Diabetes Remission, Can we Keep it there? A New Approach to Treating T2DM

By Haysook Choi, BSc.N, RN, CDE posted in Latest Diabetes News Professionals & Educators

img_grandadboyWe, at Mt. Sinai Hospital in Toronto, are currently conducting a clinical trial aimed at preservation of pancreatic β-cell function and achievement of remission of diabetes through the short-term use of intensive insulin therapy. Through this study, participants will gain greater knowledge about diabetes and the impacts of lifestyle management on their glucose level. The participants will be supported throughout insulin therapy to gain confidence in their ability to work with this treatment, thus counteracting psychological insulin resistance and resulting in more effective healthcare.

Short-term Intensive Insulin Therapy to Preserve Pancreatic β-cell Function


In Canada, more than 9 million or nearly 1 in 4 live with diabetes or prediabetes. 80% of Canadians with diabetes will die from cardiovascular disease while others are at high risk for complications and premature death. Not only is diabetes a personal challenge for people with the disease, it is also a tremendous financial burden for society as a whole. Diabetes currently costs our healthcare system and our economy $11.7 billion and will cost Canadians about $16 billion annually by 2020 (1).

Type 2 diabetes (T2DM) is a chronic complex disease that poses many challenges for the healthcare provider and patient. The UKPDS study (2) has shown that β-cell function deteriorates progressively over time in people with T2DM, irrespective of lifestyle and pharmacological interventions. This study also showed that only a fraction of pancreatic islet function remained at the time of diagnosis. Currently, T2DM is usually managed by a stepwise introduction of lifestyle interventions, oral agents alone and in combination, and finally insulin. Unfortunately, existing management protocols have failed to achieve and maintain the glycemic levels necessary to provide optimal healthcare for people with diabetes (3, 4, 5). Furthermore, due to many factors, oral antihyperglycemic agents were preferred instead of insulin injections even when insulin therapy was clearly the optimal treatment option. Our current culture and healthcare set-up thus produces “psychological insulin resistance” as patients and health care providers fear insulin initiation (6, 7).

Data from the UKPDS (2) trials has indicated the importance of tight metabolic control in delaying or preventing the progression of T2DM complications. Furthermore, the long term observational studies EDIC (DCCT-10 year follow-up) and UKPDS-10 year follow-up (8), have shown that a “metabolic memory” exists where the benefits of initial tight metabolic control are carried forward for another 10 years even though the control of diabetes was not so intense during the observational period. In ACCORD (9) and ADVANCE (10) trial, the participant’s glucose control at baseline of the study was poor and likely not at optimal levels for some time before entry in the trial. This might explain the lack of beneficial effects in the intensive arm of both trials. Clear evidence exists of the importance of tight glucose control and a good “metabolic memory” in the prevention and delayed progression of chronic diabetes cardiovascular complications.

Natural History of Type 2 Diabetes

The T2DM is a progressive disease characterized by insulin resistance (IR), progressive reduction in β-cell mass and dysfunction resulting in worsening hyperglycemia over time. The most common cause of IR is obesity since there is decreased ability of muscle and fat tissues to take up and metabolize glucose in response to insulin. Not all patients with IR will develop diabetes, however, as long as their β-cells can compensate by producing extra insulin to meet their metabolic needs and, since IR is relatively stable over the course of T2DM, it is unlikely to account for the continued long-term progressive nature of T2DM. Instead, the development of T2DM is caused by the inability of β-cells to adequately compensate for IR, which may be a consequence of β-cell dysfunction and β-cell loss (11).

Growing evidence suggests that deterioration in β-cell function occurs before IGT and hyperglycemia become apparent and may be the primary defect in many people with diabetes (11). For example, Retnakaran et al. (12) looked at women in pregnancy as their pancreatic β-cells experience extreme added stress to compensate for pregnancy induced IR, rising in glucose level, in order to maintain euglycemia. In his study, women with gestational diabetes and those with gestational impaired glucose tolerance exhibited declining β-cell function in the first year postpartum despite stable rates of dysglycemia 3-12 months postpartum. Thus, β-cell dysfunction progresses in the early stages of postpartum in women with a history of gestational dysglycemia and is likely a pathophysiologic factor contributing to the development of T2DM. This study provided the unique opportunity to identify a patient population where the early evolution of β-cell dysfunction unfolds prior to the development of diabetes or pre-diabetes.

The evolution of β-cell dysfunction in T2DM progression has been proposed by Weir (13) in 5 stages. Stage 1 is β-cell compensation: insulin secretion increases to maintain normoglycemia to compensate for IR resulting from obesity, physical inactivity and genetic predisposition. In this stage, β-cell mass may be normal or increased and insulin secretion is maintained. Stage 2 is stable β-cell adaptation: glucose levels start to rise due to loss of β-cell mass and initial disruption of β-cell function. Acute glucose-stimulated insulin secretion was reduced if the fasting plasma glucose levels were >5.6 mmol/l and completely lost when fasting glucose level increased to only 6.4 mmol/l. However, the 2nd phase of insulin release was partially preserved. The loss of the first phase and a decrease in the second phase of prandial insulin secretion is one of the earliest signs of defective β-cell function as impaired glucose tolerance (IGT). In the Diabetes Prevention Program, an individual at the upper end of stage 2 with IGT progressed to diabetes at the rate of ~11%/year while those who adhered to a diet and exercise programme progressed at a rate of only 5%/year. Stage 3 is transient unstable early decompensation:  glucose levels rise rapidly to the diabetic levels of stage 4. The increase in glucose concentration likely worsens glucotoxic effects on β-cells leading to less efficient insulin secretion. Stage 4 is stable decompensation: there is β-cell mass reduction and more severe β-cell dysfunction and rapid increase in glucose levels. T2DM is noted here. Stage 5 is severe decompensation: there is a profound reduction in β-cell mass and β-cell failure with progression to ketosis and glucose levels are typically >22 mmol/l. Type 1 diabetes, ketosis, or very severe pancreatitis is noted here. Movement across stages 1-4 can be in either direction. For example, people with T2DM can move from stage 4 to stage 1 or 2, for example, individuals with T2DM who undergo gastric reduction surgery. Even treatment with diet, exercise, and oral agents can transition patients to stage 2. In order to stop the progression of T2DM from one stage to next it is necessary to minimize β-cell loss and preserve remaining β-cell function.

There are several genetic and environmental factors contributing to the progressive loss of β-cell function in T2DM. In Tibaldi’s (11) review, the acquired factors such as, glucotoxicity from chronic exposure to hyperglycemia can prompt β-cell apoptosis from glucose-induced toxicity causing continuous decline of β-cell function in T2DM. During glucotoxicity, the first-phase insulin secretion is diminished at blood glucose of ≥ 5.6 mmol/l and completely lost at glucose of 6.4 mmol/l. Yet, the state of glucotoxicity is potentially reversible (14). Another factor is lipotoxicity due to increased circulating free fatty acid and dyslipidemia that are commonly seen in patients with diabetes, especially obese individuals with abdominal adiposity. In healthy individuals, elevated free fatty acids may also increase IR and typically prompt enhanced insulin secretion. However, persistently elevated free fatty acids and chronic hyperglycemia may contribute to progressive β-cell failure (β-cell lipotoxicity) in patients with diabetes or those predisposed to developing diabetes. Oxidative stress and pancreatic inflammation caused by hyperglycemia is likely involved in β-cell dysfunction as well. Medications such as some sulfonylureas and glucocorticoids may also promote the progression of β-cell dysfunction or even possibly β-cell death.

Both Retnakaran (14) and Tibaldi (11) noted that first treating hyperglycemia with insulin may alleviate glucotoxicity and lipotoxicity, which are known to adversely affect β-cell function and that insulin may exert antiapoptotic effects. Therefore timing may be an important factor when initiating insulin to improve β-cell function: early initiation of intensive insulin therapy appears to delay the progression of β-cell dysfunction. Thus, there may be a window of opportunity for treatment where intensive insulin therapy may slow or prevent further progression of T2DM.

Effect of Short-Term Intensive Insulin Therapy on Beta-Cell Function

Studies show that when short-term intensive insulin therapy (IIT) was implemented early in the course of T2DM, glycemic remission is sometimes induced wherein patients are able to maintain normoglycemia without any anti-diabetic agents. Weng et al. (15) looked at 382 newly diagnosed T2DM patients randomized to the continuous subcutaneous insulin (CSII) group, MDI group, and the OAD group. The Primary end point was the time of glycemic remission and remission rate at 1 year after short-term IIT. There were no serious hypoglycemia episodes. More participants achieved target glycemic control in the insulin group (97% in CSII and 95% in MDI) in less time (4.0 days in CSII group and 5.6 days in MDI group) than those treated to OAD (83.5% and 9.3 days). Remission rates after 1 year were significantly higher in the insulin groups (51.1% in CSII and 44.9% in MDI) than in OAD group (26.7%). Overall, β-cell function improved significantly after intensive interventions in all 3 groups though the secretory quality of β–cells was markedly improved in the insulin groups (mean 8.7% in CSII and 10.8% in MDI) compared to OAD group (4.1%), suggesting diminished β–cell overstimulation with insulin treatment. Additionally, the decline in IR and improved lipid profile without use of lipid-adjusting agents were also indicators of the reduction in glucotoxicity. All treatment groups achieved high rates of initial euglycemia but the insulin therapy group was associated with significantly higher rates of remission and preservation of first-phase insulin secretion after 1 year. This suggests that more profound β–cell rest by IIT is likely to go beyond glucose-lowering effects and could possibly have extended benefits such as anti-inflammatory and anti-apoptosis effects. Thus, early intensive glycemic control by IIT as compared to OAD, could provide a more beneficial type of β–cell rest and reduce excessive secretory demands on damaged β–cells by affecting the metabolic memory, impeding the progression from metabolic abnormalities to irreversible cellular alterations. These effects might further alter the natural history of diabetes and prevent or reduce the development and progression of diabetes-related complications.

A pilot study at Mt. Sinai Hospital with 34 participants with a mean duration of T2DM of 5.9 years underwent 4-8 weeks of IIT, with a 4-h meal test administered at baseline and at 1 day post-IIT. A positive clinical response was defined as fasting glucose <7.0 mmol/l off any antidiabetic therapy at the latter test. A positive response was achieved in 68% (n=23) of the subjects. From this study, it was concluded that the clinical response to short-term IIT is variable, consistent with the heterogeneity of T2DM. The responders who achieved positive response to the short-term IIT had, at baseline, higher c-peptide and lower glucose levels during the latter stages (180-240 minutes) of the meal test than the non-responders. This reflects late-phase insulin secretion and was the strongest predictors of the improvement in β–cell function following short-term IIT (16).

Studies have shown that, the degree of improvement in response to short-term IIT varies between patients. Thus, Kramer et al. (17) at Mt. Sinai Hospital sought to characterize the determinants of improvements in β–cell function in response to short-term IIT in early T2DM. Sixty-three patients with mean 3.0 years duration of T2DM underwent 4 weeks of IIT with a 75gm oral glucose tolerance test administered at baseline and 1-day post-IIT. No severe hypoglycemic episodes were reported. Overall, the study population experienced an increase in β–cell function, with a third of participants improving β–cell function by at least ≥25%. These participants also had greater changes in reduction of fasting glucose, A1C, ALT, AST and IR. At baseline, the third of participants with greatest improvement in β–cell function had higher fasting glucose, higher A1C, and lower β–cell function. Thus, their poor glycemic control at baseline may be reflective of more pronounced effects of glucotoxicity resulting in the greatest improvements in β –cell function from IIT. Upon further analyses, the reversibility of β–cell dysfunction was achieved in only those participants where IIT yielded an improvement in IR. Hence, decline in IR may be a key determinant of improvement of β–cell function in response to short-term IIT, suggesting a fundamental contribution of IR to the reversible component of β–cell dysfunction in early T2DM. This will help to identify which patients are most likely to respond to IIT and the pathophysiology underlying this effect.

Two sets of reviews looking at implementation of short-term IIT early in the course of T2DM by either CSII or MDI were performed at Mt. Sinai Hospital, Toronto.   The goal of this treatment strategy was achievement of “glycemic remission” wherein patients could maintain normal glucose levels without any antihyperglycemic medication after cessation of the short course of IIT. In Retnakaran et al.’s review (14), the vast majority of these newly diagnosed patients were able to achieve this “glycemic remission” (80-97%). Furthermore, euglycemia persisted for 1 year in about 40% of the subjects from these studies and has continued for up to 2 years or longer in some cases. There were no severe hypoglycemia episodes. Overall, these clinical studies have demonstrated that short-term IIT can have long-lasting effects on glycemic control in patients with newly diagnosed T2DM. In Kramer et al.’s meta-analyses (18), in patients with newly diagnosed T2DM, short-term IIT is associated with improvement in β-cell function and IR. The percentage of participants in long-term drug-free remissions was 66.2% at 3 months of follow-up, 58.9% at 6 months, 46.3% at 12 months and 42.1% at 24 months after IIT. Baseline characteristics associated with glycemic remission were increased BMI and decreased fasting plasma glucose, as well as low post-intensive insulin therapy fasting plasma glucose and post-challenge glucose. The baseline factors indentified as predictors of remission are possibly indicative of increased underlying residual β–cell function (eg, after elimination of the glucotoxic effects of hyperglycemia). The high BMI could be a marker for increased β–cell mass available for recovery as results from post-mortem studies show that obesity increases β–cell mass in both diabetic and non-diabetic individuals.


Predictors of the response to short-term intensive insulin therapy

Clinical studies have consistently shown that short-term IIT improves β-cell function and that euglycemia can be maintained for a long time after IIT is completed. However, these effects are not permanent and glucose levels eventually rise again. Additionally, there is heterogeneity in the patient response as seen by varying durations of the euglycemia period post IIT. Retnakaran et al (14) propose that predictors of a positive response to short-term IIT can be divided into 3 groups based on their timing.  1. Baseline predictors: better glycemic control, higher late-phase insulin secretion and shorter duration of diabetes may all reflect greater residual β-cell function. Additionally, higher BMI and IR may indicate a greater contribution of secretory stress (rather than β-cell failure alone) leading to the development of T2DM in the setting of greater residual β-cell function relative to those with non-obese T2DM. 2. Factors that emerge during treatment: the faster achievement of glycemic targets and lower requirements for exogenous insulin. This point to the recovery of greater underlying β-cell function. 3. Those that are apparent immediately after treatment: greater improvement in β-cell function from that which was observed at baseline has been a consistent predictor of the response to short-term insulin therapy in previous studies.

Support for the importance of residual underlying β-cell function observed from the studies of short-term IIT are first, patients who respond positively (achieve sustained euglycemia) have required less exogenous insulin during intensive therapy than their peers, suggestive of a comparatively greater contribution of endogenous insulin secretion. Second, when receiving IIT early in the course of T2DM, very little hypoglycemia is experienced despite targeting near-normal glycemic control. This is very different from the usual experience in clinical practice setting where there is an increased risk of hypoglycemia as one approaches normoglycemia when administering insulin therapy late in the course of disease. This is likely because only β-cells can finely regulate insulin secretion to achieve normal glycemic control without hypoglycemia (14).

The New Approach to Treating Type 2 Diabetes in Clinical Research

To avoid the increasing β-cell failure, the complication associated with hyperglycemia and the burden of T2DM, it is desirable to maintain β-cell function for as long as possible after the timely diagnosis of diabetes. Studies have shown that early IIT in early T2DM can increase insulin sensitivity at both skeletal muscle and liver where its effects include suppression of hepatic gluconeogenesis and reduction in liver fat content, improved β-cell function and reduction in IR. Thus, insulin therapy needs to be introduced at an early stage in the natural history of T2DM, when sufficient β-cell mass remains to enable functional improvement with the alleviation of reversible glucotoxicity and lipotoxicity.

Studies have shown that IIT in early T2DM is safe with very little risk of hypoglycemia.

Contrary to popular perception of both patients and providers, a short-course of IIT can result in significant improvement in QOL and treatment satisfaction, demonstrating the patient acceptability of early insulin therapy. Opsteen et al. at Mt. Sinai Hospital reported significant improvements in physical functioning, general health, general mental health, global health perception, and diabetes worry and treatment satisfaction (19). However, studies have also shown that euglycemia achieved after short-term IIT in early T2DM is temporary and glucose levels eventually rise. We also know that the “metabolic memory” exists, thus another reason for regaining glycemic control early on before much of β-cell mass and function is lost.

“Diabetes Remission”, Can we Keep it there?

A New Approach to Treating T2DM.

Participant Recruitment


We are now in the process of accepting participants to join our study and help determine if it is possible to preserve and maintain pancreatic β-cell dysfunction for longer duration, and change the natural course of T2DM. The inclusion criteria for this study are: T2DM for 0-5 year, on metformin only or lifestyle to treat diabetes.

Contact: hchoi@mtsinai.on.ca


  1. CDA. The prevalence and costs of diabetes. www.diabetes.ca/diabetes-and-you/what/prevalence/ 2012
  2. UK Prospective Diabetes Study (UKPD) Group. Intensive blood-glucose control with sulfonylurea or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 1998; 352:837-54
  3. Leahy JL, Hirsch IB, et al. Targeting β-cell function early in the course of therapy for type 2 diabetes mellitus. Journal of clinical endo & metab. 2010; 95 (9):4206-16.
  4. Harris SB, Worrall G, et al. Diabetes Management in Canada:  Baseline results of the Group Practice Diabetes Management Study. CJD 200; 30 (2): 131-137.
  5. Harris SB, Ekoe J, et al. Glycemic control and morbidity in the Canadian primary care setting (results of the diabetes in Canada evaluation study). Diabetes Research and clinical practice 2005; 70: 90-97  
  6. Polonsky W and Jackson, R. What’s so tough about taking insulin?  Addressing the problem of psychological insulin resistance in type 2 diabetes. Clinical Diabetes 2004; 22: 147-150
  7. Peragallo-Dittko V. Removing Barriers to insulin therapy. The Diabetes educator 2007; 33 (S3): 60-65.
  8. Holman RR, Paul SK, et al. 10-year follow-up of intensive glucose control in type 2 diabetes. NEJM 2008; 359:1577-89
  9. Gerstein HC, Miller ME, Byington RP, et al. Effects of intensive glucose lowering in type 2 diabetes. NEJM 2008; 358: 2545-59
  10. Patel A, MacMahon S, Chalmers J, et al. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. NEJM 2008; 358: 2560-72
  11. Tibaldi J. Preserving insulin secretion in type 2 diabetes mellitus. Expert Rev. Endocrinol Metab. 2008; 3 (2): 147-159
  12. Retnakaran R, Ying Q, et al. Beta-cell function declines within the first year postpartum in women with recent glucose intolerance in pregnancy. Diabetes Care 2010 August; 33 (8): 1798-1804
  13. Weir GC, Bonner-Weir S. Five stages of evolving beta-cell dysfunction during progression of diabetes. Diabetes 2004: 53 (3): S16-S21
  14. Retnakaran R, Zinman B. Short-term intensified insulin treatment in type 2 diabetes: long-term effects on ß-cell function. Diabetes, obesity and metabolism 2012; 14(3):161-66
  15. Weng J, Li Y et al. Effect of intensive insulin therapy on β–cell function and glycaemic control in patients with newly diagnosed type 2 diabetes: a multicentre randomized parallel-group trial. Lancet 2008;371:1753-60
  16. Retnakaran R, Yakybovich N, et al. The response to short-term intensive insulin therapy in type 2 diabetes. Diabetes, obesity and metabolism 2010; 12: 65-71
  17. Kramer KC, Choi H, et al. Determinants of reversibility of beta-cell dysfunction in response to short-term intensive insulin therapy in patients with early type 2 diabetes. American journal of physiology-endocrinology and metabolism. 2013; published online vol no. D01: 10.1152/ajpendo.00447.2013
  18. Kramer CK, Zinman B, Retnakaran R. Short-term intensive insulin therapy in type 2 diabetes mellitus: a systemic review and meta-analysis. www.thelancet.com/diabetes-endocrinology. Published online January 30, 2013.
  19. Opsteen C, Qi Y, et al. Effect of short-term intensive insulin therapy on quality of life in type 2 diabetes. Journal of evaluation in clinical practice 2012; 18:256-261.


About the Author

Haysook Choi, BSc.N, RN, CDE is a registered nurse with nineteen years’ experience as a Diabetes Educator and has dedicated the last nine years to clinical diabetes research. She holds BSc.N/ Primary Health Care Nurse Practitioner degree with Honors from Ryerson University, Toronto and is a Certified Diabetes Educator. As a diabetes educator at a community hospital, she was involved in the development and implementation of the diabetes education programmes. She was instrumental in assessing, planning and implementing the gestational diabetes service in the ambulatory care setting. Additionally, Haysook helped to bring changes to the in-hospital nursing and medical practice by developing and implementing The Hypoglycemia Protocol and The DKA Protocol. In 2004 she changed her focus from clinical practice to research in endocrinology and presently works as a research nurse coordinator. Some of the trials that she has worked in the past include the Action to Control Cardiovascular risk in Diabetes (ACCORD trial), ultra-long-acting basal insulin, GLP-1 analogues and looking at insulin-prescribing behavior of Family Physicians across Canada. For the past 4 years Haysook has been working with Dr. B. Zinman and Dr. R. Retnakaran at Mt. Sinai Hospital, Toronto on a study that aims to preserve pancreatic β-cell function and achieve remission of diabetes through the short-term use of intensive insulin therapy.