The Remission of Type 2 Diabetes Mellitus in Chinese Patients After Metabolic Surgery and Its Preoperative Contributing Factors: A Cohort Study

Background and Objective: Existing data about the contributing factors of operative effect for type 2 diabetes mellitus (T2DM) are still limited in Chinese patients, especially about preoperative blood routine and biochemical indexes. We evaluated a prospective cohort of T2DM patients in early and middle stage with obesity to assess the post-operative prognosis and investigate its contributing factors. Methods: Adult T2DM participants in early and middle stage with obesity were enrolled and received metabolic surgery. Clinical data such as age, sex, baseline body mass index, percentage excess weight loss, glycemia range and drug consumption were collected and analyzed. Patients were managed by a multi-disciplinary team and followed up for 12 months. Complete remission was dened as HbA1c < 6.0%, fasting glucose < 5.6 mmol/l, without pharmacological intervention for at least 3 months. Results: In this study, a total of 96 T2DM patients with metabolic surgery were included. Among them, 61 (63.54%) patients had complete remission and 85 (88.50%) had post-operative partial remission after a 12-month clinical follow-up. Only 1 patient was reported to have an anastomotic leak; no surgical mortality during the follow-up. According to the complete remission or not, the patients were divided into two groups. There were signicant differences on stroke history, fasting blood sugar and white blood cell between them. Furthermore, in multivariable analyses models, preoperative triglyceride (adjusted OR, 0.585; 95%CI, 0.418-0.819; p<0.01) and preoperative lymphocyte count (adjusted OR, 2.647; 95%CI, 1.141-6.142; p < 0.05) was signicantly associated with complete remission of T2DM. Conclusion: Our data suggest that T2DM patients with lower triglyceride and higher lymphocyte count tended to achieve post-operative remission. Preoperative evaluation of lipid metabolism and immune function may be helpful for the evaluation of prognosis of metabolic surgery. t-test, Mann-Whitney U test or Chi-square test was used to assessed differences among variables. Pearson’s correlation coecients were calculated to assess the relationship between the variables. Logistic regression analysis was used to nd contributing factors which associated with remission of T2DM in patients receiving metabolic surgery. Models were built for groups of confounding factors: Model 1 was adjusted for demographics; Model 2 was adjusted for model 1 covariates and medical history of stroke and hypertension; and Model 3 additionally for baseline blood routine, biochemical indexes and BMI. Considering the correlation among smoking status, alcohol consumption and gender in China, we did not bring them into the adjustment factors at the same time. The level of signicance for these descriptive comparisons was established at 0.05 for two-sided hypothesis testing. Statistical analysis was performed in SPSS 25.0. in this study were recruited only from one clinical unit, there may have retrospective bias inherent due to the insucient sample size. We used denitions of post-operative complete and partial remission similar to ADA consensus statement, but not identical(31). Considering that the general lack of regular physical examination in Chinese patients may lead to the undetected T2DM, we did not include the duration of DM into variables. The study is observational in nature. We can’t prove the causal relationship among


Introduction
As a traditionally intractable chronic medical condition, the medical management of type 2 diabetes mellitus (T2DM) typically consists of lifestyle modi cations and speci c glucose-lowering medications (1,2). Most patients bene t from these conservative approaches in a short term, however, persistent clinical remission and its associated metabolic improvement is rarely achieved (2,3). At the same time, the pharmacotherapy of T2DM is often oriented towards managing only hyperglycemia, rather than the disease's numerous metabolic disorders (4).
The pathogenesis of insulin resistance and T2DM is closely related to overall and visceral adiposity (5,6). The remarkable effects of metabolic surgery regarding metabolic amelioration and therapeutic modality for T2DM have gradually gathered attention (7). In a large sample study, a considerable number of patients (75%) achieved remission of T2DM after two years' surgery(8). Furthermore, the remission and weight loss were accompanied by a regression of microalbuminuria and a lower incidence of vascular complications (9). However, existing data about the contributing factors of operative effect, especially about preoperative blood routine and biochemical indexes, are still limited.
In this cohort study, the demographic and clinical characteristics of T2DM patients with obesity receiving metabolic surgery were collected. Meanwhile, the correlations between preoperative indicators and prognosis were examined in a 12-month follow-up.

Study Population
We conducted a prospective single-cohort study of T2DM individuals who received metabolic surgery in the First A liated Hospital of Soochow University between January 2013 and July 2018. T2DM was de ned as 1) using of diabetes medication; 2) fasting whole blood glucose ≥ 7.0 mmol/L and/or random whole blood glucose ≥ 11.1 mmol/L; 3) patients according to national or local patient registers with diabetes. Diagnosis and classi cation of T2DM was based on the criteria established by the American Diabetes Association (10). Diabetic patients were managed by a multi-disciplinary team included endocrinologist and general surgeon. The exclusion criteria were as follows: 1) patients were younger than 18 years old or older than 65 years old; 2) patients with a body mass index (BMI) ≤ 32 kg/m 2 ; 3)patients withhistory of previous metabolic surgery; 4) patients with type 1 diabetes or poor beta-cell function (diagnosed anti-GAD or islet-cell auto-antibodies, insulin use for more than ten years, fasting C-peptide < 1 ng/ml, or unresponsive to a stimulus test); 5) patients with end organ damage; 6) patients with pregnancy; 7) patients had a history of infectionwithin 2 weeks before admission that was de ned as fever(T ≥ 38•C) and at least one other typical symptoms (cough,rhinitis, hoarseness, sneezing, or vomiting); 8) patients had a history of alcoholic cirrhosis, hematological diseases, autoimmune diseases, or treatment with immunosuppressive agents.
All participants accepted nutritional, psychological and endocrinological assessments. Diabetic patients with obesity were eligible for the study if they had a BMI>32 kg/m 2 and/or poorly controlled T2DM after 6 months of nutrition intervention and hypoglycemic treatment. All post-operative patients received the same medical advice and treatment, including reasonable diet, effective blood pressure control and inactive hypoglycemic therapy(monotherapy or off medication).

Surgical Procedures
Surgical procedures included the following two types of metabolic procedures: (1) laparoscopic Roux-en-Y gastric bypass (LRYGB); (2) laparoscopic sleeve gastrectomy (LSG). Procedure would be selected after a comprehensive pre-operative conference with the multi-disciplinary team. All procedures were performed by the same experienced surgical team. Brie y, we used a standard 5-port laparoscopic technique. The LRYGB operation involves an antecolic, antegastric Roux limb, a 100-cm bilio-pancreatic limb, and a 100-cm alimentary limb. The gastric pouch was approximately 30 ml, and the gastrojejunostomy was created by a stapler technique with an anastomosis 1.0-1.5 cm in diameter. LSG was performed by creating a sleeve gastrectomy over a 36Fr bougie and leaving a 4-6 cm long antrum. Thromboembolic prophylaxis consisted of perioperative pneumatic compression and low-molecular weight heparin (4000AxaIU) during anesthetic induction.

Clinical Information Collection
Demographics were collected through electronic patient records and administrative databases. Patients body weight was measured in light clothing without shoes to the nearest 0.1 kg, and body height was measured to the nearest 0.1 cm. BMI was calculated as weight in kilograms divided by height in meters squared. Complete blood cell counts and blood biochemical parameters, including the individual components of glycemic control (levels of serum glucose, HbA1c levels), were assessed preoperatively. Peripheral venous blood samples were collected on the morning of the second day after admission with an overnight fasting.

Outcome Variables
The primary outcome was the complete remission of T2DM at post-operative 12 months. Other outcomes were partial remission of T2DM at post-operative 12 months, percentage excess weight loss (%EWL) at post-operative 6 months (good outcome: >50%), and %EWL at post-operative 12 months (good outcome: >50%). Complete remission was de ned as achieving glycemia below the diabetic range (HbA1c in the normal range [< 6.0%], fasting glucose < 5.6 mmol/l) without pharmacological interventionfor at least 3 months. Partial remission was de ned as sub-diabetic hyperglycemia (HbA1C not diagnostic of diabetes [< 6.5%], fasting glucose 100-125 mg/dl [5.6-6.9 mmol/l]) without pharmacological interventionfor at least 3 months. Diabetic patients were managed by a multidisciplinary team, but each patient was followed by the case manager. First clinical evaluation was performed 1 week after discharge. Then clinical and laboratory evaluation (fasting glucose, glycosylated hemoglobin and blood count) were performed at post-operative 3, 6, 9, and 12 months.

Data Analyses
Continuous variables were analyzed as mean and standard deviation or the median and interquartile range while categorical variables were analyzed as frequency and percentage, properly. Student's t-test, Mann-Whitney U test or Chi-square test was used to assessed differences among variables. Pearson's correlation coe cients were calculated to assess the relationship between the variables. Logistic regression analysis was used to nd contributing factors which associated with remission of T2DM in patients receiving metabolic surgery. Models were built for groups of confounding factors: Model 1 was adjusted for demographics; Model 2 was adjusted for model 1 covariates and medical history of stroke and hypertension; and Model 3 additionally for baseline blood routine, biochemical indexes and BMI. Considering the correlation among smoking status, alcohol consumption and gender in China, we did not bring them into the adjustment factors at the same time. The level of signi cance for these descriptive comparisons was established at 0.05 for two-sided hypothesis testing. Statistical analysis was performed in SPSS 25.0.

Participants and Descriptive Characteristics
The initial sample included 429 patients with diabetes. Only 104 T2DM patients with obesity were recruited. A 12-month clinical follow-up was performed.
Eight patients were lost (8/104, 7.7%) and 96 participants (60 LRYGB and 36 LSG) nally formed the basis of this report. Only 1 patient was reported to have an anastomotic leak; no surgical mortality during the follow-up. As the number of participants with severe postoperative complications and death was very small, no useful separate analysis could be made. Patient selection is illustrated in Fig. 1

Comparison Of Clinical Characteristics In Patients Receiving Metabolic Surgery
According to the remission of T2DM or not, these participants were divided into two groups: the complete post-operative remission group with 61 patients and the non-complete post-operative remission group with 35 patients. Statistical analysis indicated that there were signi cant differences on stroke history, baseline WBC and FBS (P < 0.05); However, there was no difference on age, gender, smoking, drinking, hypertension, hyperlipidemia, surgical procedure, baseline TC, TG, BMI or other factors between two groups (P > 0.05, Table 2). Furthermore, we calculated the correlation coe cients between age, gender, smoking, drinking, hypertension, coronary disease, stroke history, BMI, blood biochemical criterions, complete remission and partial remission at post-operative 12 months, %EWL at post-operative 6 months, and %EWL at post-operative 12 months. Observed correlation coe cients: r = 0.405 between preoperative BMI and WBC (p < 0.01); r = 0.220 between preoperative TG and partial postoperative remission (p < 0.05); r = − 0.251 between stroke history and complete post-operative remission (p < 0.05); r = 0.286 between %EWL at post-operative 12 months and complete post-operative remission. Post-operative remission had weak but signi cant correlations with stroke history, preoperative TG, and %EWL at post-operative 12 months (Table 3).   (Table 4).

Discussion
In our study, the operative effect in T2DM patients with obesity as well as preoperative blood biochemical indexes were analyzed. Our results showed that patients with lower preoperative TG levels were more likely to have post-operative remission of T2DM. In addition, preoperative lymphocyte count was also associated with complete post-operative remission of T2DM and %EWL in short term after operation. To our knowledge, this study was the rst time to analyze the relationship between the preoperative blood biochemical indexes and post-operative remission of T2DM in metabolic surgery.
The underlying mechanism for diabetes remission after metabolic surgery is intriguing and weight loss is the most essential part of the treatment, even in nonobese patients (11,12). Abnormal lipid metabolism is a chronic and progressive disease. Previous studies found a bimodal adipocyte distribution in morbidly obese individuals: hypertrophy of the larger adipocyte population and higher proportion of very small adipocytes (13). Large adipocyte size may reduce adipose tissue acyl-CoA synthetase and diacylglycerol acyltransferase activities, suggesting lower capacity for fatty acid storage (14). Meanwhile, hypertrophic remodeling of white adipose tissues is associated with adipose tissue dysfunction and lean organs, especially heart and liver, overexposure to circulating triglycerides (TG), ultimately leading to insulin resistance and metabolic diseases (13,15,16). Hepatic TG accumulation and hepatic insulin resistance may play an important role in impaired inhibition of gluconeogenesis (17). Previous study had shown that metabolic surgery improves fatty liver and hepatic insulin sensitivity, and reduces expression of pancreatic markers associated with diabetes(18).Moreover, glycerol uxes remained unchanged one week after RYGB, but a signi cant decrease 1 year after surgery (19,20). Although we could not nd any evidence to show the effect metabolic surgery on lipid tolerance, these results underscore the importance of impaired glycerol metabolic regulation for fasting glucose levels.
In the present study, we observed lower preoperative TG levels in complete remission group. Moreover, preoperative TG was independently associated with post-operative remission of T2DM after excluding effects of confounding. However, it is di cult to determine the causal relationship between preoperative TG, abnormal lipid metabolism and T2DM on the available evidence. We can only preliminarily infer that low preoperative TG may indicate mild lipid metabolism abnormality or short duration, and it may predict the post-operative remission of T2DM, independently of traditional risk factors.
Obesity is a condition that is associated with low grade in ammation due to hypertrophy and hyperplasia of adipose tissues (21,22). Immune cells are not only the key players in inducing low grade chronic in ammation in obesity and also are main factor responsible for pathogenesis of insulin resistance resulting Type 2 diabetes (22). Our data suggested that higher preoperative lymphocyte count was related with post-operative complete remission and shortterm %EWL. Recently studies suggested that lymphocytes were involved into the pathogenesis of obesity that associated insulin resistance (23)(24)(25).
Meanwhile, cytokines secretion from lymphocytes were involved directly in recruitment and phenotypic switch of other immune cells (26,27). The biomarkers of in ammation trigger in ammatory pathways in liver cells resulting in insulin insensitivity(28).There is a signi cant change in number of some lymphocyte subsets like NKT cells, Th1 cells and Th2 cells during obesity (29). Higher preoperative lymphocyte count may indicate a stronger immune response in obesity(28-30).However, the precise roles of lymphocyte subsets in obesity are still unclear and needed to be further investigated.
Our data should be interpreted with some caution due to limitations of the study. Since the participants in this study were recruited only from one clinical unit, there may have retrospective bias inherent due to the insu cient sample size. We used de nitions of post-operative complete and partial remission similar to ADA consensus statement, but not identical (31). Considering that the general lack of regular physical examination in Chinese patients may lead to the undetected T2DM, we did not include the duration of DM into variables. The study is observational in nature. We can't prove the causal relationship among preoperative TG, lymphocyte count and post-operative remission of T2DM. Moreover, we did not assess lipid metabolism at follow-up time points. The exact clinical relevance needs to be further studied.

Conclusion
Our data indicated that most T2DM patients bene ted from metabolic surgery. Preoperative TG and lymphocyte count were independently associated with post-operative complete remission of T2DM. Preoperative evaluation of lipid metabolism and immune function may be used to evaluate the prognosis of metabolic surgery and avoid unnecessary operation.