The rapid effects of sleeve gastrectomy on glucose homeostasis and resolution of diabetes mellitus

Abstract Aims Type 2 diabetes caused by obesity is increasing globally. Bariatric surgical procedures are known to have positive effects on glucose homeostasis through neurohormonal action mechanisms. In the present study, we aimed to investigate the factors influencing glucose homeostasis independent of weight loss after the laparoscopic sleeve gastrectomy (LSG). Methods Patients who underwent LSG for morbid obesity in a 3‐year period were evaluated. Data on demographics, clinical characteristics (duration of diabetes, resected gastric volume, antral resection margin) and laboratory parameters (preoperative and postoperative blood glucose on fasting, preoperative HbA1c levels and first‐year HbA1c levels) were retrospectively reviewed. Effect of patients' body mass index (<50 kg/m2, ≥50 kg/m2), first‐year excess weight loss (EWL%) rates, age (≥50 years, <50 years), duration of diabetes (≥5 years, <5 years) and antral resection margin (≥3 cm, <3 cm) on postoperative blood glucose profile and diabetic resolution status were investigated. Results Total of 61 patients constituted the study group. There were 40 female and 21 male patients with an average age of 43.8 ± 10.5 years (19‐67 years). Preoperatively, mean BMI, blood glucose levels and HbA1c were 48.8 ± 8.5 kg/m2, 133.6 ± 47.4 mg/dL and 7.4 ± 1.1, respectively. The mean blood glucose level at the postoperatively 5th day was 88.0 ± 16.3 mg/dL (median: 84 mg/dL) (P < .001). Fifty‐nine out of 61 patients improved their glycaemic control. Conclusions It is noteworthy that LSG can control blood glucose levels in short term after surgery regardless of weight loss. Therefore, LSG should be preferred at earlier stages in the treatment of obesity‐related T2DM in order to prevent T2DM‐related complications.


| INTRODUC TI ON
Obesity-related type 2 diabetes mellitus (T2DM) is a global ever-growing health problem, a significant burden on healthcare systems and healthcare costs. 1 Obesity is an important factor in the onset of T2DM. Therefore, weight loss has a therapeutic importance for success in T2DM treatment. In these patients, lifestyle changes and medications are the main factors for controlling the diabetes.
Problems in compliance with long-term drug use and maintaining weight loss limit the role of these treatment methods. 2 The development of minimally invasive laparoscopic techniques has provided immense progress in bariatric surgery, and many randomized controlled trials have shown that bariatric surgery reduces cardiovascular risk factors along with successful glycaemic control. 3,4 Operative results are remarkable in patients with longterm uncontrolled T2DM. 5 Similarly, a randomized controlled study conducted by Schauer et al 5 demonstrated that bariatric surgery in combination with medical therapy is more effective than medical therapy alone for improving T2DM outcomes for patients with body mass index (BMI) from 27 to 43 kg/m 2 .
Out of a host of bariatric operations for morbid obesity, laparoscopic sleeve gastrectomy (LSG) has been the most frequent surgical procedure worldwide. LSG involves removing almost all of the fundus and creating a tube-shaped stomach with a capacity of 60-100 mL, which in turn limits the capacity for food intake. LSG was originally defined as a first step of biliopancreatic diversion with duodenal switch (BPDDS), but today it is preferred as a well-tolerated stand-alone bariatric surgical procedure. 6,7 We noticed that in patients with obesity-related diabetes, blood glucose levels decreased in the first days after LSG. Therefore, we performed a retrospective analysis documenting the improved blood glucose levels after LSG, which was observed before losing significant weight in 61 morbidly obese patients with T2DM. The main focus of our study is to reveal the effect of LSG on glucose homeostasis in the very early postoperative period and to investigate the factors that may cause this effect.

| Study population and design
Sixty-one morbidly obese patients with a diagnosis of T2DM were included in the study. The patients were included in this study if the following criteria were met: (a) BMI > 40 kg/m 2 , (b) uncontrolled T2DM after 6 months of medical therapy (HbA1c > 6.5%), and (c) the absence of contraindications for bariatric surgery. Exclusion criteria were as follows: (a) history of duodenal or proximal jejunal intervention, (b) patients with malabsorptive syndrome, and (c) patients who did not attend recall visit or lost to follow up. Data related to patients' demographics (age, sex, BMI), laboratory results (blood glucose level and HbA1C level) and duration of T2DM were recorded.
HbA1c level below 6.5% without the need of antidiabetic treatment for 1 year or blood glucose level of fasting below 100 mg/dL was considered as diabetes resolution. 8 The parameters used for effectiveness of LSG and weight loss were BMI and excess weight loss ratio (EWL%).

| Statistical analysis
SPSS Statistics ver. 15.0 (Chicago, SPSS Inc) was used for statistical analysis. While data for continuous variables were reported as mean ± standard deviation (SD), median, minimum and maximum, categorical variables were presented as number and percentages.
Friedman test was used for comparison of numerical variables in dependent groups. Wilcoxon test with Bonferroni correction was performed for subgroup analyses. Chi-square test was performed for comparison of independent group. A statistical P value of .05 or lesser was considered significant.

| Glycaemic control
The mean blood glucose level on fasting was 88.0 ± 16.3 mg/dL (median: 84 mg/dL) on the 5th postoperative day (P < .001) (Tables 2  and 3). The postoperative first-year blood glucose levels of the patients in the study group are shown in Table 2. Postoperative HbA1c level at the 1st year was 5.7 ± 0.8% (median: 5.6%), and comparison of preoperative and postoperative 1st-year HbA1c was shown in

| Diabetes medications
The mean T2DM duration of the patients was 5 ± 2.6 years (range: 1-14 years). There was no statistically significant difference between the groups with diabetes duration ≥5 years and <5 years in terms of T2DM resolution (p:0.238). Fifty-nine patients ceased all medication for diabetes in the 1 year follow-up period. None needed increased dosage of antidiabetic therapy. Two patients, whose T2DM did not improve, continued using insulin in the postoperative period, albeit a lower dose of insulin. T2DM was improved in 59 out of 61 patients (Table 1).

| Weight loss
Although the mean EWL% at the end of the first month was only 28 ± 8.4%, the rate of decrease in blood glucose was 20% (mean ± SD: 104.4 ± 38.6 mg/dL) in the first month and the blood glucose level reached normal limits in 48 (78.6%) patients during the first month (P < .001). At 1st year of follow-up, the mean ± SD EWL% was 96.6 ± 18.2%, and blood glucose levels decreased by 35% and HbA1c decreased by 23% during this period ( Table 2).
Although preoperative HbA1c levels were higher in other 2 patients (3.3%) who had no improvement in T2DM, preoperative HbA1c levels and early postoperative improvement were not  Po day 3 -Pr <.001
F I G U R E 1 HbA1c levels at preoperative period and postoperative 1st year TA B L E 4 Effect of demographics and surgical findings on resolution of T2DM for potential effects of bariatric surgery on T2DM resolution is also controversial. 18 In this study, the rate of weight loss did not correlate with re-  22 In the late postoperative period after bariatric surgery, the most effective factor on glucose homeostasis is the breaking down of insulin resistance related with weight loss, which lead beta cells to rest and recover.
The first metabolic event seen after LSG is the reduction in the plasma concentration of ghrelin due to the removal of the ghrelin-secreting cells in the fundus. Increasing resected gastric volume (RcGV) suggests that the more ghrelin-secreting cell is resected and the more plasma ghrelin level will decrease. Therefore, a larger RcGV is thought to reduce peripheral insulin resistance and have a regulatory effect on insulin release. Although some authors stated that RcGV > 1200 mL was more effective on T2DM resolution, 23 in our study, no statistically significant difference was detected between groups with RcGV > 1500 mL and ≤1500 mL in terms of blood glucose homeostasis. The lowest RcGV in our study was 1150 mL.
Similar to our study, Singh et al 24 reported a T2DM resolution rate of 82.9%, regardless of the amount of RcGV with a cut-off of 1700 mL in their prospective study.
Laparoscopic sleeve gastrectomy is a purely restrictive operation with no malabsorptive effect. Improvement in blood glucose levels after malabsorptive procedures occurs within days without significant weight loss. 17 Although this is not observed in restrictive procedures, review of the literature has shown that LSG has a comparable T2DM amelioration rate in the early postoperative period at rates similar to RYGB. 15,16,25 While endocrinologists recommend nutritional regulation and strict diet at the initial stage of the disease to prevent catastrophic consequences that can be observed if T2DM is not controlled, surgical intervention performed at early stages of T2DM is likely to be more effective. Since LSG has positive effects on T2DM unrelated with weight loss in the early period, it should be considered as a treatment option in order to prevent T2DM-related microvascular and macrovascular complications in obese patients in the early period of the disease. The key positive effects of LSG on T2DM independent of weight loss has led to the development of interventional T2DM treatments that do not cause weight loss. In addition, at later stages, LSG also has a positive effect on the recovery of pancreatic beta cell functions with its weight loss effect.
Although fulfilling the sample-size requirement, a relatively small number of patients were the main limitation of this study. Despite this, we conclude that LSG should be considered as a potential primary treatment option in patients with diabetes.

ACK N OWLED G EM ENT
I would like to thank Dr Hassan Ahmed Elhassan for language editing assistance.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

AUTH O R S CO NTR I B UTI O N
EB, CK, SÖ, OG and MM conceived and designed the study; involved in data collection and processing; analysed and interpreted the data; and wrote the manuscript. EB and CK performed literature search.
EB, CK and MM critically reviewed the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
There are no new data associated with this manuscript.