Surgery for Obesity and Related Diseases 11 (2015) 1266–1272
Original article
Impact of gastrointestinal bypass on nonmorbidly obese type 2 diabetes mellitus patients after gastrectomy Jun Paka, Yeongkeun Kwon, M.D.a, Emanuele Lo Menzo, M.D., Ph.D., F.A.C.S, F.A.S.M.B.Sb, Sungsoo Park, M.D., Ph.D.a,*, Samuel Szomstein, M.D., F.A.S.M.B.S.b, Raul J. Rosenthal, M.D., F.A.C.S., F.A.S.M.B.S.b a
Division of Upper Gastrointestinal Surgery, Department of Surgery, Korea University College of Medicine, Seoul, Korea b Bariatric and Metabolic Institute, Section of Minimally Invasive Surgery, Cleveland Clinic Florida, Weston, Florida Received November 7, 2014; accepted December 8, 2014
Abstract
Background: Our objective was to investigate the predictive preoperative factors and surgical components for type 2 diabetes mellitus (T2D) improvement in patients with body mass index (BMI) o35 kg/m2. Methods: All patients undergoing curative surgical resection for gastric cancer involving Billroth I gastroduodenal anastomosis, Billroth II gastrojejunal anastomosis (B-I, B-II), or Roux-en-Y total gastrectomy (RYTG), from 2008–2011, were retrospectively reviewed. Of these, 90 patients with T2D were analyzed. The study population was divided into the “improved” and “not improved” groups. The preoperative and postoperative data were assessed using multiple logistic regression analysis. To assess the necessary surgical elements, the gastrointestinal reconstruction methods were categorized according to the presence of the fundus and gastrointestinal bypass. Results: Fifty-four patients (60%) experienced improvements in their T2D 2 years after surgery. Lower preoperative glycated hemoglobin (A1C) (odds ratio [OR]: .502; 95% confidence interval [CI]: .313–.804; P ¼ .004), not using multiple oral antidiabetic medications (OR: .341; 95% CI: .120–.969; P ¼ .043), and high BMI before surgery (OR: 1.294; 95% CI: 1.074–1.559; P ¼ .007) were identified as independent predictors of T2D improvements. RYTG was more effective at improving T2D than B-I (OR: .160; 95% CI: .032–.794; P ¼ .025). Statistical analysis according to the surgical elements showed that the bypass procedure was associated with T2D improvements (OR: 3.023; 95% CI: .989–9 240; P ¼ .052). Conclusion: Gastrointestinal bypass significantly contributes to T2D improvements in patients with BMI o35 kg/m2. Low A1C, high BMI, and not using multiple antidiabetic medications were important predictors of T2D improvement. (Surg Obes Relat Dis 2015;11:1266–1272.) r 2015 American Society for Metabolic and Bariatric Surgery. All rights reserved.
Keywords:
Type 2 DM; Bariatric surgery; BMI; Predictor; Gastrectomy
The metabolic effects of bariatric surgery have been previously investigated. However, the extent to which metabolic diseases are mitigated and the mechanisms that underlie * Correspondence: Sungsoo Park, M.D, Ph.D, Division of Upper Gastrointestinal Surgery, Department of Surgery, Korea University Anam Hospital, Korea University College of Medicine, Inchon-ro 73, Seongbukgu, Seoul 136-705, Korea. Telephone number: þ82 2 920 6772. Fax number: þ82 2 928 1631. E-mail: [email protected]
the metabolic effects of bariatric surgery are poorly understood, particularly in nonmorbidly obese individuals, that is, those with a body mass index (BMI) of o35 kg/m2; and despite the metabolic benefits, the existing national and international recommendations suggest avoiding metabolic and bariatric surgery in patients whose BMI is o35 kg/m2 [1]. Patients who have higher BMIs are at greater risk for developing diabetes [1,2]; therefore, it is difficult to determine whether improvements in type 2 diabetes mellitus
http://dx.doi.org/10.1016/j.soard.2014.12.008 1550-7289/r 2015 American Society for Metabolic and Bariatric Surgery. All rights reserved.
Surgical factor for T2D improvement in BMI o35 / Surgery for Obesity and Related Diseases 11 (2015) 1266–1272
1267
(T2D) in these patients are caused by postoperative weight reductions or by other endocrine effects associated with the bariatric surgery. Although the bariatric surgery indications are currently based solely on BMI, the International Diabetes Federation has suggested that mildly obese patients, that is, those with BMIs of 30–35 kg/m2, with poorly controlled diabetes should also be considered for metabolic surgery [3]. Previous studies of bariatric surgery in patients whose BMI was o35 kg/m2 have tended to focus on weight loss and reductions in glycated hemoglobin (A1C) levels [4,5]; hence, there is currently insufficient evidence about the metabolic effects of bariatric surgery and the impacts of these metabolic effects on patients with a BMI o35 kg/m2. Bariatric surgery and gastrectomy share some fundamental components. This study investigated which surgical components are necessary for metabolic surgery and assessed the best method for metabolic surgery based on the outcomes in patients with a BMI o35 kg/m2. Furthermore, we also analyzed the preoperative and postoperative parameters of mildly obese patients with T2D. The surgical factors that are thought to contribute to improvements in T2D were investigated in detail.
Table 1 Baseline characteristics of the study patients
Methods
BMI ¼ Body mass index; DM ¼ Diabetes mellitus; A1C ¼ Glycated hemoglobin * Continuous data are presented as mean ⫾ standard deviations.
Study sample The data of all consecutive patients who underwent surgical resection for gastric cancer at our organization over a 4-year period, from January 2008 to November 2011, were reviewed. All patients were followed up for at least 2 years postoperatively. Table 1 shows the baseline characteristics of the patients. Patients were excluded from the study if their BMI was Z35 kg/m2. Patients were included in the study if they had undergone potentially curative resections of their stage I, II, and III primary tumors, and had survived for longer than 2 years after their gastrectomy. Patients who experienced any complications associated with their gastrectomy were excluded from the study, because of the potential for unexpected metabolic effects [6]. Consequently, 90 patients who had undergone potentially curative gastric cancer resections were selected for study participation. The seventh edition of American Joint Committee on Cancer staging manual was used to stage the gastric tumors [7], and the American Diabetes Association’s criteria were used to diagnose diabetes [8]. Group definitions The patients were assigned to 2 groups based on the T2D improvements observed, namely, “improved” and “not improved.” According to the consensus statement from the American Diabetes Association, partial remission was indicated by sub-diabetic hyperglycemia, which includes an A1C level of o6.5% and a fasting blood glucose level of
Variable*
All patients (n ¼ 90)
Age (years) Gender Male:female ratio, n (%) Diabetes factors A1C (%) Fasting glucose (mg/dL) Duration of diabetes (years) Smoking Current, n (%) Anthropometrics Weight (kg) BMI (kg/m2) Blood pressure (mm Hg) Systolic Tumor and operative factors, n (%) Method of reconstruction Total, Roux-en-Y Subtotal Billroth I Billroth II Cancer stage I II III
65.4 ⫾ 10.2 63:27 (70.0:30.0) 7.1 ⫾ 1.3 151 ⫾ 56 25.6 ⫾ 9.7 41 (45.6) 64.9 ⫾ 11.5 24.8 ⫾ 3.4 119 ⫾ 12
19 (21.1) 29 (32.2) 42 (46.7) 63 (70.0) 13 (14.4) 14 (15.6)
100–125 mg/dL (5.6–6.9 mmol/L) for at least one year in the absence of active pharmacologic therapy or ongoing procedures [9]. In addition, complete remission was indicated by a return to “normal” measurements of glucose metabolism, including a normal A1C level and a fasting blood glucose level of 100 mg/dL or less for at least one year in the absence of active pharmacologic therapy or ongoing procedures [9]. The improved group included T2D patients who showed partial or complete remission of their diabetes. In addition, the improved group included patients who reduced their diabetes medication dosages or showed decreases of 41% in their A1C levels and fasting blood glucose levels of 25 mg/dL or less, without increasing the dosages of their diabetes medications [10]. Operative methods For all patients, radical oncologic resection of the stomach was performed for the treatment of gastric cancer. The resections included omentectomy and D2þα lymph node dissections. The surgical resection and reconstruction methods included radical total gastrectomies (TG) and Roux-en-Y (RY) total gastrectomy (RYTG), and radical subtotal gastrectomy with Billroth I gastroduodenal anastomosis (B-I) or Billroth II gastrojejunal anastomosis (B-II). During a subtotal gastrectomy, approximately 75% of the distal stomach was resected, and intestinal continuity was restored using a B-I or B-II reconstruction using 15–20 cm
1268
J. Pak et al. / Surgery for Obesity and Related Diseases 11 (2015) 1266–1272
whereas B-I preserves the fundus but has no gastrointestinal bypass effect.
Table 2 Surgical components of each surgical procedure. Surgical component
RYTG Subtotal, Billroth I Subtotal, Billroth II
Gastrointestinal bypass þ Preserved fundus -
þ
Statistical analysis
þ þ
RYTG ¼ Roux-en-Y total gastrectomy
of the afferent limb. During a total gastrectomy, RYTG was performed using a 40 cm Roux limb and a 20 cm afferent limb. In all patients, vagotomy was routinely performed during the lymph node dissections. Components of each surgical procedure To assess the surgical components that may contribute to T2D improvements, the gastrointestinal reconstruction methods RYTG, B-II, and B-I were reconsidered in relation to the presence of the fundus and the gastrointestinal bypass procedure (Table 2). RYTG is characterized by gastrointestinal bypass effects and removal of the fundus, B-II preserves the fundus and has gastrointestinal bypass effects,
Continuous data are presented as mean ⫾ standard deviations (SD) and were analyzed using the independent samples t test. Categorical data are presented as percentages and were analyzed using the chi-squared test. Multivariate logistic regression analysis was used to compare the improved and not improved groups, and the data were adjusted for age, sex, systolic blood pressure, A1C levels, duration of T2D, the use of multiple oral antidiabetic medications, insulin use, smoking, BMI, the total cholesterol level before surgery, changes in BMI 2 years after surgery, changes in the total cholesterol level 2 years after surgery, and the reconstruction method used. All parameters shown in Table 3 and Table 4 were used as variables in this model. Moreover, the multivariate logistic regression model was additionally applied incorporating the reconstruction method, according to the presence of the fundus and gastrointestinal bypass. Risks were evaluated for the
Table 3 Characteristics of the patients in the not improved and improved groups Variable* Age (years) Male: female, n (%) Diabetes factors A1C (%) Fasting glucose (mg/dL) Duration of diabetes (years) Hypoglycemic medication, n (%) Monotherapy Multiple oral medications Insulin Smoking Yes Anthropometrics Weight (kg) BMI (kg/m2) Laboratory values Total cholesterol (mg/dL) Blood pressure (mm Hg) Systolic Diastolic Tumor and operative factors, n (%) Method of reconstruction Total, Roux-en-Y Subtotal Billroth I Billroth II Cancer stage I II III
Not improved (n ¼ 36)
Improved (n ¼ 54)
64.8 ⫾ 11.6 25:11 (69.4:30.6)
65.8 ⫾ 9.3 38:16 (70.4:29.6)
.625 .925
7.6 ⫾ 1.5 163 ⫾ 57 25.1 ⫾ 8.9
6.8 ⫾ 1.1 143 ⫾ 56 25.9 ⫾ 10.2
.002 .088 .694
36 (100.0) 20 (55.6) 4 (11.1)
54 (100.0) 20 (37.0) 2 (3.7)
.083 .213
16 (44.4)
25 (46.3)
.863
62.2 ⫾ 11.4 23.7 ⫾ 3.0
66.7 ⫾ 11.3 25.5 ⫾ 3.5
.068 .013
156 ⫾ 34
163 ⫾ 36
.298
120 ⫾ 13 73 ⫾ 8
118 ⫾ 11 75 ⫾ 7
.383 .232
P value
.201 5 (13.9)
14 (25.9)
15 (41.7) 16 (44.4)
14 (25.9) 26 (48.1)
25 (69.4) 5 (13.9) 6 (16.7)
39 (70.4) 8 (14.8) 8 (14.8)
.969
BMI ¼ Body mass index; A1C ¼ Glycated hemoglobin. * Continuous data are presented as mean ⫾ standard deviations. † Independent samples t test. b Chi-squared test.
† b
† † †
‡ ‡ ‡ † † † † † ‡
‡
Surgical factor for T2D improvement in BMI o35 / Surgery for Obesity and Related Diseases 11 (2015) 1266–1272
1269
Table 4 Changes in the parameters 2 years after surgery Variable*
Total
Not improved (n ¼ 36)
Improved (n ¼ 54)
P value†
A1C (%) Change from baseline, % Fasting glucose (mg/dL) Change from baseline, % Body mass index Change from baseline, % Total cholesterol (mg/dL) Change from baseline, %
6.3 ⫾ 1.2 17.4 ⫾ 45.1 125 ⫾ 33 26.4 ⫾ 53.6 22.3 ⫾ 3.5 11.0 ⫾ 8.8 159 ⫾ 31 3.8 ⫾ 27.4
6.5 ⫾ 1.7 29.2 ⫾ 68.0 142 ⫾ 37 20.0 ⫾ 47.3 21.8 ⫾ 3.1 8.6 ⫾ 6.4 165 ⫾ 32 3.7 ⫾ 21.5
6.2 ⫾.7 9.5 ⫾ 14.2 113 ⫾ 24 30.8 ⫾ 57.5 22.6 ⫾ 3.7 12.6 ⫾ 9.9 154 ⫾ 29 8.8 ⫾ 29.8
.284 .096 o .001 .354 .263 .036 .122 .033
A1C: Glycated hemoglobin. * Continuous data are presented as mean ⫾ standard deviations. † Independent samples t test.
patients using odds ratios (OR) with 95% confidence intervals (CI). In the multivariate logistic regression model, a backward elimination procedure was used in the likelihood ratio test, with an entrance level of .05 and a removal level of .1. A P value of o .05 was considered statistically significant in all of the analyses, and all P values were 2sided. All statistical analyses were performed using IBMs SPSSs software version 20 (IBM, Armonk, New York, USA). Results Baseline patient characteristics before surgery Table 3 shows the baseline characteristics of the 2 patient groups. Significant differences were observed between the groups in relation to the preoperative A1C levels and the BMIs. The A1C level was significantly lower in the improved group compared with the not improved group (6.8% versus 7.6%; P ¼ .002), and the BMI was significantly higher in the improved group compared with the not improved group (25.5 versus 23.7 kg/m2; P ¼ .013). In addition, the use of multiple oral antidiabetic medications tended to be more frequent in the not improved group (not improved versus improved: 55.6% versus 37.0%; P ¼ .083). Outcomes 2 years after surgery A total of 54 (60%) patients experienced improvements in relation to the management of T2D 2 years after their gastrectomy (Table 4). Among these, 34 were taking lower dosages of their antidiabetic medications, and 22 out of these 34 patients had stopped taking oral antidiabetic medication during the 2-year follow-up period. Both patient groups had lower A1C levels compared with the baseline levels. Although the A1C level in the not improved group was not below the therapeutic goal preoperatively, there was no significant difference between the 2 groups in relation to the A1C levels attained 2 years after surgery (not improved versus improved: 6.5% versus 6.2%, P ¼ .284). Two years after the gastrectomy, the not improved
group showed a larger percentage change in the A1C level compared with the improved group, a difference that showed borderline significance (29.2% versus –9.5%, P ¼ .096). In addition, the mean fasting plasma glucose levels were 142 mg/dL and 113 mg/dL in the not improved and improved groups, respectively (P o .001). Two years after the gastrectomy, the mean BMIs were 21.8 and 22.6 kg/m2 in the not improved and improved groups, which represented reductions of 8.6% and 12.6% from the baseline value, respectively (P ¼ .036). After 2 years, the amount by which the total cholesterol level had reduced in the improved group was larger compared with in the not improved group (–8.8% versus 3.7%, P ¼ .033; 165 versus 154 mg/dL, P ¼ .122).
Associations between patient characteristics before and after surgery and improvements in type 2 diabetes mellitus The univariate logistic regression analysis showed that the A1C level, baseline BMI, and changes in the BMI and total cholesterol level after 2 years were significantly associated with T2D improvements after surgery (Table 5). The multivariate analysis identified a lower preoperative A1C level (OR: .502; 95% CI: .313–.804; P ¼ .004), not using multiple oral antidiabetic medications (OR: .341; 95% CI: .120–.969; P ¼ .043), and a high BMI preoperatively (OR: 1.294; 95% CI: 1.074–1 559; P ¼ .007) as independent predictors of T2D improvements. In addition, RYTG was more effective at achieving T2D improvements than B-I (OR: .160; 95% CI: .032–.794; P ¼ .025), whereas there was no significant difference between RYTG and B-II in relation to T2D improvements (OR: .274; 95% CI: .057–1.305; P ¼ .104). The improved group showed a significant change in the total cholesterol level after gastrectomy (OR: 1.028; 95% CI: 1.004–1.053; P ¼ .020). The multivariate logistic regression analysis incorporating surgical factors (presence of the fundus and gastrointestinal bypass) instead of the surgical method used (Tables 2 and 6) showed that the gastrointestinal bypass procedure improved T2D with a borderline significance
1270
J. Pak et al. / Surgery for Obesity and Related Diseases 11 (2015) 1266–1272
Table 5 Univariate and multivariate logistic regression analyses of the pre- and postoperative predictive factors in type 2 diabetes mellitus after surgery Variables
Age (per one-year increase) Sex, female (versus male) Systolic blood pressure (per 1 mm Hg increase) A1C (per 1% increase) Duration of DM (per one-year increase) Multiple oral antidiabetic medications (versus monotherapy) Insulin, yes (versus no) Smoking, yes (versus no) BMI (per 1 kg/m2 increase) BMI change (per 1% decrease) Total cholesterol (per 1 mg/dL increase) Total cholesterol change (per 1% decrease) Method of reconstruction Total, Roux-en-Y Subtotal, Billroth I Subtotal, Billroth II
Univariate analysis
Multivariate analysis
OR (95% CI)
P value
1.011 (.969–1.053) .957 (.382–2.398) .984 (.948–1.020) .560 (.377–.832) 1.009 (.966–1.054) .471 (.199–1.111) .308 (.053–1.777) 1.078 (.462–2.515) 1.186 (1.031–1.364) 1.060 (1.002–1.122) 1.007 (.996–1.019) 1.020 (1.001–1.039)
.620 .925 .380 .004 .690 .085 .188 .863 .017 .042 .296 .040
1 .333 (.095–1.168) .580 (.175–1.919)
.086 .373
OR (95% CI)
P value
.502 (.313–.804)
.004
.341 (.120–.969)
.043
1.294 (1.074–1.559)
.007
1.028 (1.004–1.053)
.020
1 .160 (.032–.794) .274 (.057–1.305)
.025 .104
BMI ¼ Body mass index; CI ¼ Confidence interval; DM ¼ Diabetes mellitus; A1C ¼ Glycated hemoglobin; OR ¼ Odds ratio.
(OR: 3.023; 95% CI: .989–9.240; P = .052), whereas the results were similar for all other variables (Tables 5 and 6). Discussion Sleeve gastrectomies and Roux-en-Y gastric bypasses are currently the most effective types of weight loss surgical techniques. In relation to this, we analyzed the independent effects of the presence of the fundus and the gastrointestinal bypass procedure on metabolism by examining the effects of these 2 surgical components on T2D improvements. Patients are known to experience rapid decreases in their A1C levels before losing weight after undergoing bariatric surgery [11,12], and the weight loss generally continues until 15–18 months after surgery, at which time the patients
begin to regain weight [13,14]. Two years after surgery, the weight changes typically stop, and the metabolic state stabilizes. This study focus on the metabolic effects 2 years after surgery, as opposed to short-term catabolic effects. The results from the multivariate logistic regression analysis (Table 5) suggest that the patients with better preoperative control of their A1C levels were more likely to achieve better control of their T2D after surgery (OR: .502; 95% CI: .313–.804; P ¼ .004). The patients who took oral antidiabetic medications before surgery and had A1C levels that were below the therapeutic goal of 7.0% [15] generally showed T2D improvements 2 years after surgery. These results concur with the findings from previously published studies that have shown that the benefits of metabolic surgery tend to be reduced in patients whose A1C levels
Table 6 Results of the univariate and multivariate logistic regression analyses incorporating the reconsidered surgical factors Variables
Age (per one-year increase) Gender, female (versus male) Systolic blood pressure (per 1 mm Hg increase) A1C (per 1% increase) Duration of DM (per one-year increase) Multiple oral antidiabetic medications (versus monotherapy) Insulin: yes (versus no) Smoking: yes (versus no) BMI (per 1 kg/m2 increase) BMI change (per 1% decrease) Total cholesterol (per 1 mg/dL increase) Total cholesterol change (per 1% decrease) Bypass procedure (versus no bypass procedure) Fundus preserved (versus absent fundus)
Univariate analysis
Multivariate analysis
OR (95% CI)
P value
1.011 (.969–1.053) .957 (.382–2.398) .984 (.948–1.020) .560 (.377–.832) 1.009 (.966–1.054) .471 (.199–1.111) .308 (.053–1.777) 1.078 (.462–2.515) 1.186 (1.031–1.364) 1.060 (1.002–1.122) 1.007 (.996–1.019) 1.020 (1.001–1.039) 2.041 (.830–5.018) .461 (.150–1.418)
.620 .925 .380 .004 .690 .085 .188 .863 .017 .042 .296 .040 .120 .177
OR (95% CI)
P value
.479 (.297–.771)
.002
.348 (.123–.989) .184 (.024–1.438)
.048 .107
1.223 (1.030–1.452)
.021
1.024 (1.001–1.048) 3.023 (.989–9.240)
.041 .052
BMI ¼ Body mass index; CI ¼ Confidence interval; DM ¼ Diabetes mellitus; A1C ¼ Glycated hemoglobin; OR ¼ Odds ratio.
Surgical factor for T2D improvement in BMI o35 / Surgery for Obesity and Related Diseases 11 (2015) 1266–1272
were not controlled [12,16]. In this study, patients in the not improved group showed significant decreases in their A1C levels, and these levels remained below the diabetic therapeutic goal of o7.0% 2 years after surgery. These outcomes are thought to be associated with the strict T2D control that was achieved through follow-up assessments taking place at the Endocrinology Department within our organization. Eight out of the 36 patients in the not improved group either increased their antidiabetic medication dosages or added insulin to their therapeutic regimens, whereas almost half of the patients (17/36) showed considerable reductions in their A1C levels without any changes to their antidiabetic medications, which indicated that a substantial number of patients benefited from their metabolic surgery despite being assigned to the not improved group. Thus, although we have drawn conclusions that are similar to those reported in other studies, we also found that several patients in the not improved group had the potential to achieve long-term metabolic benefits from surgery. The patients who were taking multiple medications for their diabetes had difficulties in achieving T2D improvements (OR: .341; 95% CI: .120–.969; P = .043). In the course of T2D, most patients require therapy with multiple medications to achieve their glycemic goals [17,18], and this suggests that the need for multiple medications may correlate with the severity of their T2D. The analysis of the study results in relation to these preoperative factors indicates that the failure of metabolic surgery was consistent with the progression of T2D [19,20]. In relation to T2D deterioration as determined by the A1C levels, the use of multiple oral antidiabetic medications was found to be a major determinant of the patients’ ability to achieve target glucose levels. Several studies have demonstrated that bariatric surgery has greater metabolic effects on patients with short T2D durations [19,20]; however, these studies only considered T2D of relatively short durations. In contrast, this study analyzed patients with chronic T2D, with a mean ⫾ SD duration of 25.6 ⫾ 9.7 years; and, in the present study, the T2D duration was not representative of its severity, as the patients’ T2D was already advanced. It is known that higher BMIs are associated with higher levels of visceral adipose tissue [1,21,22]. In addition, despite having lower absolute BMI values, people from East Asian countries reportedly show the highest accumulations of visceral adipose tissue and the lowest accumulations of deep subcutaneous adipose tissue as adiposity increases [21]. In both sexes, the amount of visceral adipose tissue is a predictive factor for the presence of T2D [1,22]; and accordingly, the patients in the improved group tended to have more ectopic adipose tissue before surgery, even though their BMI was o35 kg/m2. Metabolic surgery resolves issues associated with ectopic adipose tissue by reducing weight more effectively in patients with relatively high BMI, even when o35 kg/m2 [23,24]. Additionally, the total cholesterol levels in the improved group were found to
1271
be reduced during the 2-year follow-up period, which is consistent with the findings from previously reported studies [25–28]. In the present study, the multivariate logistic regression indicated that RYTG was superior to B-I with respect to T2D improvements (OR: .160; 95% CI: .032–.794; P ¼ .025). RYTG has 2 surgical components that differ from BI, namely, the elimination of the gastric fundus and the gastrointestinal bypass procedure. In contrast, there was no significant difference between RYTG and B-II in relation to T2D improvements (OR: .274; 95% CI: .057–1.305; P ¼ .104), which may be due to the fact that these 2 surgical techniques share the gastrointestinal bypass factor. It is well known that, in nonobese diabetic animal models, gastrojejunal bypass can improve glycemic control and is accompanied by significant endocrine alterations [11,12,27,28]. Intestinal bypass can lead to hindgut and foregut effects; and, indeed, it has been speculated that intestinal bypass prevents contact between the short segment of the proximal small intestine and ingested nutrients, which in turn may lead to direct antidiabetic effects [12,27– 29]. Furthermore, an increase in the levels of unabsorbed nutrients in the distal gut can accentuate the secretion of glucagon-like peptide-1, thereby improving glucose homeostasis [27,29]. In contrast, resection of the fundus did not have a significant effect on improving T2D in patients whose BMI was o35 kg/m2. While removing the ghrelinsecreting cells would have reduced the ghrelin level [28,29], the effect of this was not statistically significant in relation to its impact on diabetes in this study, likely owing to the fact that the metabolic effect of the gastrointestinal bypass is more dominant than the loss of appetite in patients with BMI o35 kg/m2. Clinically, bariatric surgery has a considerable influence on weight loss, and it is clearly the most important weight loss factor for patients with BMI 435 kg/ m2 [23,24]. In this study, however, the weight loss effect was found to be lower in patients with BMI o35 kg/m2 than what has been reported in morbidly obese patients [26,27]. This suggests that, as the weight reduction declines, the metabolic effects associated with the weight loss also decline, and that the role of the gastrointestinal bypass itself in metabolic amelioration may become more important [5,30].
Conclusion The results of this study indicate that procedures involving gastrointestinal bypass represent the most effective surgical element for T2D patients with BMI o35 kg/m2. In terms of the metabolic effects of gastrointestinal modifications, patients with low A1C levels and not taking multiple oral antidiabetic medications before surgery were most likely to achieve T2D improvements. In addition, patients with high BMI before surgery tended to benefit
1272
J. Pak et al. / Surgery for Obesity and Related Diseases 11 (2015) 1266–1272
from metabolic surgery, even though they were not morbidly obese. Limitations The study was retrospective in its design and the sample was not randomized; and, therefore, the patients did not undergo gastrointestinal modifications for metabolic reasons alone. In addition, T2D was not constantly controlled using antidiabetic medication before and after surgery. There was a tendency for T2D to be more strictly controlled after surgery, and this was likely owing to the fact that the patients were regularly followed up in the Endocrinology Department within our organization. Despite these limitations, however, we believe that this study contributes to the body of knowledge on the topic by discussing the surgical elements of gastrectomy in the context of their metabolic effects in nonmorbidly obese patients. Disclosures The authors have no commercial associations that might be a conflict of interest in relation to this article. This work was supported by the National Research Foundation of Korea, which was funded by the Korean Government (2010-0024825). References [1] de Lauzon-Guillain B, Balkau B, Charles MA, Romieu I, BoutronRuault MC, Clavel-Chapelon F. Birth weight, body silhouette over the life course, and incident diabetes in 91,453 middle-aged women from the French Etude Epidemiologique de Femmes de la Mutuelle Generale de l'Education Nationale (E3 N) Cohort. Diabetes Care 2010;33:298–303. [2] Willett WC, Dietz WH, Colditz GA. Guidelines for healthy weight. N Engl J Med 1999;341:427–34. [3] Sjöström L, Lindroos AK, Peltonen M, et al. Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. N Engl J Med 2004;351:2683–93. [4] Maggard-Gibbons M, Maglione M, Livhits M, et al. Bariatric surgery for weight loss and glycemic control in nonmorbidly obese adults with diabetes: a systematic review. JAMA 2013;309:2250–61. [5] Chiellini C, Rubino F, Castagneto M, Nanni G, Mingrone G. The effect of bilio-pancreatic diversion on type 2 diabetes in patients with BMI o35 kg/m2. Diabetologia 2009;52:1027–30. [6] Yi HW, Kim SM, Kim SH, et al. Complications leading reoperation after gastrectomy in patients with gastric cancer: frequency, type, and potential causes. J Gastric Cancer 2013;13:242–6. [7] Edge SB, Compton CC. The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM. Ann Surg Oncol 2010;17:1471–4. [8] American Diabetes Association. Standards of medical care in diabetes–2014. Diabetes Care 2014;37(Suppl 1):S14–80. [9] Buse JB, Caprio S, Cefalu WT, et al. How do we define cure of diabetes? Diabetes Care 2009;32:2133–5. [10] Schauer PR, Burguera B, Ikramuddin S, et al. Effect of laparoscopic Roux-en-Y gastric bypass on type 2 diabetes mellitus. Ann Surg 2003;238:467–84; discussion 84-85.
[11] de Luis D, Domingo M, Romero A, et al. Effects of duodenal-jejunal exclusion on beta cell function and hormonal regulation in GotoKakizaki rats. Am J Surg 2012;204:242–7. [12] Rubino F, Marescaux J. Effect of duodenal-jejunal exclusion in a nonobese animal model of type 2 diabetes: a new perspective for an old disease. Ann Surg 2004;239:1–11. [13] Nocca D, Krawczykowsky D, Bomans B, et al. A prospective multicenter study of 163 sleeve gastrectomies: results at 1 and 2 years. Obes Surg 2008;18:560–5. [14] Bozzetti F, Ravera E, Cozzaglio L, et al. Comparison of nutritional status after total or subtotal gastrectomy. Nutrition 1990;6:371–5. [15] HbA1 c targets in type 2 diabetes: guidelines and evidence. Drug Ther Bull 2013;51:42–45. [16] Lee WJ, Chong K, Ser KH, et al. Gastric bypass vs sleeve gastrectomy for type 2 diabetes mellitus: a randomized controlled trial. Arch Surg 2011;146:143–8. [17] Turner RC, Cull CA, Frighi V, Holman RR. Glycemic control with diet, sulfonylurea, metformin, or insulin in patients with type 2 diabetes mellitus: progressive requirement for multiple therapies (UKPDS 49). UK Prospective Diabetes Study (UKPDS) Group. JAMA 1999;281:2005–12. [18] U.K. prospective diabetes study 16. Overview of 6 years' therapy of type II diabetes: a progressive disease. U.K. Prospective Diabetes Study Group. Diabetes 1995;44:1249–58. [19] Hall TC, Pellen MG, Sedman PC, Jain PK. Preoperative factors predicting remission of type 2 diabetes mellitus after Rouxen-Y gastric bypass surgery for obesity. Obes Surg 2010;20: 1245–50. [20] Lee WJ, Hur KY, Lakadawala M, et al. Predicting success of metabolic surgery: age, body mass index, C-peptide, and duration score. Surg Obes Relat Dis 2013;9:379–84. [21] Nazare JA, Smith JD, Borel AL, et al. Ethnic influences on the relations between abdominal subcutaneous and visceral adiposity, liver fat, and cardiometabolic risk profile: the International Study of Prediction of Intra-Abdominal Adiposity and Its Relationship With Cardiometabolic Risk/Intra-Abdominal Adiposity. Am J Clin Nutr 2012;96:714–26. [22] Sattar N, Wannamethee SG, Forouhi NG. Novel biochemical risk factors for type 2 diabetes: pathogenic insights or prediction possibilities? Diabetologia 2008;51:926–40. [23] Buchwald H, Estok R, Fahrbach K, et al. Weight and type 2 diabetes after bariatric surgery: systematic review and meta-analysis. Am J Med 2009;122(248-56):e5. [24] Long SD, O'Brien K, MacDonald KG Jr., et al. Weight loss in severely obese subjects prevents the progression of impaired glucose tolerance to type II diabetes. A longitudinal interventional study. Diabetes Care 1994;17:372–5. [25] Sjöström L, Peltonen M, Jacobson P, et al. Bariatric surgery and longterm cardiovascular events. JAMA 2012;307:56–65. [26] Buchwald H, Avidor Y, Braunwald E, et al. Bariatric surgery: a systematic review and meta-analysis. JAMA 2004;292:1724–37. [27] Drucker DJ. The role of gut hormones in glucose homeostasis. J Clin Invest 2007;117:24–32. [28] Cummings DE, Weigle DS, Frayo RS, et al. Plasma ghrelin levels after diet-induced weight loss or gastric bypass surgery. N Engl J Med 2002;346:1623–30. [29] Thaler JP, Cummings DE. Minireview: Hormonal and metabolic mechanisms of diabetes remission after gastrointestinal surgery. Endocrinology 2009;150:2518–25. [30] Hickey MS, Pories WJ, MacDonald KG Jr., et al. A new paradigm for type 2 diabetes mellitus: could it be a disease of the foregut? Ann Surg 1998;227:637–43; discussion 43–44.
Original article
Impact of gastrointestinal bypass on nonmorbidly obese type 2 diabetes mellitus patients after gastrectomy Jun Paka, Yeongkeun Kwon, M.D.a, Emanuele Lo Menzo, M.D., Ph.D., F.A.C.S, F.A.S.M.B.Sb, Sungsoo Park, M.D., Ph.D.a,*, Samuel Szomstein, M.D., F.A.S.M.B.S.b, Raul J. Rosenthal, M.D., F.A.C.S., F.A.S.M.B.S.b a
Division of Upper Gastrointestinal Surgery, Department of Surgery, Korea University College of Medicine, Seoul, Korea b Bariatric and Metabolic Institute, Section of Minimally Invasive Surgery, Cleveland Clinic Florida, Weston, Florida Received November 7, 2014; accepted December 8, 2014
Abstract
Background: Our objective was to investigate the predictive preoperative factors and surgical components for type 2 diabetes mellitus (T2D) improvement in patients with body mass index (BMI) o35 kg/m2. Methods: All patients undergoing curative surgical resection for gastric cancer involving Billroth I gastroduodenal anastomosis, Billroth II gastrojejunal anastomosis (B-I, B-II), or Roux-en-Y total gastrectomy (RYTG), from 2008–2011, were retrospectively reviewed. Of these, 90 patients with T2D were analyzed. The study population was divided into the “improved” and “not improved” groups. The preoperative and postoperative data were assessed using multiple logistic regression analysis. To assess the necessary surgical elements, the gastrointestinal reconstruction methods were categorized according to the presence of the fundus and gastrointestinal bypass. Results: Fifty-four patients (60%) experienced improvements in their T2D 2 years after surgery. Lower preoperative glycated hemoglobin (A1C) (odds ratio [OR]: .502; 95% confidence interval [CI]: .313–.804; P ¼ .004), not using multiple oral antidiabetic medications (OR: .341; 95% CI: .120–.969; P ¼ .043), and high BMI before surgery (OR: 1.294; 95% CI: 1.074–1.559; P ¼ .007) were identified as independent predictors of T2D improvements. RYTG was more effective at improving T2D than B-I (OR: .160; 95% CI: .032–.794; P ¼ .025). Statistical analysis according to the surgical elements showed that the bypass procedure was associated with T2D improvements (OR: 3.023; 95% CI: .989–9 240; P ¼ .052). Conclusion: Gastrointestinal bypass significantly contributes to T2D improvements in patients with BMI o35 kg/m2. Low A1C, high BMI, and not using multiple antidiabetic medications were important predictors of T2D improvement. (Surg Obes Relat Dis 2015;11:1266–1272.) r 2015 American Society for Metabolic and Bariatric Surgery. All rights reserved.
Keywords:
Type 2 DM; Bariatric surgery; BMI; Predictor; Gastrectomy
The metabolic effects of bariatric surgery have been previously investigated. However, the extent to which metabolic diseases are mitigated and the mechanisms that underlie * Correspondence: Sungsoo Park, M.D, Ph.D, Division of Upper Gastrointestinal Surgery, Department of Surgery, Korea University Anam Hospital, Korea University College of Medicine, Inchon-ro 73, Seongbukgu, Seoul 136-705, Korea. Telephone number: þ82 2 920 6772. Fax number: þ82 2 928 1631. E-mail: [email protected]
the metabolic effects of bariatric surgery are poorly understood, particularly in nonmorbidly obese individuals, that is, those with a body mass index (BMI) of o35 kg/m2; and despite the metabolic benefits, the existing national and international recommendations suggest avoiding metabolic and bariatric surgery in patients whose BMI is o35 kg/m2 [1]. Patients who have higher BMIs are at greater risk for developing diabetes [1,2]; therefore, it is difficult to determine whether improvements in type 2 diabetes mellitus
http://dx.doi.org/10.1016/j.soard.2014.12.008 1550-7289/r 2015 American Society for Metabolic and Bariatric Surgery. All rights reserved.
Surgical factor for T2D improvement in BMI o35 / Surgery for Obesity and Related Diseases 11 (2015) 1266–1272
1267
(T2D) in these patients are caused by postoperative weight reductions or by other endocrine effects associated with the bariatric surgery. Although the bariatric surgery indications are currently based solely on BMI, the International Diabetes Federation has suggested that mildly obese patients, that is, those with BMIs of 30–35 kg/m2, with poorly controlled diabetes should also be considered for metabolic surgery [3]. Previous studies of bariatric surgery in patients whose BMI was o35 kg/m2 have tended to focus on weight loss and reductions in glycated hemoglobin (A1C) levels [4,5]; hence, there is currently insufficient evidence about the metabolic effects of bariatric surgery and the impacts of these metabolic effects on patients with a BMI o35 kg/m2. Bariatric surgery and gastrectomy share some fundamental components. This study investigated which surgical components are necessary for metabolic surgery and assessed the best method for metabolic surgery based on the outcomes in patients with a BMI o35 kg/m2. Furthermore, we also analyzed the preoperative and postoperative parameters of mildly obese patients with T2D. The surgical factors that are thought to contribute to improvements in T2D were investigated in detail.
Table 1 Baseline characteristics of the study patients
Methods
BMI ¼ Body mass index; DM ¼ Diabetes mellitus; A1C ¼ Glycated hemoglobin * Continuous data are presented as mean ⫾ standard deviations.
Study sample The data of all consecutive patients who underwent surgical resection for gastric cancer at our organization over a 4-year period, from January 2008 to November 2011, were reviewed. All patients were followed up for at least 2 years postoperatively. Table 1 shows the baseline characteristics of the patients. Patients were excluded from the study if their BMI was Z35 kg/m2. Patients were included in the study if they had undergone potentially curative resections of their stage I, II, and III primary tumors, and had survived for longer than 2 years after their gastrectomy. Patients who experienced any complications associated with their gastrectomy were excluded from the study, because of the potential for unexpected metabolic effects [6]. Consequently, 90 patients who had undergone potentially curative gastric cancer resections were selected for study participation. The seventh edition of American Joint Committee on Cancer staging manual was used to stage the gastric tumors [7], and the American Diabetes Association’s criteria were used to diagnose diabetes [8]. Group definitions The patients were assigned to 2 groups based on the T2D improvements observed, namely, “improved” and “not improved.” According to the consensus statement from the American Diabetes Association, partial remission was indicated by sub-diabetic hyperglycemia, which includes an A1C level of o6.5% and a fasting blood glucose level of
Variable*
All patients (n ¼ 90)
Age (years) Gender Male:female ratio, n (%) Diabetes factors A1C (%) Fasting glucose (mg/dL) Duration of diabetes (years) Smoking Current, n (%) Anthropometrics Weight (kg) BMI (kg/m2) Blood pressure (mm Hg) Systolic Tumor and operative factors, n (%) Method of reconstruction Total, Roux-en-Y Subtotal Billroth I Billroth II Cancer stage I II III
65.4 ⫾ 10.2 63:27 (70.0:30.0) 7.1 ⫾ 1.3 151 ⫾ 56 25.6 ⫾ 9.7 41 (45.6) 64.9 ⫾ 11.5 24.8 ⫾ 3.4 119 ⫾ 12
19 (21.1) 29 (32.2) 42 (46.7) 63 (70.0) 13 (14.4) 14 (15.6)
100–125 mg/dL (5.6–6.9 mmol/L) for at least one year in the absence of active pharmacologic therapy or ongoing procedures [9]. In addition, complete remission was indicated by a return to “normal” measurements of glucose metabolism, including a normal A1C level and a fasting blood glucose level of 100 mg/dL or less for at least one year in the absence of active pharmacologic therapy or ongoing procedures [9]. The improved group included T2D patients who showed partial or complete remission of their diabetes. In addition, the improved group included patients who reduced their diabetes medication dosages or showed decreases of 41% in their A1C levels and fasting blood glucose levels of 25 mg/dL or less, without increasing the dosages of their diabetes medications [10]. Operative methods For all patients, radical oncologic resection of the stomach was performed for the treatment of gastric cancer. The resections included omentectomy and D2þα lymph node dissections. The surgical resection and reconstruction methods included radical total gastrectomies (TG) and Roux-en-Y (RY) total gastrectomy (RYTG), and radical subtotal gastrectomy with Billroth I gastroduodenal anastomosis (B-I) or Billroth II gastrojejunal anastomosis (B-II). During a subtotal gastrectomy, approximately 75% of the distal stomach was resected, and intestinal continuity was restored using a B-I or B-II reconstruction using 15–20 cm
1268
J. Pak et al. / Surgery for Obesity and Related Diseases 11 (2015) 1266–1272
whereas B-I preserves the fundus but has no gastrointestinal bypass effect.
Table 2 Surgical components of each surgical procedure. Surgical component
RYTG Subtotal, Billroth I Subtotal, Billroth II
Gastrointestinal bypass þ Preserved fundus -
þ
Statistical analysis
þ þ
RYTG ¼ Roux-en-Y total gastrectomy
of the afferent limb. During a total gastrectomy, RYTG was performed using a 40 cm Roux limb and a 20 cm afferent limb. In all patients, vagotomy was routinely performed during the lymph node dissections. Components of each surgical procedure To assess the surgical components that may contribute to T2D improvements, the gastrointestinal reconstruction methods RYTG, B-II, and B-I were reconsidered in relation to the presence of the fundus and the gastrointestinal bypass procedure (Table 2). RYTG is characterized by gastrointestinal bypass effects and removal of the fundus, B-II preserves the fundus and has gastrointestinal bypass effects,
Continuous data are presented as mean ⫾ standard deviations (SD) and were analyzed using the independent samples t test. Categorical data are presented as percentages and were analyzed using the chi-squared test. Multivariate logistic regression analysis was used to compare the improved and not improved groups, and the data were adjusted for age, sex, systolic blood pressure, A1C levels, duration of T2D, the use of multiple oral antidiabetic medications, insulin use, smoking, BMI, the total cholesterol level before surgery, changes in BMI 2 years after surgery, changes in the total cholesterol level 2 years after surgery, and the reconstruction method used. All parameters shown in Table 3 and Table 4 were used as variables in this model. Moreover, the multivariate logistic regression model was additionally applied incorporating the reconstruction method, according to the presence of the fundus and gastrointestinal bypass. Risks were evaluated for the
Table 3 Characteristics of the patients in the not improved and improved groups Variable* Age (years) Male: female, n (%) Diabetes factors A1C (%) Fasting glucose (mg/dL) Duration of diabetes (years) Hypoglycemic medication, n (%) Monotherapy Multiple oral medications Insulin Smoking Yes Anthropometrics Weight (kg) BMI (kg/m2) Laboratory values Total cholesterol (mg/dL) Blood pressure (mm Hg) Systolic Diastolic Tumor and operative factors, n (%) Method of reconstruction Total, Roux-en-Y Subtotal Billroth I Billroth II Cancer stage I II III
Not improved (n ¼ 36)
Improved (n ¼ 54)
64.8 ⫾ 11.6 25:11 (69.4:30.6)
65.8 ⫾ 9.3 38:16 (70.4:29.6)
.625 .925
7.6 ⫾ 1.5 163 ⫾ 57 25.1 ⫾ 8.9
6.8 ⫾ 1.1 143 ⫾ 56 25.9 ⫾ 10.2
.002 .088 .694
36 (100.0) 20 (55.6) 4 (11.1)
54 (100.0) 20 (37.0) 2 (3.7)
.083 .213
16 (44.4)
25 (46.3)
.863
62.2 ⫾ 11.4 23.7 ⫾ 3.0
66.7 ⫾ 11.3 25.5 ⫾ 3.5
.068 .013
156 ⫾ 34
163 ⫾ 36
.298
120 ⫾ 13 73 ⫾ 8
118 ⫾ 11 75 ⫾ 7
.383 .232
P value
.201 5 (13.9)
14 (25.9)
15 (41.7) 16 (44.4)
14 (25.9) 26 (48.1)
25 (69.4) 5 (13.9) 6 (16.7)
39 (70.4) 8 (14.8) 8 (14.8)
.969
BMI ¼ Body mass index; A1C ¼ Glycated hemoglobin. * Continuous data are presented as mean ⫾ standard deviations. † Independent samples t test. b Chi-squared test.
† b
† † †
‡ ‡ ‡ † † † † † ‡
‡
Surgical factor for T2D improvement in BMI o35 / Surgery for Obesity and Related Diseases 11 (2015) 1266–1272
1269
Table 4 Changes in the parameters 2 years after surgery Variable*
Total
Not improved (n ¼ 36)
Improved (n ¼ 54)
P value†
A1C (%) Change from baseline, % Fasting glucose (mg/dL) Change from baseline, % Body mass index Change from baseline, % Total cholesterol (mg/dL) Change from baseline, %
6.3 ⫾ 1.2 17.4 ⫾ 45.1 125 ⫾ 33 26.4 ⫾ 53.6 22.3 ⫾ 3.5 11.0 ⫾ 8.8 159 ⫾ 31 3.8 ⫾ 27.4
6.5 ⫾ 1.7 29.2 ⫾ 68.0 142 ⫾ 37 20.0 ⫾ 47.3 21.8 ⫾ 3.1 8.6 ⫾ 6.4 165 ⫾ 32 3.7 ⫾ 21.5
6.2 ⫾.7 9.5 ⫾ 14.2 113 ⫾ 24 30.8 ⫾ 57.5 22.6 ⫾ 3.7 12.6 ⫾ 9.9 154 ⫾ 29 8.8 ⫾ 29.8
.284 .096 o .001 .354 .263 .036 .122 .033
A1C: Glycated hemoglobin. * Continuous data are presented as mean ⫾ standard deviations. † Independent samples t test.
patients using odds ratios (OR) with 95% confidence intervals (CI). In the multivariate logistic regression model, a backward elimination procedure was used in the likelihood ratio test, with an entrance level of .05 and a removal level of .1. A P value of o .05 was considered statistically significant in all of the analyses, and all P values were 2sided. All statistical analyses were performed using IBMs SPSSs software version 20 (IBM, Armonk, New York, USA). Results Baseline patient characteristics before surgery Table 3 shows the baseline characteristics of the 2 patient groups. Significant differences were observed between the groups in relation to the preoperative A1C levels and the BMIs. The A1C level was significantly lower in the improved group compared with the not improved group (6.8% versus 7.6%; P ¼ .002), and the BMI was significantly higher in the improved group compared with the not improved group (25.5 versus 23.7 kg/m2; P ¼ .013). In addition, the use of multiple oral antidiabetic medications tended to be more frequent in the not improved group (not improved versus improved: 55.6% versus 37.0%; P ¼ .083). Outcomes 2 years after surgery A total of 54 (60%) patients experienced improvements in relation to the management of T2D 2 years after their gastrectomy (Table 4). Among these, 34 were taking lower dosages of their antidiabetic medications, and 22 out of these 34 patients had stopped taking oral antidiabetic medication during the 2-year follow-up period. Both patient groups had lower A1C levels compared with the baseline levels. Although the A1C level in the not improved group was not below the therapeutic goal preoperatively, there was no significant difference between the 2 groups in relation to the A1C levels attained 2 years after surgery (not improved versus improved: 6.5% versus 6.2%, P ¼ .284). Two years after the gastrectomy, the not improved
group showed a larger percentage change in the A1C level compared with the improved group, a difference that showed borderline significance (29.2% versus –9.5%, P ¼ .096). In addition, the mean fasting plasma glucose levels were 142 mg/dL and 113 mg/dL in the not improved and improved groups, respectively (P o .001). Two years after the gastrectomy, the mean BMIs were 21.8 and 22.6 kg/m2 in the not improved and improved groups, which represented reductions of 8.6% and 12.6% from the baseline value, respectively (P ¼ .036). After 2 years, the amount by which the total cholesterol level had reduced in the improved group was larger compared with in the not improved group (–8.8% versus 3.7%, P ¼ .033; 165 versus 154 mg/dL, P ¼ .122).
Associations between patient characteristics before and after surgery and improvements in type 2 diabetes mellitus The univariate logistic regression analysis showed that the A1C level, baseline BMI, and changes in the BMI and total cholesterol level after 2 years were significantly associated with T2D improvements after surgery (Table 5). The multivariate analysis identified a lower preoperative A1C level (OR: .502; 95% CI: .313–.804; P ¼ .004), not using multiple oral antidiabetic medications (OR: .341; 95% CI: .120–.969; P ¼ .043), and a high BMI preoperatively (OR: 1.294; 95% CI: 1.074–1 559; P ¼ .007) as independent predictors of T2D improvements. In addition, RYTG was more effective at achieving T2D improvements than B-I (OR: .160; 95% CI: .032–.794; P ¼ .025), whereas there was no significant difference between RYTG and B-II in relation to T2D improvements (OR: .274; 95% CI: .057–1.305; P ¼ .104). The improved group showed a significant change in the total cholesterol level after gastrectomy (OR: 1.028; 95% CI: 1.004–1.053; P ¼ .020). The multivariate logistic regression analysis incorporating surgical factors (presence of the fundus and gastrointestinal bypass) instead of the surgical method used (Tables 2 and 6) showed that the gastrointestinal bypass procedure improved T2D with a borderline significance
1270
J. Pak et al. / Surgery for Obesity and Related Diseases 11 (2015) 1266–1272
Table 5 Univariate and multivariate logistic regression analyses of the pre- and postoperative predictive factors in type 2 diabetes mellitus after surgery Variables
Age (per one-year increase) Sex, female (versus male) Systolic blood pressure (per 1 mm Hg increase) A1C (per 1% increase) Duration of DM (per one-year increase) Multiple oral antidiabetic medications (versus monotherapy) Insulin, yes (versus no) Smoking, yes (versus no) BMI (per 1 kg/m2 increase) BMI change (per 1% decrease) Total cholesterol (per 1 mg/dL increase) Total cholesterol change (per 1% decrease) Method of reconstruction Total, Roux-en-Y Subtotal, Billroth I Subtotal, Billroth II
Univariate analysis
Multivariate analysis
OR (95% CI)
P value
1.011 (.969–1.053) .957 (.382–2.398) .984 (.948–1.020) .560 (.377–.832) 1.009 (.966–1.054) .471 (.199–1.111) .308 (.053–1.777) 1.078 (.462–2.515) 1.186 (1.031–1.364) 1.060 (1.002–1.122) 1.007 (.996–1.019) 1.020 (1.001–1.039)
.620 .925 .380 .004 .690 .085 .188 .863 .017 .042 .296 .040
1 .333 (.095–1.168) .580 (.175–1.919)
.086 .373
OR (95% CI)
P value
.502 (.313–.804)
.004
.341 (.120–.969)
.043
1.294 (1.074–1.559)
.007
1.028 (1.004–1.053)
.020
1 .160 (.032–.794) .274 (.057–1.305)
.025 .104
BMI ¼ Body mass index; CI ¼ Confidence interval; DM ¼ Diabetes mellitus; A1C ¼ Glycated hemoglobin; OR ¼ Odds ratio.
(OR: 3.023; 95% CI: .989–9.240; P = .052), whereas the results were similar for all other variables (Tables 5 and 6). Discussion Sleeve gastrectomies and Roux-en-Y gastric bypasses are currently the most effective types of weight loss surgical techniques. In relation to this, we analyzed the independent effects of the presence of the fundus and the gastrointestinal bypass procedure on metabolism by examining the effects of these 2 surgical components on T2D improvements. Patients are known to experience rapid decreases in their A1C levels before losing weight after undergoing bariatric surgery [11,12], and the weight loss generally continues until 15–18 months after surgery, at which time the patients
begin to regain weight [13,14]. Two years after surgery, the weight changes typically stop, and the metabolic state stabilizes. This study focus on the metabolic effects 2 years after surgery, as opposed to short-term catabolic effects. The results from the multivariate logistic regression analysis (Table 5) suggest that the patients with better preoperative control of their A1C levels were more likely to achieve better control of their T2D after surgery (OR: .502; 95% CI: .313–.804; P ¼ .004). The patients who took oral antidiabetic medications before surgery and had A1C levels that were below the therapeutic goal of 7.0% [15] generally showed T2D improvements 2 years after surgery. These results concur with the findings from previously published studies that have shown that the benefits of metabolic surgery tend to be reduced in patients whose A1C levels
Table 6 Results of the univariate and multivariate logistic regression analyses incorporating the reconsidered surgical factors Variables
Age (per one-year increase) Gender, female (versus male) Systolic blood pressure (per 1 mm Hg increase) A1C (per 1% increase) Duration of DM (per one-year increase) Multiple oral antidiabetic medications (versus monotherapy) Insulin: yes (versus no) Smoking: yes (versus no) BMI (per 1 kg/m2 increase) BMI change (per 1% decrease) Total cholesterol (per 1 mg/dL increase) Total cholesterol change (per 1% decrease) Bypass procedure (versus no bypass procedure) Fundus preserved (versus absent fundus)
Univariate analysis
Multivariate analysis
OR (95% CI)
P value
1.011 (.969–1.053) .957 (.382–2.398) .984 (.948–1.020) .560 (.377–.832) 1.009 (.966–1.054) .471 (.199–1.111) .308 (.053–1.777) 1.078 (.462–2.515) 1.186 (1.031–1.364) 1.060 (1.002–1.122) 1.007 (.996–1.019) 1.020 (1.001–1.039) 2.041 (.830–5.018) .461 (.150–1.418)
.620 .925 .380 .004 .690 .085 .188 .863 .017 .042 .296 .040 .120 .177
OR (95% CI)
P value
.479 (.297–.771)
.002
.348 (.123–.989) .184 (.024–1.438)
.048 .107
1.223 (1.030–1.452)
.021
1.024 (1.001–1.048) 3.023 (.989–9.240)
.041 .052
BMI ¼ Body mass index; CI ¼ Confidence interval; DM ¼ Diabetes mellitus; A1C ¼ Glycated hemoglobin; OR ¼ Odds ratio.
Surgical factor for T2D improvement in BMI o35 / Surgery for Obesity and Related Diseases 11 (2015) 1266–1272
were not controlled [12,16]. In this study, patients in the not improved group showed significant decreases in their A1C levels, and these levels remained below the diabetic therapeutic goal of o7.0% 2 years after surgery. These outcomes are thought to be associated with the strict T2D control that was achieved through follow-up assessments taking place at the Endocrinology Department within our organization. Eight out of the 36 patients in the not improved group either increased their antidiabetic medication dosages or added insulin to their therapeutic regimens, whereas almost half of the patients (17/36) showed considerable reductions in their A1C levels without any changes to their antidiabetic medications, which indicated that a substantial number of patients benefited from their metabolic surgery despite being assigned to the not improved group. Thus, although we have drawn conclusions that are similar to those reported in other studies, we also found that several patients in the not improved group had the potential to achieve long-term metabolic benefits from surgery. The patients who were taking multiple medications for their diabetes had difficulties in achieving T2D improvements (OR: .341; 95% CI: .120–.969; P = .043). In the course of T2D, most patients require therapy with multiple medications to achieve their glycemic goals [17,18], and this suggests that the need for multiple medications may correlate with the severity of their T2D. The analysis of the study results in relation to these preoperative factors indicates that the failure of metabolic surgery was consistent with the progression of T2D [19,20]. In relation to T2D deterioration as determined by the A1C levels, the use of multiple oral antidiabetic medications was found to be a major determinant of the patients’ ability to achieve target glucose levels. Several studies have demonstrated that bariatric surgery has greater metabolic effects on patients with short T2D durations [19,20]; however, these studies only considered T2D of relatively short durations. In contrast, this study analyzed patients with chronic T2D, with a mean ⫾ SD duration of 25.6 ⫾ 9.7 years; and, in the present study, the T2D duration was not representative of its severity, as the patients’ T2D was already advanced. It is known that higher BMIs are associated with higher levels of visceral adipose tissue [1,21,22]. In addition, despite having lower absolute BMI values, people from East Asian countries reportedly show the highest accumulations of visceral adipose tissue and the lowest accumulations of deep subcutaneous adipose tissue as adiposity increases [21]. In both sexes, the amount of visceral adipose tissue is a predictive factor for the presence of T2D [1,22]; and accordingly, the patients in the improved group tended to have more ectopic adipose tissue before surgery, even though their BMI was o35 kg/m2. Metabolic surgery resolves issues associated with ectopic adipose tissue by reducing weight more effectively in patients with relatively high BMI, even when o35 kg/m2 [23,24]. Additionally, the total cholesterol levels in the improved group were found to
1271
be reduced during the 2-year follow-up period, which is consistent with the findings from previously reported studies [25–28]. In the present study, the multivariate logistic regression indicated that RYTG was superior to B-I with respect to T2D improvements (OR: .160; 95% CI: .032–.794; P ¼ .025). RYTG has 2 surgical components that differ from BI, namely, the elimination of the gastric fundus and the gastrointestinal bypass procedure. In contrast, there was no significant difference between RYTG and B-II in relation to T2D improvements (OR: .274; 95% CI: .057–1.305; P ¼ .104), which may be due to the fact that these 2 surgical techniques share the gastrointestinal bypass factor. It is well known that, in nonobese diabetic animal models, gastrojejunal bypass can improve glycemic control and is accompanied by significant endocrine alterations [11,12,27,28]. Intestinal bypass can lead to hindgut and foregut effects; and, indeed, it has been speculated that intestinal bypass prevents contact between the short segment of the proximal small intestine and ingested nutrients, which in turn may lead to direct antidiabetic effects [12,27– 29]. Furthermore, an increase in the levels of unabsorbed nutrients in the distal gut can accentuate the secretion of glucagon-like peptide-1, thereby improving glucose homeostasis [27,29]. In contrast, resection of the fundus did not have a significant effect on improving T2D in patients whose BMI was o35 kg/m2. While removing the ghrelinsecreting cells would have reduced the ghrelin level [28,29], the effect of this was not statistically significant in relation to its impact on diabetes in this study, likely owing to the fact that the metabolic effect of the gastrointestinal bypass is more dominant than the loss of appetite in patients with BMI o35 kg/m2. Clinically, bariatric surgery has a considerable influence on weight loss, and it is clearly the most important weight loss factor for patients with BMI 435 kg/ m2 [23,24]. In this study, however, the weight loss effect was found to be lower in patients with BMI o35 kg/m2 than what has been reported in morbidly obese patients [26,27]. This suggests that, as the weight reduction declines, the metabolic effects associated with the weight loss also decline, and that the role of the gastrointestinal bypass itself in metabolic amelioration may become more important [5,30].
Conclusion The results of this study indicate that procedures involving gastrointestinal bypass represent the most effective surgical element for T2D patients with BMI o35 kg/m2. In terms of the metabolic effects of gastrointestinal modifications, patients with low A1C levels and not taking multiple oral antidiabetic medications before surgery were most likely to achieve T2D improvements. In addition, patients with high BMI before surgery tended to benefit
1272
J. Pak et al. / Surgery for Obesity and Related Diseases 11 (2015) 1266–1272
from metabolic surgery, even though they were not morbidly obese. Limitations The study was retrospective in its design and the sample was not randomized; and, therefore, the patients did not undergo gastrointestinal modifications for metabolic reasons alone. In addition, T2D was not constantly controlled using antidiabetic medication before and after surgery. There was a tendency for T2D to be more strictly controlled after surgery, and this was likely owing to the fact that the patients were regularly followed up in the Endocrinology Department within our organization. Despite these limitations, however, we believe that this study contributes to the body of knowledge on the topic by discussing the surgical elements of gastrectomy in the context of their metabolic effects in nonmorbidly obese patients. Disclosures The authors have no commercial associations that might be a conflict of interest in relation to this article. This work was supported by the National Research Foundation of Korea, which was funded by the Korean Government (2010-0024825). References [1] de Lauzon-Guillain B, Balkau B, Charles MA, Romieu I, BoutronRuault MC, Clavel-Chapelon F. Birth weight, body silhouette over the life course, and incident diabetes in 91,453 middle-aged women from the French Etude Epidemiologique de Femmes de la Mutuelle Generale de l'Education Nationale (E3 N) Cohort. Diabetes Care 2010;33:298–303. [2] Willett WC, Dietz WH, Colditz GA. Guidelines for healthy weight. N Engl J Med 1999;341:427–34. [3] Sjöström L, Lindroos AK, Peltonen M, et al. Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. N Engl J Med 2004;351:2683–93. [4] Maggard-Gibbons M, Maglione M, Livhits M, et al. Bariatric surgery for weight loss and glycemic control in nonmorbidly obese adults with diabetes: a systematic review. JAMA 2013;309:2250–61. [5] Chiellini C, Rubino F, Castagneto M, Nanni G, Mingrone G. The effect of bilio-pancreatic diversion on type 2 diabetes in patients with BMI o35 kg/m2. Diabetologia 2009;52:1027–30. [6] Yi HW, Kim SM, Kim SH, et al. Complications leading reoperation after gastrectomy in patients with gastric cancer: frequency, type, and potential causes. J Gastric Cancer 2013;13:242–6. [7] Edge SB, Compton CC. The American Joint Committee on Cancer: the 7th edition of the AJCC cancer staging manual and the future of TNM. Ann Surg Oncol 2010;17:1471–4. [8] American Diabetes Association. Standards of medical care in diabetes–2014. Diabetes Care 2014;37(Suppl 1):S14–80. [9] Buse JB, Caprio S, Cefalu WT, et al. How do we define cure of diabetes? Diabetes Care 2009;32:2133–5. [10] Schauer PR, Burguera B, Ikramuddin S, et al. Effect of laparoscopic Roux-en-Y gastric bypass on type 2 diabetes mellitus. Ann Surg 2003;238:467–84; discussion 84-85.
[11] de Luis D, Domingo M, Romero A, et al. Effects of duodenal-jejunal exclusion on beta cell function and hormonal regulation in GotoKakizaki rats. Am J Surg 2012;204:242–7. [12] Rubino F, Marescaux J. Effect of duodenal-jejunal exclusion in a nonobese animal model of type 2 diabetes: a new perspective for an old disease. Ann Surg 2004;239:1–11. [13] Nocca D, Krawczykowsky D, Bomans B, et al. A prospective multicenter study of 163 sleeve gastrectomies: results at 1 and 2 years. Obes Surg 2008;18:560–5. [14] Bozzetti F, Ravera E, Cozzaglio L, et al. Comparison of nutritional status after total or subtotal gastrectomy. Nutrition 1990;6:371–5. [15] HbA1 c targets in type 2 diabetes: guidelines and evidence. Drug Ther Bull 2013;51:42–45. [16] Lee WJ, Chong K, Ser KH, et al. Gastric bypass vs sleeve gastrectomy for type 2 diabetes mellitus: a randomized controlled trial. Arch Surg 2011;146:143–8. [17] Turner RC, Cull CA, Frighi V, Holman RR. Glycemic control with diet, sulfonylurea, metformin, or insulin in patients with type 2 diabetes mellitus: progressive requirement for multiple therapies (UKPDS 49). UK Prospective Diabetes Study (UKPDS) Group. JAMA 1999;281:2005–12. [18] U.K. prospective diabetes study 16. Overview of 6 years' therapy of type II diabetes: a progressive disease. U.K. Prospective Diabetes Study Group. Diabetes 1995;44:1249–58. [19] Hall TC, Pellen MG, Sedman PC, Jain PK. Preoperative factors predicting remission of type 2 diabetes mellitus after Rouxen-Y gastric bypass surgery for obesity. Obes Surg 2010;20: 1245–50. [20] Lee WJ, Hur KY, Lakadawala M, et al. Predicting success of metabolic surgery: age, body mass index, C-peptide, and duration score. Surg Obes Relat Dis 2013;9:379–84. [21] Nazare JA, Smith JD, Borel AL, et al. Ethnic influences on the relations between abdominal subcutaneous and visceral adiposity, liver fat, and cardiometabolic risk profile: the International Study of Prediction of Intra-Abdominal Adiposity and Its Relationship With Cardiometabolic Risk/Intra-Abdominal Adiposity. Am J Clin Nutr 2012;96:714–26. [22] Sattar N, Wannamethee SG, Forouhi NG. Novel biochemical risk factors for type 2 diabetes: pathogenic insights or prediction possibilities? Diabetologia 2008;51:926–40. [23] Buchwald H, Estok R, Fahrbach K, et al. Weight and type 2 diabetes after bariatric surgery: systematic review and meta-analysis. Am J Med 2009;122(248-56):e5. [24] Long SD, O'Brien K, MacDonald KG Jr., et al. Weight loss in severely obese subjects prevents the progression of impaired glucose tolerance to type II diabetes. A longitudinal interventional study. Diabetes Care 1994;17:372–5. [25] Sjöström L, Peltonen M, Jacobson P, et al. Bariatric surgery and longterm cardiovascular events. JAMA 2012;307:56–65. [26] Buchwald H, Avidor Y, Braunwald E, et al. Bariatric surgery: a systematic review and meta-analysis. JAMA 2004;292:1724–37. [27] Drucker DJ. The role of gut hormones in glucose homeostasis. J Clin Invest 2007;117:24–32. [28] Cummings DE, Weigle DS, Frayo RS, et al. Plasma ghrelin levels after diet-induced weight loss or gastric bypass surgery. N Engl J Med 2002;346:1623–30. [29] Thaler JP, Cummings DE. Minireview: Hormonal and metabolic mechanisms of diabetes remission after gastrointestinal surgery. Endocrinology 2009;150:2518–25. [30] Hickey MS, Pories WJ, MacDonald KG Jr., et al. A new paradigm for type 2 diabetes mellitus: could it be a disease of the foregut? Ann Surg 1998;227:637–43; discussion 43–44.
