In obese diabetic patients, weight loss is accompanied by a corresponding increase in insulin sensitivity and increase in β-cell activity [17, 18]. It is also well established that significant caloric restriction improves glucose tolerance in diabetic patients [19]. Altered insulin sensitivity after bariatric surgery reflects the impact of both caloric restriction and weight loss.

Profound improvements in fasting glucose concentration and insulin action have been noted early after bariatric surgery, often before significant weight loss has even occurred [20–22]. Caloric restriction, which occurs immediately after all surgical interventions, likely plays an important role in the favorable metabolic changes observed, particularly in the early postoperative period [19, 22]. For example, after losing an equivalent amount of weight over about 3 weeks through RYGB or a 500 kcal per day diet, obese diabetic patients exhibited similar improvements in insulin sensitivity, acute insulin response, and β-cell function as assessed by an intravenous glucose challenge [23]. This early improvement reflects mostly improved hepatic insulin sensitivity from caloric restriction, as the profound negative calorie balance leads to decreased liver fat and increased insulin sensitivity [24], reduced basal glucose production, and increased liver clearance of insulin [25, 26].

Subsequent change in peripheral insulin sensitivity varies by surgical procedure. After LAGB, further improvement in peripheral insulin sensitivity occurs more gradually and is in proportion to the amount of weight loss [12]. In contrast, RYGB leads to profound changes in insulin sensitivity and insulin action much more rapidly [27–32]. Moreover, several studies have noted greater improvement in T2DM and greater increase in insulin sensitivity after RYGB over time, even after equivalent weight loss by dietary or restrictive interventions, suggesting weight-independent mechanisms [27, 28, 33]; however, these studies did not necessarily control for equivalent daily caloric intake. Others observe that the increased insulin sensitivity after RYGB occurs in proportion to weight loss, which is greater after malabsorptive surgery [34, 35].

Improved peripheral insulin sensitivity is greater after BPD than after other procedures and occurs as early as 2 weeks postoperatively [6, 36, 37]. As noted by Mari et al., BPD increases insulin sensitivity regardless of the baseline glucose tolerance, and leads to a significant decrease in insulin hypersecretion. Insulin sensitivity after BPD has even been observed to exceed that of obese patients with normal glucose tolerance [37]. Neither weight loss nor GLP-1 secretion after oral glucose or meals appears to correlate with improved insulin sensitivity after BPD [36]. The factors contributing to improved insulin sensitivity remain unclear, although insulin sensitivity after BPD is greater after an oral than an intravenous challenge, consistent with a mechanism linked to bypass of the duodenum and jejunum [38]. In addition, the diversion of bile acid to the ileum leads to decreased lipid absorption and increased bile acid reabsorption, which may affect glucose metabolism through increased binding of bile acids to farnesoid X receptor (FXR ) [39, 40]. Increased activation of FXR in the ileum activates fibroblast growth factor-19 (FGF-19), which then binds to the fibroblast growth factor receptor-4 (FGFR-4) and suppresses hepatic gluconeogenesis [40]. This mechanism is also likely to contribute to improved insulin sensitivity after RYGB and SG.

While the long-term effects of SG on weight loss and glycemic control need further study, SG has demonstrated metabolic effects beyond that of purely restrictive procedures such as LAGB. Changes in insulin sensitivity after SG are thought to be similar to RYGB, although results have been discrepant between studies [41–44]. Abbatini et al. compared LAGB, SG, and RYGB and observed that T2DM improved similarly for RYGB and SG. Sleeve gastrectomy actually led to greater improvement in insulin sensitivity than RYGB when assessed by a euglycemic hyperinsulinemic clamp in the absence of any hypoglycemic medication [43]. Another study that assessed insulin sensitivity by applying a mathematical model to data from a mixed meal tolerance test found greater improvement in insulin sensitivity after RYGB than SG [41]. Differences in study population and mechanistic studies may account for the discordant results, but further research into the impact of sleeve gastrectomy on glucose homeostasis is needed.

In summary, insulin sensitivity improves after bariatric surgery and weight loss. Initial improvement occurs largely due to caloric restriction and primarily reflects improved hepatic insulin sensitivity, while subsequent improvement in peripheral insulin sensitivity occurs secondary to weight loss and possibly due to mechanisms specific to the surgical technique.

Assessment of β-cell function is complicated because β-cells adapt to chronic stimuli with a standard set point of secretory capacity, but must also be able to respond to acute challenges such as feeding by promptly releasing sufficient insulin to control glycemia. Bariatric surgery leads to greater insulin sensitivity and insulin action which relieves secretory pressure on the β-cell and leads to a decrease in total insulin secretion; it also improves β-cell response to dynamic changes.

Basal β-cell function is reflected in fasting insulin and total insulin secretion. β-cell responsiveness to dynamic change is assessed using parameters such as the insulinogenic index (change in insulin relative to change in glucose over an interval of time), acute insulin response (AIR) to intravenous glucose, and β-cell glucose sensitivity (measured as the slope of the insulin secretion to plasma glucose dose-response relationship) [45, 46]. Disposition index (DI) is an overall measure of β-cell function that combines both insulin sensitivity and insulin secretion [15]. Comparison of insulin secretion patterns and changes in peripheral insulin and glucose levels should take into account the altered dynamics of accelerated transit and absorption of nutrients after procedures such as RYGB, SG, and BPD.

After LAGB, the insulin curve is characterized by a parallel downward shift in concentration, consistent with increased insulin sensitivity [47, 48]. Early and dynamic insulin secretion is not necessarily improved, but DI increases after at least moderate weight loss [30, 33, 47]. Overall peripheral insulin levels decrease in proportion to the degree of weight loss and are more consistent with increased hepatic clearance than reduced insulin secretion [12].

Insulin secretion after oral ingestion of nutrients is singularly altered after RYGB, with an earlier and exaggerated rise in insulin concentration followed by rapid decline [28, 35, 48–53]. Intrinsic β-cell function recovers early on after RYGB, with several studies reporting an improved acute insulin response [18, 26, 31, 35, 47, 52, 54, 55], although residual glucose intolerance persists [35, 49, 53]. Fasting insulin and total insulin output decrease after RYGB [31, 35, 48]. Insulinogenic index also increases after RYGB [31, 49, 53, 57]. In contrast to LAGB, β-cell glucose sensitivity increases early after RYGB in response to an oral glucose challenge, although responsiveness to intravenous glucose is most notable several months after surgery, in parallel with weight loss [25, 35, 47]. Dutia et al. found that β-cell glucose sensitivity , DI, and insulin secretion rapidly and markedly improved after an oral glucose tolerance test but these early changes were less apparent after an intravenous glucose tolerance test, highlighting the importance of the oral route and gastrointestinal factors in the improvement of β-cell function after RYGB.

Further evidence of the impact of altered nutrient transport and glucose absorption after RYGB on β-cell function and insulin action is provided by the rare cases of severe postprandial hypoglycemia that arise only after RYGB but not LAGB [57–59]. Islet cell hypertrophy may be involved based on surgical specimens from symptomatic patients treated with partial pancreatectomy, but this has been disputed by later studies [58–60]. While the pathogenesis of this syndrome remains unclear, the condition is characterized by excessive insulin response despite improved insulin sensitivity [57, 61]. Given the exaggerated GLP-1 response following RYGB, it has been hypothesized that enhanced GLP-1 secretion contributes to excessive insulin response, and affected patients have been found to have greater insulin and GLP-1 response to meal tests relative to other postsurgical patients without postprandial hypoglycemia [57, 61–63]. One study suggested that postprandial hypoglycemia could be successfully treated with a GLP-1 receptor antagonist, which avoided postprandial hypoglycemia by decreasing the exaggerated release of GLP-1 and insulin [63].

In contrast to the exaggerated insulin response after RYGB, the major effect of BPD is a rapid and significant decrease in insulin resistance [37]. Total insulin secretion decreases in parallel with the increase in insulin sensitivity in all patients after BPD, regardless of baseline glucose tolerance [36, 64]. Improvement in insulin sensitivity and β-cell glucose sensitivity, as well as acute insulin response and DI, occurs as early as 1 week after BPD, before significant weight loss has occurred [36–38, 64]. The early increase in insulin sensitivity and greater insulin response after oral glucose challenge but not to intravenous glucose challenge implicate early and rapid delivery of unabsorbed nutrients to the distal small intestine as a mechanism for improvement in T2DM [38].

Sleeve gastrectomy , like RYGB, leads not only to improved insulin sensitivity but improved insulin secretion and an exaggerated postprandial increase in GLP-1 [42, 43, 65, 66]. Insulin sensitivity improves rapidly and markedly after SG and fasting insulin decreases similarly to RYGB [7, 42, 67]. While further studies are needed, an increase in the insulinogenic index and early insulin response after SG correlates with baseline C-peptide levels, suggesting that SG similarly increases β-cell function [67].

In summary, β-cell function progressively increases over time according to weight loss after restrictive procedures, leading to lower fasting insulin and total insulin secretion. RYGB and BPD also improve the dynamic responsiveness of the β-cell often before significant weight loss has even occurred. While altered nutrient transit and glucose absorption may partly explain the changes in acute insulin response and β-cell glucose sensitivity, concurrent changes in gut hormone secretion also influence further adaptation of β-cell function .

Caloric restriction and weight loss clearly play a significant role in the improvement of insulin resistance and β-cell dysfunction after bariatric surgery. However, early and profound improvement in insulin sensitivity and insulin action, before significant weight loss has even occurred, as well as the magnitude of improvement after surgeries such as RYGB and BPD, suggests that mechanisms independent of weight loss and specific to the surgical intervention may also account for the sustained and marked improvement of T2DM.