Long limb compared with standard limb Roux-en-Y gastric bypass for type 2 diabetes and obesity: the LONG LIMB RCT

Article history

The research reported in this issue of the journal was funded by the EME programme as project number 13/121/07. The contractual start date was in February 2015. The final report began editorial review in May 2019 and was accepted for publication in December 2019. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The EME editors and production house have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the final report document. However, they do not accept liability for damages or losses arising from material published in this report.

Copyright statement

© Queen’s Printer and Controller of HMSO 2021. This work was produced by Miras et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

2021 Queen’s Printer and Controller of HMSO

Background

At least 11 randomised controlled trials (RCTs) have demonstrated that bariatric surgery, and, in particular, the Roux-en-Y gastric bypass (RYGB), is substantially more effective than intensive medical care for the treatment of the hyperglycaemia of type 2 diabetes mellitus (T2DM). 1,2 The effects of surgery are so profound that approximately 50% of patients achieve ‘diabetes mellitus remission’, (i.e. euglycaemia) in the absence of glucose-lowering medications. 3

The anatomical rearrangements of RYGB result in three intestinal segments or ‘limbs’: the ‘alimentary limb’, through which food enters the small intestine; the ‘biliopancreatic limb’, which includes the bypassed segments of duodenum and proximal jejunum, through which the biliopancreatic secretions flow; and the ‘common limb’, in which food and biliopancreatic secretions mix (Figure 1).

FIGURE 1.

Schematic drawing of the standard limb and the long limb RYGB. a, Standard limb; b, long limb. BP limb, biliopancreatic limb. Reprinted with permission from The American Diabetes Association. Copyright 2020 by the American Diabetes Association. 4

Reprinted with permission from The American Diabetes Association. Copyright 2020 by the American Diabetes Association. 4

The profound improvements in glucose control after RYGB have led to the recognition of the intestine as an organ with a major impact on glucose regulation. Thus, surgeons have experimented with different intestinal limb lengths to enhance the clinical effect of RYGB. However, the optimal length of each of these limbs remains controversial, with substantial variation in practice. The reason underlying this inconsistent clinical practice is that the physiological role of each of the limbs in glucose regulation has until recently been unclear. Indeed, it is challenging to determine the precise physiological impact of each of these intestinal segments because changes in the length of one will invariably result in the change in the length of the others. The matter is complicated further by the variability in the total length of the human small intestine. 5

Although many of the benefits of RYGB on glucose control can be attributed to weight loss, both early and longer-term substantial improvements in glycaemia also take place independently. This has led to the concept of ‘metabolic surgery’. 6,7 Human and rodent studies suggest that the bypass of the proximal intestine might be the component of RYGB underlying, at least in part, its weight loss-independent effects on glucose regulation. 8 Beta cell function and early postprandial release of insulin are enhanced after RYGB. 9 The prevailing view is that the dominant mechanism underlying this observation is the early and enhanced secretion of the incretin hormone glucagon-like peptide 1 (GLP–1). 10 It is thought that the rapid delivery of nutrients to the enteroendocrine cells (EECs) of the distal small intestine triggers the exaggerated release of GLP-1 within the gut and the circulation. 11 At the same time, the simultaneous postprandial release of other hormones from the EECs, such as peptide YY (PYY) and oxyntomodulin, leads to synergistic effects on increased satiety and, perhaps, increased energy expenditure. 12,13

Hypothesis

The aim was to address the gap in knowledge with regard to the optimal lengths of the RYGB limbs through the understanding of the physiology of glucose regulation after surgery. Therefore, a reductionist approach was applied to examine the effects of a longer biliopancreatic limb on glucose control in this double-blind, mechanistic RCT. It was hypothesised that a long biliopancreatic limb RYGB would enable an even faster delivery of undigested nutrients to the distal small intestine, resulting in an even greater release of GLP-1 and insulin, than a ‘standard’ biliopancreatic limb RYGB.

Trial design

This was a prospective, randomised, double-blind RCT. Fifty-three patients with T2DM and obesity due to undergo RYGB surgery were recruited from the Imperial College Healthcare NHS Trust and King’s College Hospital NHS Foundation Trust by the clinical and the research team and randomised at a ratio of 1 : 1 to either a 150-cm (long limb) or a 50-cm (standard limb) RYGB while keeping the alimentary limb constant at 100 cm (see Figure 1). Both the patient and the clinical/research teams (except the operating surgeon) were blinded to treatment disposition. Participants were randomised to either long limb or standard limb RYGB surgery in a 1 : 1 ratio using an online randomisation program (www.randomisation.com; accessed 1 July 2015) by Dr Victoria Salem, who was not otherwise involved in the trial. No stratification variables were used.

Inclusion and exclusion criteria

Key inclusion criteria were aged 18–70 years, a diagnosis of T2DM treated with at least one glucose-lowering medication, a body mass index (BMI) ≥ 30 kg/m² and eligible for metabolic surgery based on the UK’s National Institute for Health and Care Excellence Clinical Guideline Number 189. 14 Key exclusion criteria were any surgical, medical or psychological contraindications to metabolic surgery, pregnancy and currently breastfeeding.

Ethics approval

The trial was approved by the West London Research Ethics Committee (reference number 15/LO/0813) and registered in the International Standard Randomised Controlled Trial registry (as ISRCTN 15283219). Written informed consent was obtained from all patients prior to participation in the trial.

Intervention and follow-up

Patients were assessed by the multidisciplinary clinical team as part of routine NHS care preoperatively and at 2 weeks, and 3, 6 and 12 months after surgery, unless clinical need dictated more frequent consultations. Operations were performed laparoscopically by four surgeons, who followed a standard operating protocol agreed before the trial commenced (see Appendix 1). The procedures were filmed to enable independent assessment of the consistency of the surgical technique among the operating surgeons. The total length of the small intestine was measured from the ligament of Treitz to the ileocaecal valve. This was performed using set distance markers on laparoscopic graspers, running the bowel segment by segment along the antimesenteric border. The management of glucose-lowering medications was performed by a single consultant diabetologist (ADM), who was blind to treatment allocation. Glucose-lowering medications were discontinued during the course of the 12-month follow-up depending on glycated haemoglobin (HbA_1c) concentrations and capillary glucose measurements, and when clinically safe. Glycaemic remission was defined based on a variation of the American Diabetes Association’s criteria15 as an HbA_1c concentration < 48 mmol/mol and fasting glucose level of < 5.6 mmol/l in the absence of glucose-lowering medication for a minimum of 12 months. Micronutrient supplementation was based on the British Obesity & Metabolic Surgery Society’s guidance. 16

Mechanistic visits

Mechanistic assessments took place at three time points: preoperatively, at 2 weeks after surgery to examine the effects of the interventions before substantial weight loss has taken place and when 20% of weight loss was achieved in order to remove weight loss as a confounding variable. Five days prior to the mechanistic visits, all glucose-lowering medications were discontinued and intermediate-acting insulin used as ‘rescue’ treatment if necessary. Patients were asked to refrain from consuming alcohol and strenuous physical activity for 48 hours before the visit. The patients were admitted to the Imperial College London or King’s College London National Institute for Health Research (NIHR) clinical research facilities in the evening and consumed a standardised meal. The next morning the patients underwent a two-stage hyperinsulinaemic–euglycaemic clamp with the stable isotope [6,6-²H₂]glucose using a validated protocol. 17 Stage 1 consisted of an insulin infusion at 0.5 mU/kg/minute (low dose) for 120 minutes to measure hepatic insulin sensitivity based on endogenous glucose production; and stage 2 consisted of an insulin infusion at 1.5 mU/kg/minute (high dose) for 120 minutes to measure peripheral insulin sensitivity based on glucose uptake. On the morning of the third, and final, day of their visit the patients underwent a mixed-meal tolerance test. Blood samples were obtained before and at 180 minutes following a liquid meal [Ensure^® Compact (Abbott Laboratories, Abbott Park, IL, USA), 300 kcal in 125 ml].

Trial outcomes

Sample analysis

Plasma and serum samples were stored at –80 °C until further analysis. Glucose was measured on the ARCHITECT ci8200 (Abbott Laboratories) platform using a hexokinase method. Insulin was measured using the ARCHITECT i2000SR (Abbott Laboratories) immunoassay. GLP-1, PYY and glucose-dependent insulinotropic polypeptide (GIP) were measured using the MAGPIX^® (Luminex Corporation, Austin, TX, USA) assay. Glucose isotopic enrichment was measured by gas chromatography–mass spectrometry on a HP 5971 A mass selective detector (Agilent Technologies, Inc., Santa Clara, CA, USA). Rates of glucose appearance (Ra) and disappearance (Rd) from plasma were calculated using non-steady-state equations proposed by Steele et al. 18 and modified for stable isotopes.

Sample size calculations

The majority of published studies have shown that peak active GLP-1 concentrations are approximately twofold greater after standard limb RYGB,19,20 compared with preoperatively. It was estimated that peak active GLP-1 levels after long limb RYGB will be tripled at 2 weeks after surgery. The trial was powered to detect a minimum clinically significant difference between the groups in mean peak active GLP-1 of 10.0 pmol/l, assuming a standard deviation (SD) of 10.8 pmol/l in each group. With a sample size of 20 completers in each arm, the statistical power was 80% to detect this difference at an alpha of 0.05. Based on the centre’s experience, a predicted 20% dropout rate was predicted; therefore, the recruitment target was 25 patients in each arm.

Statistical analyses

A detailed statistical analysis plan is available in Appendix 2. In summary, continuous variables were summarised using the number of (non-missing) data points, mean and SD if found to follow a normal distribution. Continuous variables not found to be normally distributed were summarised by the number of data points, median and interquartile range (IQR). Categorical variables were summarised by the frequency and percentage (based on the non-missing sample size) of values in each category. All the analyses presented in this report were based on the full analysis population, which consisted of patients in the groups to which they were randomised, regardless of deviation from the protocol or whether or not they received the allocated surgery. Patients with completely missing data at the outcome time point were excluded from this data set for the particular outcome for which they had missing data.

The analysis of the primary outcome was performed using analysis of covariance (ANCOVA). In the analysis, the peak of active GLP-1 concentration at the early mechanistic postoperative visit at 2 weeks was considered as the outcome measure, whereas the baseline peak of active GLP-1 was included as a covariate. The baseline-adjusted differences in outcome values between groups were reported, along with a corresponding 95% confidence interval (CI). Secondary outcomes were measured on a continuous scale, with a baseline measurement, and were analysed using a similar approach to that outlined for the primary efficacy outcome. The data from each postoperative time point will be analysed in a separate analysis. For continuous secondary outcomes with no baseline measurement, the two groups were compared using the unpaired t-test. Alternatively, the Mann–Whitney U-test was used if the assumptions of the t-test were not met. Binary and nominal outcomes were compared between the two study groups using either the chi-squared test or Fisher’s exact test if the number of responses in some categories was low. Ordinal outcomes were analysed using the Mann–Whitney U-test to allow for the natural ordering of the response categories. Statistical significance was defined as a p-value of < 0.05. Association between outcomes were performed using Pearson’s correlation. Alternatively, Spearman’s rank-order correlation was used if the Pearson’s correlation assumptions (e.g. non-linear relationship, both variables non-normally distributed) were not met. The data analyses were performed using the statistical software packages Stata^® (version 15.1; StataCorp LP, College Station, TX, USA), IBM SPSS Statistics version 20 (IBM Corporation, Armonk, NY, USA) and GraphPad Prism (version 6; GraphPad Software Inc., CA, USA).

Public and patient involvement

The research and clinical teams engaged closely and formed an active partnership with the following key contacts during the application process: Ms Georgina Hayman, lead of the British Obesity Surgery Patient Association (BOSPA) west-London branch; and Dr Shamil Chandaria, Patron of the National Obesity Forum, an independent charity supporting patients and health-care professionals.

Ms Danielle Neal, the communications and public and patient involvement (PPI) officer, NIHR North West London, approached the Diabetes Research Network PPI group and asked members the following questions:

• What are your initial feelings about the research?

• Do you think the research question is important?

• What issues do you feel will prevent people from taking part in the study?

• Do you feel that the treatment and assessment plan will be acceptable to the participants?

This feedback direct from the patient contact influenced the direction of both the clinical trial and the mechanistic studies in this proposal.

All three PPI representatives contributed to the development of the grant application, starting from its design, and to undertaking the research and the choice of research topics, and will help with dissemination of the study findings through their organisations.

During the course of the project Ms Hayman conducted patient support groups every 2 months. These support groups served to support patients following their operations and acted as an avenue for the patient voice to be heard. The Trial Steering Committee and researchers conducting the day-to-day running of the trial obtained feedback from patients and optimised the conduct of the trial to make it more acceptable. Numerous minor and major modifications were made to the way the clinical and mechanistic assessments and follow-up were performed as a result this feedback. This helped the trial immensely with recruitment and retention. Only one patient dropped out of the trial. Patients were so excited about contributing to this important study that many refused to accept the allocated reimbursement at the end of the trial.

Patients did not take part in the data analyses. However, the above patient support groups will be approached to help disseminate the findings of the trial to the local patient and health-care communities. How this will take place practically is currently being decided on. The contribution of patients has been acknowledged in both the report and the draft manuscripts for publication.

Trial participants

Fifty-three participants were recruited into the study between August 2015 and November 2017. Twenty-seven participants were randomised to standard limb and 26 to long limb RYGB (Figure 2). For anatomical reasons, one patient in the standard limb group underwent a vertical sleeve gastrectomy and one patient in the long limb group underwent a one-anastomosis gastric bypass. The final visit of the last patient took place in December 2018. After dropouts resulting from failure to undergo mechanistic visits and loss to follow-up, 24 participants completed the 12-month mechanistic visit in the long limb group and 24 in the standard limb group.

FIGURE 2.

The Consolidated Standards of Reporting Trials (CONSORT) flow diagram. a, For more details on the full analysis population see the statistical analysis plan (in Appendix 2).

Clinical parameters at baseline

Clinical parameters were well balanced between trial groups at baseline (Table 1). The majority of the patients were middle-aged, white, European and female. The mean (SD) BMI was 42 kg/m² (6 kg/m²) in the standard limb group and 43 kg/m² (8 kg/m²) in the long limb group. Patients in the standard limb group had a mean HbA_1c level of 73 mmol/mol (SD 17 mmol/mol), a median duration of T2DM of 8 years (IQR 6–10 years) and were taking a median number of three (IQR 2–3) glucose-lowering medications (Table 2). Patients in the long limb group had a mean HbA_1c level of 76 mmol/mol (SD 16 mmol/mol), a median duration of T2DM of 8 years (SD 6–9 years) and were taking a median of 3 (IQR 2–3) glucose-lowering medications.

Primary outcome

There were significant increases compared with baseline in the postprandial peak of active GLP-1 concentration in both trial groups at the 2-week time point, but there were no significant differences between the standard and long limb trial groups (Table 3 and Figure 3). There were significant increases at the point of 20% weight loss compared with baseline in the postprandial peak active GLP-1 concentration and the area under the curve (AUC) in both trial groups; however, there were no significant differences between the trial groups.

FIGURE 3.

Glucagon-like peptide 1 response during the mixed-meal tolerance test. (a) Plasma-active GLP-1 concentration preoperatively; (b) plasma-active GLP-1 concentration at 2 weeks postoperatively; and (c) plasma-active GLP-1 concentration at 20% weight loss time point. Data were plotted as means (SDs). n = 24 in each trial group. Reprinted with permission from The American Diabetes Association. Copyright 2020 by the American Diabetes Association. 4

Reprinted with permission from The American Diabetes Association. Copyright 2020 by the American Diabetes Association. 4

Secondary outcomes

Glucose tolerance

At the 2-week time point and at the point of matched 20% weight loss, fasting and postprandial glucose concentrations (as assessed via AUCs) during the mixed-meal tolerance test were significantly reduced compared with baseline in both trial groups. However, there were no significant differences between the standard and long limb trial groups (Tables 3 and 4 and Figure 4).

TABLE 4 - Glucose, insulin and gut hormone responses during the mixed-meal tolerance test and at the point of matched 20% weight loss

Notes

Continuous data are presented as a mean (SD) when normally distributed or a median (IQR) when non-normally distributed.

Between-group differences are denoted as a mean (95% CI).

The statistical test used was ANCOVA.

All data were analysed using ANCOVA, with adjustment of between-group differences (treatment effect or odds ratio) for outcome at baseline.

The AUC was calculated from time point 0 to 120 minutes.

Outcome	Trial group	n	Time point, mean (SD)		Treatment effect (95% CI)	p-value
			Preoperatively	20% weight loss
GLP-1 active peak (pmol/l)	Standard limb	24	24 (33)	70 (32)	0	0.43
	Long limb	23	16 (14)	70 (19)	5 (–8 to 18)
GLP-1 active AUC (pmol/l/minute)	Standard limb	18	2496 (4219)	5312 (3711)	0	0.18
	Long limb	18	1292 (1497)	3812 (1327)	–529 (–1315 to 256)
GLP-1 total peak (pmol/l)	Standard limb	24	13 (6)	117 (57)	0	0.68
	Long limb	24	14 (10)	112 (38)	–6 (–35 to 23)
GLP-1 total AUC (pmol/l/minute)	Standard limb	24	1017 (394)	6281 (2586)	0	0.69
	Long limb	24	1044 (543)	6039 (2555)	–285 (–1726 to 1155)
Insulin peak (mU/l)	Standard limb	24	29 (14)	42 (20)	0	0.37
	Long limb	23	28 (16)	38 (18)	–3 (–10 to 4)
Insulin AUC (mU/l/minute)	Standard limb	24	5281 (2464)	6433 (3058)	0	0.34
	Long limb	23	5128 (2833)	5716 (2879)	–594 (–1828 to 641)
Outcome	Trial group	n	Time point, median (IQR)		Odds ratio (95% CI)	p-value
			Preoperatively	20% weight loss
PYY peak (pmol/l)	Standard limb	16	56 (31–75)	104 (85–158)	1	0.29
	Long limb	12	33 (27–69)	101 (74–120)	0.88 (0.68 to 1.13)
PYY AUC (pmol/l/minute)	Standard limb	8	8163 (5419–10,539)	14,472 (10,444–16,385)	1	0.78
	Long limb	3	5966 (2576–7536)	11,189 (5514–12,851)	0.96 (0.68 to 1.35)
GIP peak (pmol/l)	Standard limb	24	75 (39–110)	129 (61–259)	1	0.20
	Long limb	23	101 (31–171)	88 (38–181)	0.56 (0.23 to 1.37)
GIP AUC (pmol/l/minute)	Standard limb	15	5139 (1489–8472)	9185 (4006–15,486)	1	0.69
	Long limb	16	11,203 (2614–13,665)	6041 (4174–11,893)	0.87 (0.44 to 1.74)
Glucose peak (mmol/l)	Standard limb	24	15.3 (13.2–17.2)	9.3 (7.7–11.0)	1	0.32
	Long limb	24	14.4 (11.4–17.5)	8.0 (6.9–9.3)	0.93 (0.81 to 1.07)
Glucose peak (mmol/l)	Standard limb	24	2828 (2450–3172)	1564 (1276–1896)	1	0.38
	Long limb	24	2647 (2103–3221)	1301 (1170–1580)	0.94 (0.80 to 1.09)

FIGURE 4.

Plasma glucose excursion during the mixed-meal tolerance test. (a) Plasma glucose concentrations preoperatively; (b) plasma glucose concentrations 2 weeks postoperatively; and (c) plasma glucose concentrations at the 20% weight loss time point. Data were plotted as means (SDs). n = 24 in each trial group. Reprinted with permission from The American Diabetes Association. Copyright 2020 by the American Diabetes Association. 4

Reprinted with permission from The American Diabetes Association. Copyright 2020 by the American Diabetes Association. 4

Insulin secretion

At the 2-week time point and at the point of matched 20% weight loss, peak postprandial insulin concentrations during the mixed-meal tolerance test were significantly increased compared with baseline in both trial groups. However, there were no significant differences between the standard and long limb trial groups (see Tables 3 and 4; Figure 5).

FIGURE 5.

Serum insulin excursion during the mixed-meal tolerance test. (a) Serum insulin concentrations preoperatively; (b) serum insulin concentrations 2 weeks postoperatively; and (c) serum insulin concentrations at the 20% weight loss time point. Data were plotted as means (SDs). n = 24 in each trial group. Reprinted with permission from The American Diabetes Association. Copyright 2020 by the American Diabetes Association. 4

Reprinted with permission from The American Diabetes Association. Copyright 2020 by the American Diabetes Association. 4

Insulin sensitivity

At the 2-week time point and at the point of matched 20% weight loss, the rate of glucose appearance during the low-dose phase of the hyperinsulinaemic–euglycaemic clamp (i.e. Ra), which is a measure of hepatic insulin sensitivity, had decreased significantly compared with baseline in both trial groups. This decrease of Ra denotes an improvement in insulin sensitivity. However, there were no significant differences between the standard and long limb trial groups (Tables 5 and 6 and Figure 6).

At the 2-week time point and at the point of matched 20% weight loss, the rate of glucose disappearance during the high-dose phase of the hyperinsulinaemic–euglycaemic clamp (i.e. Rd), which is a measure of peripheral insulin sensitivity, had increased significantly compared with baseline in both trial groups. This increase in Rd denotes an improvement in insulin sensitivity. However, there were no significant differences between the standard and long limb trial groups (Tables 5 and 6 and Figure 6).

TABLE 5 - Rate of glucose appearance (Ra), disappearance (Rd) and metabolic clearance rate of glucose (MCR) in the hyperinsulinaemic–euglycaemic clamps 2 weeks postoperatively

High, measured in the phase of high-dose insulin infusion; Low, measured in the phase of low-dose insulin infusion; MCR, metabolic clearance rate of glucose.

Notes

Continuous data are presented as means (SDs) when normally distributed.

Between-group differences are denoted as a mean (95% CI).

All data were analysed using ANCOVA, with adjustment of between-group differences (treatment effect) for outcome at baseline.

Outcome	Trial group	n	Time point, mean (SD)		Treatment effect (95% CI)	p-value
			Preoperatively	2 weeks postoperatively
Ra
Basal	Standard limb	23	11.1 (2.0)	9.6 (1.0)	0	0.23
	Long limb	23	10.9 (1.5)	10.0 (1.5)	0.4 (–0.3 to 1.1)
Low	Standard limb	23	5.1 (1.7)	3.4 (0.9)	0	0.94
	Long limb	23	5.0 (2.4)	3.4 (1.4)	0.0 (–0.6 to 0.7)
High	Standard limb	23	1.9 (2.4)	0.6 (1.7)	0	0.86
	Long limb	23	1.1 (2.0)	0.7 (1.7)	0.1 (–0.9 to 1.1)
Rd
Basal	Standard limb	23	11.2 (2.0)	9.7 (1.0)	0	0.23
	Long limb	23	10.9 (1.5)	10.1 (1.5)	0.4 (–0.3 to 1.2)
Low	Standard limb	23	9.8 (1.6)	11.8 (2.9)	0	0.18
	Long limb	23	10.6 (3.6)	13.6 (4.2)	1.3 (–0.6 to 3.2)
High	Standard limb	23	18.7 (7.6)	29.0 (9.1)	0	0.98
	Long limb	23	19.1 (9.4)	29.2 (9.9)	0.1 (–5.2 to 5.3)
MCR
Basal	Standard limb	23	1.9 (0.3)	1.8 (0.2)	0	0.05
	Long limb	23	1.9 (0.3)	1.9 (0.3)	0.1 (0.0 to 0.3)
Low	Standard limb	23	1.7 (0.3)	2.1 (0.6)	0	0.23
	Long limb	23	1.9 (0.7)	2.4 (0.8)	0.2 (–0.1 to 0.6)
High	Standard limb	23	3.2 (1.3)	5.3 (1.8)	0	0.89
	Long limb	22	3.6 (2.2)	5.6 (2.0)	0.1 (–1.0 to 1.1)

TABLE 6 - Rate of glucose appearance (Ra), disappearance (Rd) and metabolic clearance rate of glucose (MCR) in the hyperinsulinaemic–euglycaemic clamps at a matched 20% weight loss

High, measured in the phase of high-dose insulin infusion; Low, measured in the phase of low-dose insulin infusion; MCR, metabolic clearance rate of glucose.

Notes

Continuous data are presented as means (SDs) when normally distributed.

Between-group differences are denoted as a mean (95% CI).

All data were analysed using ANCOVA, with adjustment of between-group differences (treatment effect) for outcome at baseline.

Outcome	Trial group	n	Time point, mean (SD)		Treatment effect (95% CI)	p-value
			Preoperatively	20% weight loss
Ra
Basal	Standard limb	24	11.1 (2.0)	10.6 (1.4)	0	0.28
	Long limb	23	10.9 (1.5)	11.0 (2.2)	0.6 (–0.4 to 1.5)
Low	Standard limb	24	5.2 (1.7)	2.8 (1.3)	0	0.62
	Long limb	23	5.0 (2.4)	2.6 (1.7)	–0.2 (–1.0 to 0.6)
High	Standard limb	24	2.0 (2.4)	0.0 (1.8)	0	0.09
	Long limb	23	1.2 (2.0)	–1.0 (1.3)	–0.8 (–1.8 to 0.1)
Rd
Basal	Standard limb	24	11.2 (2.0)	10.7 (1.4)	0	0.28
	Long limb	23	10.9 (1.5)	11.1 (2.2)	0.5 (–0.4 to 1.5)
Low	Standard limb	24	9.8 (1.6)	15.2 (2.9)	0	0.36
	Long limb	23	10.6 (3.6)	16.5 (4.1)	0.9 (–1.1 to 2.9)
High	Standard limb	24	18.5 (7.6)	36.1 (8.5)	0	0.47
	Long limb	23	19.1 (9.4)	38.1 (9.2)	1.8 (–3.2 to 6.9)
MCR
Basal	Standard limb	24	1.9 (0.3)	2.1 (0.2)	0	0.17
	Long limb	23	1.9 (0.3)	2.2 (0.5)	0.1 (–0.1 to 0.3)
Low	Standard limb	24	1.7 (0.3)	2.7 (0.4)	0	0.10
	Long limb	23	1.9 (0.7)	3.1 (0.8)	0.3 (–0.1 to 0.7)
High	Standard limb	24	3.2 (1.2)	6.7 (1.7)	0	0.57
	Long limb	23	3.6 (2.1)	7.1 (1.7)	0.3 (–0.7 to 1.2)

FIGURE 6.

(a) Rate of glucose appearance (Ra) at low-dose insulin infusion; and (b) rate of glucose disappearance (Rd) at high-dose insulin infusion. Data are plotted as means (SDs). n = 23 in each trial group. Reprinted with permission from The American Diabetes Association. Copyright 2020 by the American Diabetes Association. 4

Reprinted with permission from The American Diabetes Association. Copyright 2020 by the American Diabetes Association. 4

Glycaemic control and weight loss

There were no significant differences in the levels of HbA_1c between the standard limb and long limb trial groups at any time point postoperatively. At 3 months postoperatively, the level of HbA_1c in the standard limb group was 50 mmol/mol (SD 10 mmol/mol), compared with 48 mmol/mol (SD 10 mmol/mol) in the long limb group (p = 0.40). At 6 months postoperatively, the level of HbA_1c in the standard limb trial group was 44 mmol/mol (SD 8 mmol/mol), compared with 41 mmol/mol (SD 6 mmol/mol) in the long limb trial group (p = 0.07). At 12 months postoperatively, the level of HbA_1c in the standard limb trial group was 43 mmol/mol (SD 10 mmol/mol), compared with 41 mmol/mol (SD 5 mmol/mol) in the long limb trial group (p = 0.20) (see Table 2 and Figure 7). There were no significant differences in the percentage of patients achieving glycaemic remission at 12 months between the trial groups (standard limb 62% vs. long limb 77%; p = 0.23).

FIGURE 7.

Level of (a) glycated haemoglobin and (b) total body weight loss within the first postoperative year. Data are expressed as means (SDs). n = 26 in each trial group. Asterisks in aqua and navy compare the values with baseline within each group. ***p < 0.001.

The usage of glucose-lowering medications decreased in both trial groups (see Appendix 3). At baseline, 100% of patients were on pharmacotherapy. At 3 months postoperatively, 38% of standard limb and 24% of long limb trial group participants required medications. At 6 months the corresponding figures were 28% and 24%. Only one participant from the long limb trial group was on glucose-lowering pharmacotherapy at 12 months, with none in the standard limb group.

At 2 weeks postoperatively, patients in both trial groups lost similar percentages of body weight [standard limb 6.2% (SD 2.3%) vs. long limb 6.1% (SD 1.6%); p = 0.97]. As per protocol, both trial groups were studied again at the point of matched 20% weight loss; this occurred, on average, 4.5 months after surgery. At this time point, mean percentage weight loss was 21.5% (SD 2.8%) in the standard limb group and 20.6% (SD 2.7%) in the long limb trial group. There were no differences in total body weight loss percentage between the trial groups at any time point postoperatively. At 3 months postoperatively, mean weight loss was 19% (SD 4%) in both trial groups (p = 0.99). At 6 months postoperatively, mean weight loss was 26% (SD 6%) in the standard limb group and 24% (SD 4%) in the long limb trial group (p = 0.11). At 12 months postoperatively, the corresponding figures were 30% (SD 8%) and 29% (SD 8%) (p = 0.52) (see Figure 7).

Recommendations

It is hoped that the trial design and findings lay the foundation for a new generation of experimental medicine studies that aim to optimise the clinical efficacy of metabolic surgery, and indeed non-surgical interventions, through interrogation of the elusive physiology of the intestine and the impact of its various segments on metabolic regulation.

Based on the findings of this trial, an application for a follow-up extension to 5 years has been submitted and successfully secured. It will include a mixed-meal tolerance test 2 years after surgery, followed by yearly clinical follow-ups. This will enable the measurement of any differences in the long-term impact of the long limb compared with the standard limb RYGB.

Should further trials on the biliopancreatic limb length be planned, this study recommends consideration of a longer biliopancreatic limb length and using limb lengths in proportion to the total small bowel length instead of the absolute values. Furthermore, investigating intestinal remodelling and its impact on the postoperative outcomes could be considered by taking intestinal biopsies intraoperatively and then endoscopically at least 12 months after the surgery.

However, it is possible that no further improvement in glucose homeostasis can be achieved with manipulation of the biliopancreatic limb length and, therefore, the research should be directed towards investigating the common limb.

Acknowledgement of other investigators not in list of authors

Dr Julian Marchesi and Professor Elaine Holmes are co-investigators. They are currently collaborating on further analysis, which includes the gut microbiota data.

Dr Paul Bassett is the trial statistician and is independent of the study.

Dr Victoria Salemis is an independent researcher from Imperial College London and is responsible for the randomisation of trial patients.

Contributions of authors

Alexander Dimitri Miras (https://orcid.org/0000-0003-3830-3173) (Co-investigator; Senior Clinical Lecturer in Endocrinology, Imperial College London) contributed to the study concept, design and conduct, data analysis and manuscript writing.

Anna Kamocka (https://orcid.org/0000-0002-6242-0639) (Clinical Research Fellow and Specialty Registrar in Bariatric Surgery, Imperial College London) contributed to the conduct of the study, data collection and analysis, and manuscript writing.

Tricia Tan (https://orcid.org/0000-0001-5873-3432) (Co-investigator; Professor of Metabolic Medicine and Endocrinology, Imperial College London) contributed to the study concept and design, and data analysis and manuscript writing.

Belén Pérez-Pevida (https://orcid.org/0000-0001-8632-488X) (Clinical Research Fellow, Imperial College London) contributed to the conduct of the study, data collection and analysis, and manuscript editing.

Harvinder Chahal (Co-investigator; Consultant Endocrinologist, Imperial College London) contributed to the study design and manuscript editing.

Krishna Moorthy (Co-investigator; Senior Clinical Lecturer in Upper Gastrointestinal and Bariatric Surgery, Imperial College London) contributed to the study design, was an operating surgeon and was involved with manuscript editing.

Sanjay Purkayastha (Co-investigator; Senior Clinical Lecturer in Bariatric Surgery, Imperial College London) contributed to the study design, was an operating surgeon and was involved with manuscript editing.

Ameet Patel (https://orcid.org/0000-0002-8637-0963) (Co-investigator; Professor of Surgery, King’s College London) was an operating surgeon and contributed to manuscript editing.

Anne Margot Umpleby (https://orcid.org/0000-0001-6147-7919) (Co-investigator; Professor of Diabetes and Metabolic Medicine, University of Surrey) contributed to the study design, data analysis and manuscript editing.

Gary Frost (https://orcid.org/0000-0003-0529-6325) (Co-investigator, Imperial College London) contributed to the study concept and design.

Stephen Robert Bloom (https://orcid.org/0000-0003-1542-2348) (Principal Investigator; Professor of Diabetes, Endocrinology and Metabolism, Imperial College London) contributed to the study concept and design and manuscript editing.

Ahmed Rashid Ahmed (Co-investigator; Senior Clinical Lecturer in Bariatric Surgery, Imperial College London) contributed to the study concept and design, was an operating surgeon and was involved in manuscript writing.

Francesco Rubino (https://orcid.org/0000-0001-8581-2515) (Co-investigator; Professor of Metabolic and Bariatric Surgery, King’s College London) contributed to the study concept and design, was an operating surgeon and was involved in manuscript writing.

Publication

Miras AD, Kamocka A, Pérez-Pevida B, Purkayastha S, Moorthy K, Patel A, et al. The effect of standard versus longer intestinal bypass on GLP-1 regulation and glucose metabolism in patients with type 2 diabetes undergoing Roux-en-Y gastric bypass: The Long-Limb Study. Diabetes Care 2020;43:dc20076.

Data-sharing statement

Data are archived at the National Institute for Health Research Imperial Clinical Research Facility. All data requests should be submitted to the corresponding author for consideration. Access to anonymised data may be granted following review.

Patient data

This work uses data provided by patients and collected by the NHS as part of their care and support. Using patient data is vital to improve health and care for everyone. There is huge potential to make better use of information from people’s patient records, to understand more about disease, develop new treatments, monitor safety, and plan NHS services. Patient data should be kept safe and secure, to protect everyone’s privacy, and it’s important that there are safeguards to make sure that it is stored and used responsibly. Everyone should be able to find out about how patient data are used. #datasaveslives You can find out more about the background to this citation here: https://understandingpatientdata.org.uk/data-citation.

Disclaimers

This report presents independent research. The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, the MRC, NETSCC, the EME programme or the Department of Health and Social Care. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the EME programme or the Department of Health and Social Care.