Long-Term Results of Single-Anastomosis Duodeno-ileal Bypass with Sleeve Gastrectomy (SADI-S)

The SADI-S (Fig. 1) was introduced in 2007 in our institution [1] with the intention of simplifying a difficult but very effective surgical technique to treat morbid obesity, the DS [2‐5]. The operation differed substantially from the previous one-anastomosis technique, the one-anastomosis gastric bypass (OAGB) [6], first because of the pylorus preservation, which warranted the absence of pathologic bile reflux, and second because the common channel length was always measured, as it was in the Scopinaro procedure [7, 8] and the duodenal switch [2], the operations known as biliopancreatic diversions.

SADI-S has been demonstrated to be a safe and reproducible technique [9] and offers good weight loss results in the short term [10, 11]. It results equally effective as a second-step or revisional operation after sleeve gastrectomy [12‐14]. Results on diabetes are comparable to those obtained with the duodenal switch [15], and some studies have even found a better metabolic effect thanks to the longer common channel [16].

The present manuscript reviews the experience with the SADI-S in a series of patients operated consecutively in our department and followed for at least 5 years. Second-step and revisional surgeries have been excluded to avoid bias in the interpretation of the results.

From May 2007 to December 2015, 199 patients underwent SADI-S, among which 164 had a primary SADI-S and constitute our study group. Indications for single-step SADI-S in our Department are BMI over 45 kg/m² or severe metabolic disease independently of the BMI. The mean age of the patients was 47 years (22–71), the mean weight was 124 kg (72–180) and the mean BMI was 45.8 kg/m² (34–67). There were 99 women and 65 men. Type 2 diabetes was present in 101 cases (61.5%), of which 47 were on oral therapy (46.5%), 42 were on insulin therapy (41.5%), and 12 were under diet control. Arterial hypertension (AT) was present in 91 cases (55.4%), obstructive sleep apnea (OSA) in 81 (49.3), and dyslipidemia in 118 (71.9%).

All patients underwent a thorough endocrine and psychiatric evaluation previous to surgery, as recommended by international guidelines.

The operation has been described elsewhere [1, 10‐13] (Fig. 1).

The present work reports the 5- to 10-year outcome of a series of patients consecutively submitted to primary SADI-S in one institution, and it constitutes the largest series with more than 5 years of follow-up, 139 patients, and the only one reporting data at 10 years from surgery. SADI-S is a simplification of the DS. Although the procedure has been demonstrated to behave differently to previous malabsorptive procedures—firstly because when the common channel is equal to or longer than 250 cm, it does not give rise to malabsorption—the relation with BPD-DS is so tight that we still consider SADI-S a modification of the former instead of a totally new procedure.

The first challenge of this operation was to demonstrate that weight loss was comparable to that obtained with the DS. The initial common limb was 200 cm, which signified a twofold to fourfold increment from the lengths chosen by Hess [2], but as the alimentary limb had been eliminated, it was difficult to estimate the weight loss after both modifications. The average excess weight loss at two years was 100%. However, some patients developed severe problems derived from malnutrition and had to be revised. In 2009 the common limb length was changed to 250 cm, which is the current standard length, and, in selected cases, to 300 cm. As a whole, patients submitted to SADI-S achieved a 5-year EWL of 87% and a 10-year EWL of 80%. Hess’ report of DS at 10 years [17] with an excellent 92% follow-up rate (167 patients) showed an average EWL of 75%, with a revisional rate of 3.7%, mostly due to undernutrition and protein insufficiency. Comparable weight losses are reported by Marceau and Topart [18, 19]. The only article communicating a significant higher EWL is the work of Bolckmans and Himpens [20], in which they obtain a 93% mean EWL at 10 years performing a DS with a sleeve gastrectomy made over a 34 French bougie, an alimentary limb of 250 cm and a common limb of 75 to 100 cm; the main drawback of their work is that more than 10% of their patients had to be submitted to revisional surgery for undernutrition. Recently Surve et al. have communicated an 80% EWL and 36% TWL at 6 years in a series of 750 consecutively operated patients, among which 46 were followed for more than 6 years [21]. Patients with an initial BMI over 50 kg/m² achieved a significantly lower EWL, 81 versus 90% at 5 years and 75 versus 81% at 10 years; however, when TWL was considered, no differences could be found [22].

Hypoabsorptive surgery offers a greater weight loss than less aggressive techniques, but it has the drawback of inducing a higher rate of nutritional deficiencies and the possible development of malnutrition. To counterbalance this, three rules must be strictly respected: follow-up, dietary compliance, and supplementation. We have attained a 75% follow-up rate at 10 years, which is difficult to improve. Attending to diet and supplementation, patients are encouraged to include in their daily intake at least 100 g of proteins and no more than 30% of fat. Even though supplementation is prescribed to every patient, deficiencies are detected. At 5 years, hypoproteinemia affects 25% of the patients, low vitamin D is detected in 70%, low ferritin in 56%, iron in 18%, vitamin A in 40%, and zinc in 32%. At 10 years, secondary to intestinal adaptation, these numbers are fairly improved. Deficiencies detected are comparable to those reported by Himpens and Topart [19, 20] after the duodenal switch, by Scopinaro after the classical form of biliopancreatic diversion [23] or after the distal gastric bypass [24]. Excluding the initial series with a short common channel, the rate of revisional surgery for undernutrition was 4.3%.

The results of SADI-S on diabetics were expected as biliopancreatic diversions are the most powerful metabolic operations [25, 26]. This is the reason why we offer SADI-S to our diabetic patients independently of the BMI, and so the rate of diabetic patients in the present series is higher than the observed in other works. The anastomosis of the duodenum with the proximal ileum reduces carbohydrate absorption because most of the hexose transporters of the intestinal cells are located in the jejunal mucosa [27]. This is shared by the Scopinaro procedure, DS, and SADI-S, but in this last one, as the common channel is longer glucose absorption is higher what results in a higher postprandial glycemia [16]. On the other hand, the longer absorptive intestinal length increases the time of interaction between the bile acids and the membrane of the enterocytes where the TGR5 receptors stimulate the secretion of GLP-1, hence improving glycemic control in diabetics [16]. Regarding dyslipidemia, although 53% of the patients had some alteration of the lipid profile, the individual analysis revealed triglyceride alterations in 20% of the patients, HDL in 10%, LDL in 36%, and total cholesterol in 8.5%, numbers slightly worse than those reported after DS [20], what is surely secondary to the increased length of the common channel were fat is absorbed.

Intestinal diseases, as inflammatory bowel diseases, mesenteric ischemia, or malabsorptive bariatric operations enhance colonic oxalate absorption in response to fat malabsorption [23, 28], probably secondarily to the binding of colonic fats to free calcium what increases the amount of unbound oxalate that is able to pass through the intestinal mucosa into the bloodstream [29]. After SADI-S, we observed a small number of cases with nephrolithiasis and hyperoxaluria, with a higher rate in patients with a shorter common limb. However, the rate of urolithiasis that we found in our series is not higher than that reported after Roux-en-Y gastric bypass [30].

A frequent concern after sleeve-based operations is the development of gastroesophageal reflux disease. We found advanced esophagitis in only 8.3% of the cases submitted to upper gastrointestinal endoscopy, which is a significantly lower proportion than could be expected, and we attribute it to the wider sleeve gastrectomy associated with the procedure.

SADI-S offers a greater and more stable weight loss in the long term than gastric bypass [31‐33]. But more important is the absence of secondary complications requiring reoperation. Obeid NR et al. reported a 20% rate of reoperation in the first 10 to 13 years after a Roux-en-Y gastric bypass for long-term complications of the initial operation, internal hernia, obstruction-, or gastro-gastric fistula [32]. Duvoisin et al. report 14% of complications requiring surgery after gastric bypass, mainly anastomotic complications and internal hernias [31]. In the present series, excluding cholecystectomy, the rate of reoperation for causes related to the primary surgery is 8.5%, and it has decreased significantly after abandoning the short common channel operation. The number of patients requiring cholecystectomy in the follow-up was 7.7%, comparable to that reported by Higa after gastric bypass [33]; no prophylactic cholecystectomy is recommended.

The incidence of cancer in our study, 4.2% or 6 cases per 1000 patient-years is comparable to that reported by Sjöstrom et al. corresponding to the Swedish Study [34] and close to the reported by Adams et al. in 2009 with more than 6000 patients followed for a mean time of 24 years; they report a total cancer incidence of 3.8% with a total incidence density rate of 3.1 cases per 1000 person-years follow-up [35]. However, in other reported studies, the incidence seems to be much lower, as in the work from Canada of Christou et al., who report a 5-year incidence of 2.03% [36]. In a meta-analysis published in 2014 [37], the cancer incidence density rate was found to be between 1.06 and 1.08 cases per 1000 person-years. In the present series, we have not discarded the neoplastic events presented in the first 3 postoperative years, so the final results could be biased as some cancers were surely present at the moment of surgery. A longer follow-up will surely contribute to the decrease of the total incidence density rate.

Finally, the quality of life after SADI-S is much better than after old biliopancreatic diversions [18, 23], as the stool frequency in the long term is almost normal and only increased after dietary violations.

In summary, we present the long-term evolution of patients submitted to SADI-S, a safe and effective surgery for the treatment of obesity, and its comorbid conditions. This is a prospective study with the limitations of the selection of patients for the procedure and the inclusion of patients submitted to SADI-S with different limb lengths. On the other hand, the strength of the report is the long follow-up of a series of consecutive patients submitted to SADI-S in only one institution, with a 75% follow-up rate at 10 years.