Rubino et al. [56] demonstrated that by excluding the duodenum and proximal jejunum without restriction of gastric volume, good glycemic control was achieved in nonobese diabetic rats in the absence of weight loss or decreased caloric intake.

One of the possible mechanisms that underlie this glucose-lowering effect is jejunal nutrient sensing. Breen et al. [57] reported that intrajejunal nutrient administration lowered endogenous glucose production in normal and streptozotocin-induced uncontrolled diabetic rats through a gut-brain-liver network without changes in insulin concentration. Moreover, when these rats were submitted to duodenal-jejunal bypass (DJB), higher concentrations of nutrients were delivered to the jejunum, causing a more profound reduction in glucose concentrations 2 days after surgery, independently of changes in plasma insulin concentrations, food intake, and body weight.

Another potential mechanism reported by Salinari et al. [58] is the action of jejunal hormones inducing insulin resistance. In this study, the authors were able to isolate jejunal conditioned medium proteins from insulin-resistant diabetic animals and insulin-resistant humans. The authors found that these proteins impaired insulin signaling, reducing glucose uptake by skeletal muscle cell cultures. A similar effect was obtained with human serum from insulin-resistant subjects, suggesting that there are circulating duodenal factors that induce insulin resistance by impairing insulin signaling. The implication of such findings is that by rerouting the food passage (e.g., after DJB), endocrine factors present in the duodenum and proximal jejunum that induce insulin resistance may halt, causing an immediate and long-standing metabolic response.

We reported the first two patients to undergo DJB with no gastric manipulation (“classic DJB”) [59]. The surgical technique involved Roux-en-Y duodenojejunostomy with a 50-cm biliopancreatic limb and 80-cm Roux limb. Both patients showed a decrease in HbA1c with no correlation to weight variation.

Later, we published our experience with 36 non-morbidly obese diabetic patients who underwent classic DJB [60]. Diabetes remission (HbA1c < 7 % and fasting plasma glucose FPG < 126 mg/dL) was achieved in 40 % of patients at 1 year of follow-up. Remission was not related to weight change in this study. We further assessed glucose and β-cell response to an oral glucose load before and at 6, 9, and 12 months after surgery [61] and compared the results with subjects with normal glucose tolerance. DJB improved β-cell function and glycemic control in overweight and class I obese subjects with T2DM. It did not normalize β-cell function when compared with the subjects with normal glucose tolerance but increased it two- to threefold compared with baseline.

Geloneze et al. [62] published their results on DJB in 12 overweight diabetic patients. Remission (HbA1c < 6.5 %, no medications) occurred in two (16.7 %) patients. This result was due to the selection of patients with a long history of diabetes and/or established macrovascular disease, slightly worsening their results regarding T2DM control. In their study, although all patients were undergoing insulin therapy before surgery, 10 (83 %) patients began taking only oral medications 24 weeks after surgery and experienced a significant decrease in HbA1c levels (8.78 to 7.84 %).

Seeking better outcomes and attempting to reproduce the results found in class 1 or morbidly obese patients, we moved forward with some technical and pathophysiological modifications as follows. We conducted our second protocol and performed a “sleeved duodenal exclusion” or “short duodenal switch” by adding sleeve gastrectomy with a 50/60-F bougie in 47 patients. The primary end points were fasting and postprandial glycemic control, and the secondary end points were lipid and hypertension control and carotid intima-media thickness, an important surrogate marker for atherosclerosis progression. In addition, based on our own studies on better metabolic/diabetes outcomes with longer limb lengths in the morbidly obese population [63], we have increased the biliary limb to 100 cm and the alimentary limb to 150 cm. We believe that resecting the gastric fundus longitudinally, thus removing part of the major ghrelin production site, may lead to slower gastric emptying, decreasing the glucose load to the intestine. Preserving the pylorus may be key in decreasing the glycemic peaks after food ingestion, leading to an improved first-phase insulin response and better glycemic outcomes. Ghrelin has the capability to decrease pancreatic insulin secretion through direct and counter-regulatory mechanisms [64, 65]. Thus, removing the main ghrelin production site would allow for better control of diabetes. With an average follow-up of 1 year, we found that adding sleeve gastrectomy and increasing the limb lengths does not add any excessive weight loss to this leaner group (total body weight loss of 6 %). In addition, so far we have seen diabetes resolution in approximately 71 % of patients, and 100 % if we include patients from remission to improvement (unpublished data). As secondary end points, we achieved control of hypertension in 67 % of patients at 12 months (≤130/80 mmHg, no or minimal medications), normalization of triglycerides in 77 % of patients, and normalization of low-density-lipoprotein (LDL) cholesterol in 81 % of patients. The carotid intima-media thickness was significantly reduced from baseline in 12 months.

Several other authors have reported their experience with DJB, as described in Table 2 [60, 62, 66–68].

Finally, a new tool for T2DM control was introduced. The DJB liner (DJBL) is an endoscopically placed device that prevents contact between partially digested nutrients and the proximal intestine, mimicking the exclusion of the proximal bowel, which is a component of several effective metabolic surgeries [69]. Escalona et al. [70] implanted the DJBL in 39 morbidly obese patients, and after 12 months, all achieved good loss of excess body weight (47.0 ± 4.4 %). The patients also showed significant improvements in waist circumference, blood pressure, total and LDL cholesterol, triglycerides, and fasting glucose.

The only report in the literature regarding low-BMI T2DM was recently published by Cohen et al. [71]. Sixteen of 20 subjects implanted with the DJBL completed the 1-year study (mean BMI of 30 kg/m²). Ten of 16 subjects (62.5 %) who completed the study demonstrated HbA1c levels of <7 % at week 52, and statistically significant lipid control was achieved (LDL and triglycerides). No significant correlations between changes in body weight and changes in FPG or HbA1c were observed. Based on the results of that study, the DJBL appears to reproduce some aspects of metabolic surgery in terms of its ability to improve HbA1c, FPG, and lipid parameters without a direct relation to weight modification. Interestingly, after some mathematical modeling of data extracted from the oral test after a mixed meal challenge glucose excursions, C-peptide deconvolution, and insulin curves, we found almost immediate, weight loss-independent improved insulin sensitivity after the placement of the DJBL. Moreover, this effect was maintained throughout the year that the device was kept in place. No improvement in insulin secretion was seen. This finding supports the animal studies of Breen and Jiao [57, 72]. The DJBL seems to be an effective tool for metabolic control, allowing some potential associations with GLP-1 analogues and DPP4 inhibitors because the device itself possibly does not change insulin secretion. It is an outpatient-based endoscopic procedure that is associated with virtually no major complications, but the main drawback is that its design is only safe for 1 year of implantation. It may be a good screening tool for the effectiveness of duodenal exclusion before a surgical procedure or an excellent way to quickly improve glycemic and metabolic control in individuals with glucotoxicity and lipotoxicity who may need prompt, effective, and safe intervention.

Laparoscopic sleeve gastrectomy with transit bipartition was proposed by Santoro et al. [73] in an observational study of 1,020 morbidly obese patients. The surgical procedure involves gastroileal anastomosis added to sleeve gastrectomy, without exclusion of the duodenum and proximal jejunum. The mean BMI was 42.2 kg/m², and excess BMI loss was 91 % and 74 % in the first and fifth postoperative years, respectively. T2DM was diagnosed in 32.6 % of patients, and 86 % went into complete remission (HbA1c < 6.5 % without medication).

Although the authors emphasize the enterohormonal pathways of metabolic surgery and the mechanisms involved in metabolic syndrome amelioration, this paper was not a diabetes treatment study. Rather, it was a study on the treatment of morbid obesity (BMI of 33–72 kg/m²) and the improvement observed in glycemic control and other comorbidities which was the result of diminished food intake and weight loss. Furthermore, the follow-up period was relatively short, and a high proportion of patients were lost to follow-up. Allowing two routes for food passage (ileal and duodenal) associated with sleeve gastrectomy makes the exact rationale behind this proposed operation unclear.

Every new T2DM treatment must be safe and effective. It must not only correct hyperglycemia but also prevent or mitigate the complications of this chronic disease. The continuing morbidity and mortality in individuals with T2DM diabetes and the lack of control even with new medications are a sign that the best management in terms of maximizing metabolic control remains elusive. Given this scenario, the option of metabolic surgery must be considered in appropriately selected individuals.