Caloric Restriction and Insulin Secretion

In obese patients with type 2 diabetes mellitus, lifestyle modifications resulting in weight loss improve or even normalize blood glucose. This beneficial effect on glucose control is accounted for by improvements in both insulin secretion and insulin sensitivity. However, the metabolic effects of caloric restriction per se may be, at least in part, independent of body weight reduction. Furthermore, improved control of blood glucose in type 2 diabetes by very low calorie diet (VLCD) for 40 days was documented during the first 10 days of caloric restriction, when weight loss was still trivial. When caloric intake was increased after weight reduction, plasma glucose increased in spite of no weight rebound. The mechanism(s) underlying these early improvements caused by a VLCD in patients with type 2 diabetes have been assessed only in a few studies. Thus, a fall in hepatic glucose production and a modest increase in insulin sensitivity were reported as early as 7 days after a very low calorie diet. A subsequent study replicated the effects of short term VLCD on hepatic glucose production, but not on whole body insulin sensitivity.

As to beta cell function, earlier studies reported an apparent improvement in insulin secretion rate during the oral glucose tolerance test (OGTT), after short-term application of a VLCD. However, no formal investigation of beta cell sensitivity to glucose was performed, and, since glucose was given per os, other factors (e.g. incretins, ghrelin) might have been involved. One study reported an improvement of beta cell response during hyperglycemic clamps, i.e. excluding gut related factors, after 8 weeks of a VLCD and during a weight stabilization period, i.e. in the absence of the negative energy balance signal. Furthermore, by study design, it did not explore the first phase secretory response to glucose.

Quite recently, Lim and coll. reported that, in patients with type 2 diabetes, a VLCD markedly improved glucose control in a few days, and that ameliorations of both liver insulin sensitivity and beta cell sensitivity to glucose were the mechanisms primarily involved. Prolongation of the VLCD for 8 weeks led to an apparent remission of diabetes. This and most of earlier mechanistic studies were performed in patients with BMI around 30-35, in whom, according to most current guidelines, metabolic surgery should not be considered a treatment option. In patients with type 2 diabetes undergoing bariatric surgery, improvements in glucose control and in beta cell function are detectable before a significant weight loss occurred, and there is a strong suggestion that intestinal bypass procedures may have metabolic effects (e.g. on beta cell) that are independent of their effect on body weight, possibly involving the incretin axis. Furthermore, remission rates of type 2 diabetes following bariatric surgery as high as 70-80% have been reported.

Thus, it would be important to know whether in severely obese patients with type 2 diabetes, who are potential candidates for metabolic surgery, a short term VLCD exerts effects similar to those reported in less obese patients.

The investigators therefore performed a preliminary proof-of-concept study to assess whether in severely obese patients with type 2 diabetes 7 days of VLCD affect glucose control through changes in either beta cell function or insulin sensitivity or both.

Participants were studied at baseline and then after 7 days of caloric restriction (VLCD). VLCD consisted of a 400 kcal/day diet, with percentage distribution of lipids, proteins and carbohydrates, according to Italian Standards of Care.

Both at baseline and at the end of VLCD, a hyperglycemic insulin clamp study was performed in all patients, as previously described. All studies were carried out at 08.00, after a 12-hour overnight fast, while the subjects were lying in bed, and lasted 180'. In all subjects two intravenous catheters were inserted into an ante-cubital vein and (retrogradely) into a wrist vein for substance infusion and sampling of arterialized blood, respectively, according to the hot box technique. After a 60 min period to establish baseline (-60'- 0'), at time 0' a hyperglycemic glucose clamp was carried out for the following 120', as previously described. Plasma glucose was measured at bedside every 2'-5' as needed and was clamped at 7.0 mmol/L (126 mg/dl) above baseline values. Under these conditions of constant hyperglycemia, the normal beta cell secretory response is biphasic with an early burst of insulin release within the first 10 minutes (first phase), followed by a later monotonically increasing hormone release (second phase).

Blood samples for glucose, C-peptide and insulin determinations were drawn every 2.5 min from 0 to 15 min and every 15 min from 15 to 120 min.

The acute insulin response (AIR) was calculated as the average incremental plasma insulin concentration at 2.5, 5.0, 7.5 and 10 min of the hyperglycemic clamp.

Second phase insulin response (2ndIR) was computed as the average incremental insulin concentration between 60' and 120' of the hyperglycemic clamp.

Glucose disposal during the clamp was computed as the rate of exogenous glucose infusion corrected for the (minimal) changes in the glucose pool (M value; units: µmol.min-1.m-2 BSA).

The metabolic clearance rate of glucose during the clamp was computed as the ratio between the M value and the prevalent glucose concentration. Under these experimental conditions, the metabolic clearance rate of glucose is a direct experimental measurement of the Disposition Index of second phase insulin secretion (DI; units: ml. min-1.m-2 BSA), in that it is the whole body use of glucose achieved by the beta cell at the same experimentally fixed glucose concentration. It measures whole body capability to dispose an intravenous glucose load, and it reflects beta cell adequacy to adjust to prevailing insulin resistance and insulin clearance. This DI has two advantages: 1. It is a direct experimental measure, not the product of two different experimental assessments; 2. At variance with all other DIs, it requires no assumptions regarding the mathematical relationship linking insulin sensitivity to beta cell secretory response or glucose-stimulated insulin concentrations.

Insulin sensitivity during the hyperglycemic clamp was calculated as the ratio of metabolic clearance rate of glucose divided by the average insulin concentration achieved between 60' and 120' (IS; units: [(ml. min-1.m-2 BSA)/ (pmol/L)].

The analyses of the glucose and C-peptide curves during the hyperglycemic clamp follow the general strategy proposed by several laboratories with some slight modifications, which have been previously described in detail.

The main outputs of this model are:

1. First phase parameters

• Total amount of insulin secreted due to first phase (1stISR; units: pmol.m-2 BSA)
• Glucose sensitivity of first phase secretion (σ1), expressed as the amount of insulin secreted in response to a rate of increase in glucose concentration of 1 mmol/liter between time 0 and 1 min of the study (units: [(pmol.m-2 BSA)/(mM.min-1)]

1. Second phase parameters

• Total amount of insulin secreted due to second phase (2ndISR; units: pmol.m-2 BSA);
• Glucose sensitivity of second-phase secretion (σ2), expressed as the steady-state insulin secretion rate in response to a step increase in glucose concentration of 1 mmol/liter above baseline (units: [(pmol.min-1.m-2 BSA/(mmol/L)].

Finally, an index of insulin clearance (InsClearIndex; units: L.min-1.m-2 BSA) was calculated as the ratio of the average insulin secretion rate divided by the average insulin concentration during the hyperglycemic clamp.