Introduction

Laparoscopic bariatric surgery has become the gold-standard treatment of morbid obesity [1, 2]. Although a recent meta-analysis has shown similar long-term outcomes for both the sleeve gastrectomy (SG) and Roux-en-Y gastric bypass (RYGB) [2], weight regain and comorbidities relapse prompted the bariatric surgeons to seek modification of currently established or introduction of new techniques [3]. Duodenal switch has proved to be the most effective procedure in terms of the long-term weight loss outcome and comorbidity resolution. However, its technical difficulty and potential adverse events have limited its widespread use [4].

Single-anastomosis duodeno-ileal switch (SADIS) is a modification of the original biliopancreatic diversion with duodenal switch (BPDDS) [5]. Due to its simpler technique and reduced number of anastomosis, SADIS has shown potentials in bariatric surgery [6]. Currently, there are case series with limited follow-up on postoperative outcome of patients undergoing this new procedure [6,7,8,9,10,11,12,13,14,15,16,17,18,19]. Our systematic review aimed to pool available data in the literature on weight loss outcome and co-morbidity resolution as well as postoperative complication and nutritional deficiency in patients undergoing SADIS for morbid obesity.

Methods

Review Design

In adherence to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines [20], a systematic review was designed using the Cochrane handbook for systematic reviews of interventions [21]. Two independent reviewers reviewed the literature and retrieved the full texts of the eligible papers for data extraction. Any conflict was resolved by consensus.

Search Strategy

PubMed/Medline, ISI Web of Science, and Scopus were queried through July 2016 using the following combination of keywords: [(“single anastomosis” or “single-anastomosis”) and (“ileoduodenal” or “duodenoileal” or “duodeno-ileal” or “duodenal” “ileo-duodenal”)] or [(“stomach intestinal pylorus sparing” or “SIPS”)]. Manual screening of the bibliography of the retrieved papers was also performed to supplement our search protocol.

Study Selection

English papers reporting the outcomes of a single-anastomosis duodeno-ileal bypass procedure for weight loss in human were included. Review articles, commentaries, and surgical technique descriptions were excluded.

Data Extraction and Analysis

The primary endpoint of this study was the efficacy of SADIS in treatment of morbid obesity, defined as the weight loss outcome and comorbidity resolution. The secondary endpoint was the safety of SADIS, defined as the incidence rate of postoperative complications and nutritional deficiencies.

Extracted data were pooled to estimate an overall result on weight loss outcome, comorbidity resolution rate, and the incidence of postoperative technical and metabolic complication. The values are presented as mean ± standard deviation (SD) or number (percentage, %).

Results

Search Result

A total of 117 records were identified through initial literature search. Additionally, 11 papers were identified by manual search of the supplementary sources. After duplicate removal and title/abstract screening, full-texts of 34 articles were reviewed. Of these, 14 papers [5,6,7,8,9,10,11,12,13,14,15,16,17,18] were eligible of which two papers [5, 14] were excluded from quantitative analysis due to overlapping data on the same group of patients (Fig. 1). Publication time frame ranged from 2013 to 2016 with the majority of the SADISs performed in 2015 (280 patients, 48.2%), 2016 (181 patients, 31.1%), and 2013 (100 patients, 17.2%) (Fig. 2).

Fig. 1
figure 1

PRISMA flowchart for this systematic review search protocol

Full size image
Fig. 2
figure 2

Number of overall SADIS performed per year

Full size image

Characteristics of Included Studies and SADIS Patients

Twelve eligible studies comprised five cohorts [6,7,8, 13, 18], four case series [9,10,11,12], and three case reports [15,16,17] encompassing a total of 581 SADIS patients (217 males and 364 females). Patients’ age and BMI ranged between 18 and 71 years and 33–71.5 kg/m2. Of 581 SADIS surgeries, 508 (87.4%) were a primary and 73 (12.6%) were a revision procedure. T2DM (354 patients, 60.1%), HTN (286 patients, 49.2%), and GERD (136 patients, 23.4%) were the most common reported comorbidity in SADIS patients. Follow-up was available between 6 and 60 months after SADIS (Table 1 ).

Table 1 Characteristics of included studies on single-anastomosis duodeno-ileal switch (SADIS)
Full size table

Technical Considerations of SADIS According to Different Studies

SADIS was variably named in different studies as stomach intestinal pylorus sparing (SIPS) surgery (258 patients, 44.4%) [8, 12, 13, 15], single anastomosis duodeno-ileal bypass with sleeve (SADI-S) (197 patients, 33.9%) [6, 9], single anastomosis loop duodenal switch (SALDS) (57 patients, 9.8%) [16, 17], single anastomosis duodenal-jejunal bypass with sleeve gastrectomy (SADJB-SG) (50 patients, 8.6%) [7], and single anastomosis duodeno-ileal bypass (19 patients, 0.3%) [16] (Table 2).

Table 2 Technical considerations of SADIS
Full size table

Length of common limb was reported by nine studies (528 patients, 90.9%) [6, 8,9,10,11,12,13, 17, 18]. Of these, the common limb was 300 cm in 315 patients (54.2%), followed by 250 cm and 200 cm in 135 (23.2%) and 78 (13.4%) patients, respectively. Similarly, bougie size was reported by ten studies (560 patients, 96.4%) [6,7,8,9,10, 12, 13, 15, 17, 18]. Of these, bougie size 42 (252 patients, 43.4%) and 54 (197 patients, 33.9%) were the most common ones to be used.

Anastomosis technique was described by eight studies (564 patients, 97%). Of these, linear stapler was used in three studies (151 patients, 26.7%) [6, 7, 17] but a hand sewn suture technique in seven studies (413 patients, 73.3%) [6, 8, 11,12,13, 15, 18]. Only two studies (113 patients, 19.4%) reported the hand sewn or linear stapler [9, 10]. Additionally, two studies (150 patients, 25.8%) stated the reinforcement of the staple line in the sleeve invagination (100 patients, 66.7%) [7] and staple line (50 patients, 33.3%) [6].

Postoperative Complications

Sixteen types of complications were reported postoperatively by the included studies in 28 patients. The overall complication rate following SADIS was 4.8%. Diarrhea was the most common complication (six patients, 1.2%) reported by three studies (188 patients) [8, 12, 18] (Table 3). However, diarrhea in one patient was due to clostridium difficile [12]. Other complications included staple line leak, anastomosis leak, wound infection, dysphagia, sleeve stricture, bowel obstruction, gastroesophageal reflux disease each in two patients (0.34%) and abscess, incisional hernia, hemoperitoneum, postoperative bleeding, internal hernia, umbilical hernia, and ileus each in one patient (0.17%). The average number of bowel movements was reported by three studies (213 patients, 36.6%) ranging between 2.1 and 2.5 times per day [8, 12, 18].

Table 3 Postoperative complications after SADIS
Full size table

Co-morbidity Resolution

Co-morbidity resolution was observed in 183 patients out of 247 ones with T2DM (74.1%), 131 patients out of 136 ones with HTN (96.3%), 69 patients out of 101 ones with dyslipidemia (68.3%), 19 patients out of 30 ones with OSA (63.3%), and 7 patients out of 8 ones with GERD (87.5%) (Table 4).

Table 4 Comorbidity resolution in patients of the included studies
Full size table

Weight Loss Outcome

Weight loss was reported by nine studies (575 patients, 98.9%) at variable time intervals after the surgery [6,7,8,9,10, 12, 13, 15, 18]. The maximum reported %EWL was 49% at 3 months (17.8–49%), 80% at 6 months (41–80%), 95% at 1 year (63.2–95%), and 100% at 2 years (72–100%) (Table 5). On the other hand, the average %EWL was 30% at 3 months (17.8–49%), 55% at 6 months (41–80%), 70% at 1 year (63.25–95%), and 85% at 2 years (72–100%) (Fig. 3 and Table 5).

Table 5 Postoperative weight loss outcome (%EWL) after SADIS
Full size table
Fig. 3
figure 3

Pattern of %EWL at different time intervals after SADIS

Full size image

Nutritional Disturbances

Nutrient deficiencies were inconsistently reported in the literature and only by four of our included studies (213 patients, 36.6%) [6, 9, 10, 18]. Selenium, zinc, and iron were the most common deficient minerals (in up to 50% of the reported cases), vitamin A deficiency in up to 53% of the reported patients, and protein deficiency in up to 34% of patients (Table 6). Overall, vitamin A, selenium, and iron deficiency were the most common nutritional disturbances after SADIS (Fig. 4).

Table 6 Nutritional deficiency after SADIS
Full size table
Fig. 4
figure 4

Nutrient deficiency at the longest follow-up after SADIS

Full size image

Revisional vs. Primary SADIS

There was no technical difference between the primary or the revisional SADIS in terms of the length of the efferent limb, bougie size, and the stapler type used for duodeno-ileal anastomosis (Table 2). On the other hand, staple line leak, wound infection, diarrhea, dysphagia, upper GI bleeding, hemoperitoneum, internal herniation, GERD, and bowel obstruction were more commonly reported with the revisional SADIS. There was no sufficient data to comment on any potential difference between these types of SADIS in terms of weight loss, comorbidity resolution, or nutritional deficiencies.

Discussion

SADIS was introduced in 2007 by Sánchez-Pernaute as a modification of biliopancreatic diversion with duodeno-ileal switch (BPDDS) in which after sleeve gastrectomy, the duodenum is anastomosed to an ileal loop in a Billroth-II fashion [4]. Although the authors reported the first clinical outcome of SADIS later in 2010 [5], a series of studies were successively published on the utility and feasibility of this new technique [5,6,7,8,9,10,11,12,13,14,15,16,17,18]. Numerous terms have been used for SADIS, including single-anastomosis duodenal-jejunal bypass, single anastomosis loop duodenal switch, single-anastomosis duodeno-ileal bypass with sleeve gastrectomy, stomach intestinal pylorus sparing surgery, and single-anastomosis duodeno-ileal switch, all simply refer to a pylorus sparing technique gastrectomy accompanied by a proximal duodenal-ileal end-to-side bypass.

Our systematic review of 581 SADIS patients (508 primary and 73 conversion procedures) demonstrated the evolution of SADIS technique from inception over the last decade. Currently, follow-up data on SADIS outcome is available up to 60 months after the surgery. Little data is available on durability of SADIS in weight loss outcome and comorbidity resolution as well as long-term development of postoperative complications.

Technical Considerations

Almost a decade has passed since the introduction of SADIS to bariatric surgery during which surgeons have adopted a longer efferent limb [14,15,16, 18] by creating a shorter afferent limb. The modification occurred as a result of early postoperative complication and late nutritional deficiency [8, 17, 18]. A similar trend was noted over time for bougie size decreasing from 54 French [6, 9, 10] to 40 French. Having only one anastomosis makes SADIS more feasible technically than the classical duodenal switch with two anastomoses.

Postoperative Complications

Although SADIS has been developed to simplify the technique for BPDDS, it still bears the complications of a combined restrictive and malabsorptive procedure. Postoperative staple line leak, anastomosis leak, bleeding, hernia, infection and abscess formation, ileus, bowel obstruction, and diarrhea are among the reported early postoperative complications [4,5,6,7,8,9,10,11,12,13,14, 16,17,18]. Our review determined diarrhea as the most common complication after SADIS. On average, SADIS patients experience up to three bowel movements per day [6, 9, 10, 14, 18]. This is much lower than the number of bowel movements reported by patients after BPDDS [21, 22]. In a matched comparison of SADIS and RYGB, Cottom et al. showed that diarrhea/steatorrhea in SADIS patients occurs as a result of a miscounted alimentary loop, resembling the length of a common limb in the traditional BPDDS [18]. Subsequently, the diarrhea was resolved after lengthening of the loop to a 450 cm common channel. In our review, the average number of bowel movements was reported between 2.1 and 2.5 per day [8, 12, 18].

Weight Loss Outcome

The pooled percentage of excess weight loss after SADIS ranges from 17.8% in the first 3 months up to 100% after 2 years. SADIS has shown comparable weight loss to that of RYGB [18] but a superior effect than LSG [14]; however, no study has compared its efficacy against the traditional BPDDS. It is explained that early weight loss occurs due to the creation of the restrictive slim tube while the sustained weight loss is a result of the malabsorptive component. Only two studies are available with a follow-up longer than 2 years, demonstrating a sustainability of 72–95% of %EWL after the SADIS as a second step of weight loss surgery after the primary sleeve gastrectomy [6, 10]. In comparison, revisional BPDDS after a failed RYGB has shown a %EWL of 29.2% at 2-year follow-up [23]. Nevertheless, SADIS data on the long-term weight loss is lacking and only 127 patients have a minimum of 2-year follow-up.

Comorbidity Resolution

Our pooled analysis revealed a resolution rate of 74.1% and 96.3% for T2DM and HTN, respectively. For other obesity comorbidities, SADIS achieved a resolution rate of > 60%. The superiority of BPDDS than that of other bariatric procedures in comorbidity resolution has been well established by previous meta-analysis [2, 24]. Rapid transition of gastric content into the small bowel and early exposure to the terminal segment of the gastrointestinal tract has been accounted responsible for incretin secretion [1]. This is responsible for the discrete effect of bypassed GI route, besides the weight loss effect, in improvement of insulin resistance and post-prandial glucose. On the other hand, it seems that the resolution of other comorbidities such as HTN, HLP, and OSA is more compatible with the weight loss pattern after duodeno-ileal bypass.

Nutritional Deficiency

Nutritional deficiency is directly correlated with the length of the common alimentary limb in duodeno-ileal switch [25]. Although Vitamin A and D deficiency, anemia, and hyperparathyroidism secondary to a Vitamin D deficiency and hypocalcaemia are among the most common nutritional disturbance after a BPD/DS [26], our review also identified selenium and protein deficiency in up to 50% and 34% of SADIS patients, respectively. Compared to a traditional BPDDS with a common limb of 100 cm, a modified procedure with a longer loop of 200 cm or more, as is in SADIS, can result in similar weight loss and comorbidity resolution but potentially a lower rate of protein deficiency, micronutrient disturbance, and lower amount of prescribed vitamin A and D [14, 18, 25]. The benefits of a longer common limb in reduction of the nutrient deficiency after SADIS can be observed in the study of Cottom et al. implementing a 300-cm loop [18]. The authors reported a significantly lower rate of micronutrient and vitamin deficiency, while still achieving a %EWL of 86% at 12 months. Unfortunately, this is the only study on SADIS with a long common limb (> 200 cm) which dynamically investigated the frequency of postoperative micronutrient deficiency. Additionally, follow-up of SADIS patients in terms of nutritional and metabolic disturbance is still premature.

Limitations and Perspectives for Future Studies

This is the first systematic review on SADIS demonstrating where we stand on our way toward a modified rescue procedure in bariatric surgery. Although the pooled results collectively support the efficacy and safety of this SADIS in the treatment of morbid obesity, the technique is evolving in terms of malabsorptive and restrictive component. On the other hand, authors have repeatedly published overlapping data of patients undergoing SADIS, in forms of case reports and small case series, for which stratification and meta-analysis seem impractical. Additionally, mid-term and long-term outcomes of SADIS in comparison to the currently established procedures are not available. Future randomized controlled trials are warranted to compare SADIS outcome in terms of weight loss and co-morbidity resolution with commonly performed RYGB and LSG.

Conclusion

As a modification of traditional BPDDS, SADIS is a bariatric procedure with a combined restrictive and malabsorptive components. SADIS showed a promising short-term weight loss outcome and comorbidity resolution rate. Moreover, postoperative complications and nutritional deficiencies are comparable to the BPDDS but long-term data are missing and there is currently a high level of technical variability. On the other hand, further studies are required to measure its cost-effectiveness compared to the currently popular bariatric procedures, LSG and RYGB.