Study protocol of the DUCATI-study: a randomized controlled trial investigating the optimal common channel length in laparoscopic gastric bypass for morbid obese patients

Obesity is a global problem in Western societies. The World Health Organization (WHO) states that currently obesity is the most serious chronic health problem worldwide [1]. In some western industrial countries over 50 % of the population is overweight. The mortality risk is doubled for women with BMI of >40 kg/m² and tripled for men. Moreover, the earlier obesity occurs with regard to age, the stronger the rise in mortality risk [2].

Treating morbid obesity (defined as a BMI > 40 kg/m²) by non-surgical methods has only resulted in minor effects. Surgical treatment has gained widespread acceptance in recent decades due to its superior outcomes in weight loss [3–5]. In 1992 the WHO stated that bariatric surgery is the only effective treatment to achieve long lasting weight loss [1]. Furthermore, bariatric surgery has profound effects on obesity-related co-morbidities, such as type 2 diabetes mellitus, hypertension and sleep apnea, [6]. Bariatric surgery is also safer than conservative treatment for obesity in terms of a lower lifelong mortality risk caused by morbid obesity [7].

Surgical interventions for morbid obesity can be divided into two categories: interventions that cause food malabsorption and interventions that cause food restriction. Laparoscopic Roux-en-Y gastric bypass (LRYGB) has been acknowledged to be safe and effective for the treatment of morbid obesity, and is globally considered the golden standard surgical technique [1, 7–9].

After one year follow-up, LRYGB results in an average of 60–80 % excess weight loss (% EWL) with reduction in comorbidities. However, a partial weight regain can be seen during long-term follow-up [10].

Consensus does exist on the need for a small calibrated gastric pouch in the LRYGB in order to achieve improved % EWL results. However, to date little attention has been given to determine the optimal lengths of the various intestinal limbs that comprise the operation (the alimentary or ‘Roux’ limb and the biliopancreatic limb or afferent limb) (see Fig. 1). These limb-lengths are important as they determine the amount of weight loss and the incidence of adverse events related to malnutrition and malabsorption.

No consensus has been reached on the optimal lengths of the limbs. In an internet-based study performed by Madan et al. on preferred technique of 215 bariatric surgeons, all of whom are members of the American Society for Metabolic and Bariatric Surgery, showed a wide range of the AL and BL limbs lengths (average 110 [range 35–250]cm for AL, and 48 [range 10–250]cm for BL) [11].

The common channel (where the food arriving through the alimentary limb (AL) is mixed with gastric, bilary, and pancreatic fluids coming from the biliopancreatic limb (BL) (Fig. 1), absorbs most of the nutrients and preserves enterohepatic circulation of bile salts and fat [12]. Some absorption of amino acids and glucose may occur in the AL as well, through the saliva and succus which may digest proteins and carbohydrates. In the BL, as the digestive fluids pass through its mucosa, reabsorption of digestive components such as lipases may also occur, which decrease the digestive potential of the fluid reaching the food in the common channel [12].

In the specific population of super-obese patients it has been suggested that an increase in malabsorbtion may be established by creating a longer AL, resulting in an increase in weight loss in this selected group of patients [13, 14].

A review of the available literature on different limb lengths in LRYGB identified four randomized controlled trials (RCT).

It has been concluded that the currently available literature supports the notion that a longer AL (at least 150 cm) may be associated with a modest weight loss in the short-term for superobese patients, but without a significant impact on patients with BMI < 50 [13, 14]. Limitations of the available literature include the heterogeneity of studies, small study populations, and lack of standardization. Furthermore, the length of the common channel was not accounted for in these studies.

Nevertheless, there is evidence that the degree of malabsorption after gastric bypass is influenced mainly by the length of the common channel [15]. Therefore it has been suggested that good quality studies that focus on the impact of the length of the common channel on weight loss are urgently needed [14].

When searching for the optimal lengths of the bypass limbs, it is known that one cannot lengthen the alimentary limb without facing metabolic complications. Complications associated with biliopancreatic diversion and very long-limb RYGB are mostly due to a too short common channel [16]. Kalfarentzos et al. [17] found that 100 cm common channel in biliopancreatic diversion with RYGB and long limbs is a safe length for achieving satisfactory % EWL without major nutritional deficiencies.

In 2006 Nelson et al. [18] described a “very, very long limb” Roux-en-Y gastric bypass in a series of super obese patients in which a 100 cm CC was created, with a 60 cm BL and a “very, very” long Roux or AL of 400 to 500 cm. In this study 82 % of patients lost >50 % of excess body weight. Nine (4 %) patients required re-operation due to protein/calorie malnutrition in this group by proximal relocation of the enteroenterostomy with symptom resolution in all patients.

Recently Thurnheer et al. [19] published on a so called distal very long RYGB with an average CC of 76 cm (±7 cm), BL of 79 cm (±14 cm) and variable AL lengths of 604 cm (±99 cm). Average EWL in this population of 355 patients was >74 % up to 5 years after the operation and failure rate was relatively low (<6 %). The incidence of severe malnutrition states was relatively low (1.7 %) and the complication rate (4.5 % major and 10.4 % minor complications) was comparable to the literature.

We hypothesized that distal LRYGB as stated in the present proposal will have similar results in terms of % EWL as the distal very long RYGB described by Thurnheer on the % EWL after 1 year (69.7 %; SD 18.3), this will be used to calculate group sizes for this study. The proposed technique was first described by the group of Sarr [20] and termed very, very long limb Roux-en-Y gastric bypass (VVLL-RYGB). We feel that very long Roux limb gastric bypass (VLRL-RYGB) describes the proposed technique more clearly and will therefore be used in this article.

Data required for the calculation of group sizes was not reliable and therefore the data from the study by Thurnheer [19] will be used for this purpose.

In conclusion, currently there is no consensus about the optimal length of the alimentary, biliopancreatic limb and common channel for achieving best weight loss results, and more studies are needed to determine the role of the common channel in LRYGB.

We hypothesize that VLRL-RYGB (with standardised common channel length of 100 cm) will result in a better weight loss result, in terms of % EWL after one year when compared to standard LRYGB.

As mentioned in the background section, there are currently no randomized controlled trials addressing the role of the common channel in the treatment of the morbidly obese patient. Savassi et al. [24] addressed this issue by evaluating the length of the common channel in 100 patients that underwent LRYGB. They found a negative correlation (though weak) between weight loss and common limb length. Average common limb length was approximately 500 cm (range 268–829 cm) in patients receiving a standardized 110 cm AL and variable BL (average 56 cm, range 30–105 cm). They concluded that the common channel length does not play a role in weight loss for proximal (or short limbs) LRYGB but the variability in jejunoileal and common channel length should be considered when making modifications in the AL and BL.

Hernandez-Martinez et al. [25] performed LRYGB with a fixed common channel length of 230 cm in a series of 565 patients. After measuring the entire small bowel a length of 230 cm common channel was standardized and the rest of the small intestine was distributed as 60 % AL and 40 % BL. They selected 230 cm based on the assumption that this is the normal length of small bowel in adult non-obese persons. They found promising results in achieving long lasting EWL of over 70 % during 8-year follow-up with few metabolic complications.

In Scopinaro’s biliopancreatic diversion, distal gastrectomy is performed creating a 200–500 ml gastric pouch, with a 200-cm AL, a common channel of 50 cm, and a BL of 3–4 metres. Although this approach seems to be superior to RYGB for long term weight loss in super-obese patients [16, 26–29], high rates of metabolic complications are a concern [16, 28, 30, 31].

We hypothesize that our proposed VLRL-RYGB will provide for improved results concerning % EWL with an acceptable rate of (metabolic) complications in accordance with the results described by Thurnheer et al. [19] and Kalafarentzos et al. [17].

Our main point of interest is to determine if the VLRL-RYGB is a superior alternative to standard LRYGB in terms of percentage excess weight loss and to put more focus on the role of the common channel. Therefore we will perform this randomized controlled trial comparing both techniques, with % EWL as a primary outcome.