Is β-cell failure in type 2 diabetes mellitus reversible ?

β-Cell dysfunction and insulin resistance are known to be the two major mechanisms involved in the pathophysiology of type 2 diabetes mellitus (T2DM). The β-cell dysfunction is initially characterized by impairment in the fi rst phase of insulin secretion following glucose stimulation, resulting in impaired glucose tolerance (a prediabetic state) and postprandial hyperglycemia. [1-3] As the disease progresses, the second phase secretion declines, resulting in fasting hyperglycemia, i.e., either impaired fasting glucose (IFG) or T2DM.[4] This β-cell dysfunction is thought to be progressive and irreversible. However in the UKPDS, many subjects achieved the glycemic goal (HbA1c < 7.0%) at 9 years while being treated with oral monotherapy, denoting the lack of a progressive β-cell failure.[5,6] Moreover, β-cell function improved to 80% from the 50% seen at diagnosis following therapy with sulphonylureas.[7,8] Several other studies have also documented an improvement in insulin secretion after administration of sulfonylureas as well as diazoxide;[9-11] β-cell recovery is also noted with the weight loss that follows gastric bypass surgery in morbidly obese subjects with T2DM who have been using high-dose insulin for several years.[12-15] Finally, a recent study clearly demonstrated that β-cell failure in T2DM may be neither universal nor inevitable.[16] Therefore this study was conducted to determine the insulin secretion prior to and after withdrawal of exogenous insulin, while attaining desirable glycemic control (HbA1c ≤ 7.0%) with initiation of oral hypoglycemic drugs as well as weight loss.


Introduction
β-Cell dysfunction and insulin resistance are known to be the two major mechanisms involved in the pathophysiology of type 2 diabetes mellitus (T2DM).[3] As the disease progresses, the second phase secretion declines, resulting in fasting hyperglycemia, i.e., either impaired fasting glucose (IFG) or T2DM. [4]This β-cell dysfunction is thought to be progressive and irreversible.However in the UKPDS, many subjects achieved the glycemic goal (HbA 1c < 7.0%) at 9 years while being treated with oral monotherapy, denoting the lack of a progressive β-cell failure. [5,6]Moreover, β-cell function improved to 80% from the 50% seen at diagnosis following therapy with sulphonylureas. [7,8][14][15] Finally, a recent study clearly demonstrated that β-cell failure in T2DM may be neither universal nor inevitable. [16]Therefore this study was conducted to determine the insulin secretion prior to and after withdrawal of exogenous insulin, while attaining desirable glycemic control (HbA 1c ≤ 7.0%) with initiation of oral hypoglycemic drugs as well as weight loss.

Materials and Methods
Ten obese men with T2DM in the age range of 50-65 years and 10 healthy age-matched men participated in the study after signing the informed consent.The duration of DM was 10-15 years and all subjects were receiving insulin at doses of over 1.0 U/kg body weight, once or twice daily, for 2-10 years.Selection criteria included presence of morbid obesity, with BMI > 35 kg/m 2 ; duration of diabetes of over 10 years and treatment with insulin alone for over 2 years, with the daily dose being more than 1.0 U/kg body weight; and a HbA 1c of over 8.0% on two successive determinations at an interval of 3-4 months prior to enrollment.Subjects with serum creatinine ≥1.5 mg/dl, AST and ALT ≥ 2.5 times the upper limit of normal laboratory values, hospitalization for any cause during the previous 6 months, unstable coronary artery disease, cerebrovascular disease, or peripheral vascular disease were excluded.Subjects with known mental instability, as well as those who refused to provide informed consent, were also excluded.
The demographics are shown in Table 1.The subjects were hospitalized for 30 days in the metabolic ward of the medical center.They received a 1200-1500-kcal American Diabetes Association (ADA) diet and were also subjected to daily exercise for 30 min, twice a day, in the physical therapy department.They were started on glimepiride 8 mg and metformin 850 mg daily; exogenous insulin was withdrawn simultaneously.Metformin was titrated at weekly intervals to a maximum of 2550 mg/ day.In addition, dietary counseling was provided at weekly intervals by a registered dietician.The subjects were then followed at intervals of 4 weeks as outpatients, with recurrent counseling being given for compliance with diet, exercise, and oral agents.Serum C-peptide and glucose levels were determined after an overnight fast and up to 180 min during OGTT with 75 g glucose in subjects with T2DM prior to treatment (pre-treatment) and after attainment of HbA 1C < 7.0% with weight loss, metformin, and glimepiride (post-treatment); these values were also determined in the 10 healthy volunteers.Insulin and glucose responses during OGTT were assessed as cumulative responses as calculated by adding the differences between levels at each time period and fasting level.Cumulative response has been well documented to be a reliable expression of the integrated response as determined by the area under the curve over the duration of the OGTT. [17]Insulin secretion was expressed by fasting level of C-peptide (FCPEP) as well as an insulinogenic index as calculated by CRCPEP/CRG determined during OGTT. [18]Insulin sensitivity was expressed as a product of FG and FCPEP as recently documented. [19]Liver enzymes (i.e., ALT and AST), serum creatinine levels, and lipid profi les were also determined after an overnight fast, at the time of OGTT.

Results
At enrollment, serum creatinine and liver enzymes in subjects with T2DM were not signifi cantly different from that in healthy volunteers and remained unchanged at the end of the study [Table 2].Subjects with T2DM achieved a desirable glycemic control over a period of 6-9 months, as refl ected by HbA 1c ≤ 7.0% [Table 3].However, even with treatment the glycemia remained signifi cantly higher in these subjects than in the normal healthy subjects [Table 3].Insulin secretion improved  markedly on achieving the desirable glycemic goal, as evidenced by the signifi cant rises in FCPEP level as well as the insulinogenic index (CRCPEP/CRG) [Table 3].Moreover, both early (within 30 min) and late (90-120 min) insulin secretion following glucose ingestion increased significantly, as shown by the C-peptide responses [Figure 1].Finally, insulin sensitivity in the T2DM group also improved markedly following treatment, as assessed by the product of FG and FCPEP [Table 2].However, neither the insulin secretory patterns nor the insulin sensitivity normalized despite the subjects attaining the desirable glycemic goal [Figure 1; Table 3].Lipid profi les improved markedly in these subjects on achieving the desirable glycemic goal [Table 2].

Discussion
This study demonstrates that reinitiation of oral agents and weight loss can help subjects in attaining the desirable glycemic goal without the need for insulin administration.[22][23][24][25][26] Earlier studies have indicated that improvement in insulin sensitivity is the probable mechanism.However, the pattern of insulin secretion prior to and following reinitiation of oral agents and weight loss has not been well studied.This study clearly demonstrates the improvement in insulin sensitivity as well as enhancement of insulin secretion on achieving desirable glycemic control following withdrawal of insulin and reinitiation of oral agents and weight loss.[29][30][31] The markedly decreased β-cell function present at initial evaluation in our subjects may be attributed to suppression by long-term administration of exogenous insulin, especially in extremely high daily dose, as demonstrated in our recent study with various sufonylurea drugs. [32]Alternatively, the decline in β-cell function, especially in terms of postprandial insulin secretion, may also be attributed to extreme insulin resistance at the level of the β-cell itself in the presence of morbid obesity as has been described recently. [33,34]5][29][30][31] Even a moderate weight loss following a hypo-caloric diet is shown to improve insulin action and secretion. [28,30][30][31][35][36][37][38] Another similar study showed normalization of insulin sensitivity and restoration of a normal β-cell acute insulin response (AIR) to glucose ingestion, as well as a normal relationship of AIR to insulin sensitivity. [29]Therefore, the progressive β-cell failure documented in the UKPDS may also be attributed to increase in insulin resistance induced by the signifi cant weight gain noted in all therapeutic arms of the study. [5,6,39]Moreover, in this study many subjects were at the glycemic goal (HbA 1C < 7.0%) at 9 years, denoting lack of β-cell failure. [5,6]Thus, the progressive β-cell failure was found to be neither universal nor inevitable in this study, a fi nding that was confi rmed in another recent study. [16]Finally, we believe that the decline in β-cell function may be reversible, as documented in the UKPDS as well as in other studies following initiation of treatment with sulphonylureas at the onset of illness and even in the later stage of the disease, as noted in morbidly obese subjects who achieve weight loss with gastric bypass surgery alone or with hypo-caloric diet, exercise, and addition of oral agents as noted in this study. [5,6]Several physiologic mechanisms may explain this improvement in β-cell function.32] Decrease in clearance of C-peptide may be another possible mechanism but is unlikely since both renal and hepatic function remained intact at the time of the repeated OGTT.Alternatively, improvement in the recently recognized decline in the incretin effect in T2DM may have contributed to the reversal of β-cell failure as well. [40]Finally, it is plausible that β-cell failure in T2DM may be an expression of microvascular involvement of the β-cells themselves, with increasing fi brosis resulting in reduction in the number of β-cells as well as deranged function of the remaining cells; in the UKPDS, β-cell failure, as reflected by rising HbA 1c (above 7.0%) while on an oral agent, occurred at around the same time as the onset of microvascular complications. [5,16,39]The concept of microvascular disease involving the β-cells themselves received further support recently by the demonstration of progressively rising prevalence of β-cell failure in association with increasing number of microvascular complications and preserved β-cell function in patients without microvascular complications. [41,42]Microvascular disease is attributed to deposition of glycated proteins in organs and tissues.Deposition of amyloid, a glycoprotein, is well known to occur in T2DM. [43]Finally, fi brosis of the pancreatic islets in subjects with T2DM of long duration, which was documented in a recent study, may add credence to this hypothesis. [44]Therefore we believe that β-cell failure could be delayed or prevented by attaining and maintaining glycemic control, which is known to provide benefi cial effects with regard to the other well-known microvascular complications in both type 1 and type 2 DM. [5,6,45]

Table 2 : Comparison of pre-treatment and post-treatment hepatorenal function and lipids
a P < 0.05 vs normal; b P < 0.01 vs pre-treatment; c P < 0.05 vs pre-treatment

Table 3 : BMI, HbA 1c , fasting glucose (FG), fasting C-peptide (FCPEP), insulin sensitivity index (FG × FCPEP), and insulinogenic index during OGTT (CRCPEP/CRG) in 10 healthy men and in 10 morbidly obese subjects with type 2 DM prior to treatment (pre-treatment) and following attainment (post-treatment) of desirable glycemic goal (HbA 1C ≤ 7.0%) after weight loss and therapy with oral agents
a Cumulative response as determined by summation of differences between the level at each time period and basal level up to 180 min; b P < 0.01 vs pre-treatment; c P < 0.05 vs pre-treatment; d P < 0.05 vs normal; e P < 0.01 vs normal Jain et al.: Reversible β-cell failure in T2DM[Downloaded free from http://www.ijddc.