Efficacy of self-monitoring of blood glucose versus retrospective continuous glucose monitoring in improving glycaemic control in diabetic kidney disease patients.

AIMS: Patients with diabetic kidney disease (DKD) on anti-diabetic agents, are at greater risk of glycemic variations, both hypoglycemia and hyperglycemia. We aimed to compare glycemic control (using HbA1c) and hypoglycemia incidence in patients with Stage 3 DKD (eGFR 30-60 mL/min per 1.73 m2 ), receiving retrospective CGM-guided anti-diabetic therapy versus self-monitoring of blood glucose (SMBG) over 3 months. METHODS: Thirty patients with HbA1c >8% were randomized to 6-day retrospective CGM or SMBG. In the CGM group, CGM was worn at the beginning and 6 weeks. HbA1c, assessment of hypoglycaemia events (self-reported and BG < 4 mmol/L from CGM/SMBG data) and medication adjustment were performed at baseline and 3 months. All patients received education on hypoglycaemia avoidance. RESULTS: Fourteen patients were allocated to CGM and 16 to SMBG. Mean (±SD) eGFR was 42.9 ± 10.3 mL/min. Majority (86.7%) of patients had diabetes duration >10 years and on insulin therapy (90%). HbA1c improved significantly from baseline 9.9 ± 1.2 to 9.0 ± 1.5% (P < 0.001) at 3 months, with no difference between CGM (9.8 ± 1.2 to 8.8 ± 1.8%, P = 0.009) or SMBG (9.9 ± 1.3 to 9.1 ± 1.1%, P = 0.007) groups (P = 0.869 between groups). In the CGM group, percentage duration in hyperglycaemia (BG > 10 mmol/L) reduced from baseline 65.4 ± 22.4% to 54.6 ± 23.6% (P = 0.033) at 6 weeks, with a non-significant rise in percentage duration in hypoglycaemia from 1.2 ± 2.2% to 4.0 ± 7.0% (P = 0.176). There was no difference in self-reported and documented hypoglycaemia events. CONCLUSION: In a pilot study of DKD patients, short-term episodic use of CGM reduced time spent in hyperglycaemia range without significantly increasing time-exposure to hypoglycaemia. However, both CGM and SMBG were equally effective in improving glycaemic control.

Long-term outcomes of patients with type 2 diabetes attending a multidisciplinary diabetes kidney disease clinic.

BACKGROUND: The best model of care to retard diabetic kidney disease (DKD) in the clinic is underexplored. In this study we investigated the long-term renal outcomes of a joint endocrinologist-nephrologist clinic. METHODS: The present study was a nested case-control study derived from a cohort of patients with type 2 diabetes mellitus (T2DM) seen prospectively at a secondary care diabetes center (DC). Cases ("DKD clinic group") were patients seen at the CKD clinic after being referred by physicians in DCs for management of DKD. Controls ("non-DKD clinic group") were patients from the same DC (i.e. same source population) with the same inclusion criteria of Stages 3-4 chronic kidney disease (CKD) at baseline but not seen at the DKD clinic. The outcome was Stage 5 CKD, defined as an estimated glomerular filtration rate <15 mL/min per 1.73 m2 . RESULTS: During the median follow-up period of 3.0 years (interquartile range 1.2-5.1 years), 240 patients (28.7%) reached Stage 5 CKD, with 45.8% and 54.2% of those reaching Stage 5 CKD in the DKD and non-DKD clinic groups, respectively. Multivariable Cox regression revealed that the DKD clinic group had a lower risk of progressing to Stage 5 CKD (hazard ratio 0.55; 95% confidence interval 0.36-0.83; P = 0.004) compared with the non-DKD clinic group. CONCLUSIONS: Multidisciplinary endocrinology and nephrology care in the DKD clinic is associated with a lower risk of end-stage renal disease. These findings may inform future management strategies targeted at patients with T2DM and CKD, especially with regard to joint specialist management involving endocrinologists and nephrologists.

Effect of long-term glycemic variability on estimated glomerular filtration rate decline among patients with type 2 diabetes mellitus: Insights from the Diabetic Nephropathy Cohort in Singapore.

BACKGROUND: In the present study, we examined the association between HbA1c variability and renal disease progression based on estimated glomerular filtration rate (eGFR) decline in patients with type 2 diabetes mellitus (T2DM) in Singapore. METHODS: Glycemic burden and renal function were retrospectively assessed in 1628 patients in 2002-2014. Multivariable logistic regression was used to assess the relationships between HbA1c variability (expressed as HbA1c coefficient of variation [HbA1c-CV] in quartiles), HbA1c intrapersonal mean (HbA1c-IM), and eGFR decline, adjusted for baseline covariates. RESULTS: Among patients with relatively good glycemic control (i.e. HbA1c-IM below the median cohort value [8.0%]), HbA1c-CV Quartile 4 was associated with eGFR decline (odds ratio [OR] 1.88; 95% confidence interval [CI] 1.10-3.25). The OR for HbA1c-CV Quartile 4 was 2.20 (95% CI 1.24-3.89) after additional adjustment for HbA1c-IM. Where HbA1c-IM was above the median cohort value, HbA1c-CV Quartiles 3 and 4 were associated with eGFR decline, with ORs of 2.60 (95% CI 1.48-4.55) and 3.29 (95% CI 1.89-5.76) respectively. After further adjusting for HbA1c-IM, the ORs for Quartiles 3 and 4 were 2.69 (95% CI 1.53-4.74) and 3.51 (95% CI 1.98-6.21), respectively. CONCLUSIONS: Variability in HbA1c is strongly and independently associated with eGFR decline in patients with T2DM independent of mean HbA1c. The findings may highlight the importance of sustained stable glycemic control in management of diabetes mellitus.

Genetic variants in the receptor for advanced glycation end products (RAGE) gene were associated with circulating soluble RAGE level but not with renal function among Asians with type 2 diabetes: a genome-wide association study.

BACKGROUND: The soluble receptor for advanced glycation end products (sRAGE) has been shown to play an important role in diabetic complications. We conducted genome-wide association study (GWAS) of sRAGE in Asian type 2 diabetes mellitus (T2DM) patient and validated the association in an independent cohort of T2DM. METHODS: GWAS for sRAGE was performed in 2058 T2DM patients. Associations between single-nucleotide polymorphisms (SNPs) and plasma sRAGE level were analyzed in an additive model using a linear mixed model. To validate the associations, we performed de novo genotyping in an independent cohort (n = 1984). We selected the top SNP for assessment with diabetic kidney disease (DKD). RESULTS: The strongest SNP, rs2070600C>T (P = 1.21 × 10-52), was a genotyped, missense SNP located on chromosome 6, corresponding to the RAGE (AGER) gene locus, the gene encoding RAGE. Conditioning analysis on rs2070600 revealed that rs2071288C>T was the top genotyped independent SNP (P = 8.36 × 10-10). Both SNPs were strongly and dose-dependently correlated with sRAGE level (TT = 399.6 pg/mL, CT = 737.0 pg/mL and CC = 967.0 pg/mL, P < 0.001 for rs2070600; TT = 687.9 pg/mL, CT = 737.6 pg/mL and CC = 904.7 pg/mL, P < 0.001 for rs2072188). Both SNPs were robustly replicated in the independent cohort, especially among Chinese patients (P = 9.02 × 10-72 for rs2070600; P = 1.13 × 10-9 for rs2071288). Log-transformed sRAGE was associated with DKD after adjustment for age, gender and ethnicity in pooled cohorts [odds ratio 2.536 (95% confidence interval 1.864-3.450), P < 0.001]. However, we did not observe any significant association between rs2070600 and DKD. CONCLUSIONS: Common variants in RAGE are strongly associated with plasma sRAGE level, which is associated with DKD. However, we did not find a causal link between sRAGE and renal function by Mendelian randomization.

Onset and progression of kidney disease in type 2 diabetes among multi-ethnic Asian population.

AIM: To elucidate the natural history of chronic kidney disease(CKD), which is defined as estimated glomerular filtration rate(eGFR)<60ml/min/1.73m(2) and/or increase of urinary albumin-to-creatinine ratio (uACR)≥30mg/g), and to identify factors associated with its onset and progression. METHODS: Prospective cohort study on individuals with T2DM attending Diabetes Centre in a regional hospital in Singapore from 2002. There were 553 patients with no pre-existing CKD for "onset" analysis and 967 patients with pre-existing CKD for "progression" analysis. Multivariable logistic regression was performed to determine risk factors of the outcomes. RESULTS: The mean follow-up period was 5.8years (4.5-7.1) and 5.3years (3.9-6.9) for the onset and progression cohorts respectively. About 45% of individuals developed CKD and 41% had progression. Among subjects with CKD onset, albuminuria-only occurred in 75% of them. Majority of the patients remained in the same CKD risk-category during follow-up. Progression and regression occurred across all CKD-categories. Transitions to adjacent risk-category were much more likely than transitions bypassing adjacent state. Risk factors for CKD onset included baseline albuminuria, eGFR, HbA1c variability, body mass index, triglycerides and age (all P<0.05). The predictors for CKD progression or rapid-progression included HbA1c variability, baseline albuminuria, systolic blood pressure, LDL-cholesterol, eGFR, HbA1c and ethnicity (all P<0.05). CONCLUSIONS: Albuminuria was the first manifestation of CKD in most T2DM patients. Transition across CKD-category occurred bi-directionally, but evolved largely in a stepwise fashion. The onset and progression of CKD were predicted by multiple risk factors, some of which were modifiable.

Long-term diabetes outcomes in multi-ethnic Asians living in Singapore.

AIMS: This study aims to assess ethnic and gender disparities on long-term complications among multi-ethnic Asians with Diabetes Mellitus (DM) living in Singapore. METHODS: We conducted a retrospective cohort study involving 3006 patients who attended a diabetes centre in a hospital from 2003 to 2011. Demographics and clinical data were obtained from standardised questionnaire and patient's case records. Age at onset of diabetes was calculated as current age minus duration of DM in years. Outcomes on Acute Myocardial Infarction (AMI), End-Stage Renal Failure (ESRF) and all-cause death were ascertained by data linkage with national registries. RESULTS: The mean duration of diabetes exposure was 15.6 ± 9.1 years for AMI, 15.4 ± 9.0 years for ESRF and 17.0 ± 9.0 years for death. After adjusting for traditional cardiovascular risk factors, Malay and Indian with diabetes remained significantly associated with AMI with HRs 2.81(95%CI, 1.81-4.37) and 2.03 (95%CI, 1.15-3.59), respectively. The effect of Malays on ESRF and death became attenuated post-adjustment. Besides mortality, there was preponderance for other adverse outcomes associated with male. CONCLUSIONS: Ethnic (Malay worse) and gender (male worse) disparities were observed in DM-related outcomes. The results may inform allocation of finite resources and to organize care targeted at high-risk groups.

Fasting during Ramadan and Associated Changes in Glycaemia, Caloric Intake and Body Composition with Gender Differences in Singapore.

INTRODUCTION: Millions of Muslim patients with diabetes mellitus (DM) fast during Ramadan. However, little is known about the metabolic impact of Ramadan fasting. We aimed to study the changes in body composition and metabolic profile in this group of patients. MATERIALS AND METHODS: We studied 29 Southeast Asian Muslim patients with type 2 diabetes; all underwent pre-Ramadan education. Study variables were weight change, body composition (using multifrequency bioimpedance method, InBody S20®, Biospace, South Korea), blood pressure (BP), glycated haemoglobin (HbA1c), fasting lipid profile, and caloric intake assessment using FoodWorks® nutrient analysis software. RESULTS: Twenty-three subjects fasted ≥15 days; mean ± SD: 57 ± 11 years; 52% were males. HbA1c improved significantly (8.6 ± 2.4% pre-Ramadan vs 8.0 ± 2.3% end-Ramadan, P = 0.017). Despite similar body weight, there was reduction in body fat mass (BFM) (30.9 ± 11 kg vs 29.2 ± 12.2 kg, P = 0.013). Multivariate analysis suggested that the reduction in HbA1c was attributed by reduction in BFM (ß = -0.196, P = 0.034). There was no change in visceral adiposity (visceral fat area (VFA)) but stratification by gender showed a reduction amongst females (137.6 ± 24.5 cm2 to 132.5 ± 25.7 cm2, P = 0.017). These changes occurred despite similar total caloric intake (1473.9 ± 565.4 kcal vs 1473.1 ± 460.4 kcal, P = 0.995), and proportion of carbohydrate (55.4 ± 6.3% vs 53.3 ± 7.5%, P = 0.25) and protein intake (17.6 ± 4.1% vs 17.3 ± 5.4%, P = 0.792), before and during Ramadan respectively, but with increased proportion of fat intake (11.9 ± 2.4% vs 13 ± 11.7%, P = 0.04). Seven out of 23 patients had medications adjusted to avert symptomatic hypoglycaemia but none of the patients developed severe hypoglycaemia. CONCLUSION: Ramadan fasting can be practiced safely with prior patient education and medication adjustment. It also confers modest benefits on metabolic profile and body composition, especially among females.

Diabetes management and hyperglycemia in safety sensitive jobs.

The chronic and acute effects of hyperglycemia affecting cognition and work are as important as those of hypoglycemia. Its impact, considering that majority of diabetic patients fail to reach therapeutic targets, would be potentially significant. Self monitoring of blood glucose, recognition of body cues and management interventions should be geared not only towards avoidance of disabling hypoglycemia, but also towards unwanted hyperglycemia. Over the long term, chronic hyperglycemia is a risk for cognitive decline. Acute episodes of hyperglycemia, above 15 mmol/L have also been shown to affect cognitive motor tasks. Maintaining blood sugar to avoid hyperglycemia in diabetic workers will help promote safety at work.