association of HbA1c with 1-year survival and hemoglobin values.

Ethical issues
The research followed the tenets of the Declaration of Helsinki.This study was approved by ethics committee of Madras Medical Mission hospital.All participants were informed about the objectives of the study and assured that the information will remain confidential.Participants also signed out consent forms.

Statistical analysis
Statistical analysis was done using EZAnalyze software (http://www.ezanalyze.com,Timothy Poynton, USA).A P value of <0.05 was considered significant.Continuous variables were expressed as mean and standard deviation.Categorical variables were expressed as percentages.To look at the association between 2 continuous variables, Pearson's correlation coefficient method was used.To compare means of 2 continuous variables, t test was used.To look at the association between 2 categorical variables, chi-square test was used.

Results
The mean age of the patients was 60.21 ± 10.74 years.There were 18 female patients and 43 male patients.Eleven patients (18.03%) had evidence of systolic dysfunction on echocardiography.At the end of one year, 38 patients had survived, 10 patients died, 3 patients were transferred to continuous ambulatory peritoneal dialysis, and 10 patients were transferred to other centers.All these patients were receiving erythropoiesis stimulating agents (ESA) on a regular basis.
There was a significant relationship between HbA1c and blood hemoglobin values using Pearson's correlation coefficient (r = 0.245, P = 0.05).Figure 1 shows the correlation between the two values.
Table 1 shows the clinical variables of the patients involved in the study.Table 2 shows the comparison of various parameters between those patients who survived and died at the end of 1 year.Table 3 shows the survival rates of patients with Hba1c levels ≤6%, 6%-8% and ≥8%. Figure 2 shows the comparison between survived and non-survived patients.

Discussion
In patients with advanced CKD, targeting a HbA1c of <7% may be associated with increased risk of developing hypoglycemia (5-7).HbA1c level is largely affected by glycemic status, acidosis, the degree of anemia and the RBC life span (1,2).Similar to the diabetic non-CKD population, high levels of HbA1c are strongly     (10).A large prospective study of 24 875 patients reported that intensity of glycemia represented by a HbA1c of ≤5% and >11% were associated with poor outcomes (9).This is the first study from India, looking at HbA1c values and mortality in Indian dialysis patients.
In our study of HbA1c of our diabetic dialysis population, there seems to be a trend for those patients who survived had a lower HbA1c value (mean = 6.768 ± 1.602%) despite having longer dialysis duration (3.347 ± 3.032 years).However, owing to small sample size, statistical significance could not be obtained.This is probably because of a higher proportion (40%) of nonsurvived patients in the group having >8% HbA1c.In our patients, HbA1c levels were directly correlated with blood hemoglobin levels.This is probably related to the lifespan of the red blood cells, which is often reduced in uremic patients (2,3).None of the patients had hemolytic anemia on assessment.About 60% of the deaths were due to cardiovascular disease, and the others died at home, so the cause of death is not known.

Conclusion
Diabetic patients have poor survival rates on hemodialysis compared with non-diabetic patients (13,14).This study shows poor glycemic control could be one of the factors that could be responsible for the poor survival in diabetic patients on hemodialysis.Larger prospective studies need to be done to establish this relationship.

Limitation of the study
The main limitations of this study are the small sample size, lack of serial measurements of HbA1c, which has been planned as an ongoing study.

Authors' contribution
GA, AKD and VM designed the project.SM collected the data.SN, RP, and MV analysed the data.MV and GA wrote the manuscript.DG, MM, AKD and VM edited the final draft.All authors and read and signed the final manuscript.

Conflicts of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.The results presented in this paper have not been published previously in whole or part, except in abstract format.

Ethical considerations
Ethical issues (including plagiarism, data fabrication, double publication) have been completely observed by the authors.

Funding/Support
None.

Figure 2 .
Figure 2. The comparison of HbA1c in the patients who survived and expired at the end of one year after the measurement (P = 0.326).

Table 1 .
Clinical variables of these patients

Table 2 .
Comparison of parameters between survived and non-survived patients by independent samples t test

Table 3 .
Comparison of survival rates of groups based on HbA1c values using chi-square test