Abstract
Background
Although gastric cancer patients generally experience drastic weight decrease post‐gastrectomy, the impact of weight decrease on type 2 diabetes risk remains unclear. We investigated the type 2 diabetes risk after gastric cancer surgery according to postoperative weight decrease in gastric cancer survivors in South Korea, the country with the world's highest rate of gastric cancer survival.
Methods
This retrospective nationwide cohort study included gastric cancer surgery recipients between 2004 and 2014 who survived for ≥5 years post‐surgery. We included patients without a history of diabetes at the time of surgery and those who had not received adjuvant chemotherapy before or after the surgery. Postoperative weight loss was defined as the per cent body weight loss at 3 years post‐surgery compared with the baseline. The type 2 diabetes risk was evaluated using Cox regression analyses for five groups of postoperative weight decrease.
Results
In 5618 included gastric cancer surgery recipients (mean age, 55.7 [standard deviation, SD, 10.9] years; 21.9% female; mean body mass index, 23.7 [SD, 2.9] kg/m2), 331 patients (5.9%) developed postoperative type 2 diabetes during follow‐up duration of 8.1 years (median; interquartile range, 4.8 years; maximum, 15.2 years). Compared with those who gained weight post‐surgery, patients with ≥ −15% to < −10% of postoperative weight decrease (hazard ratio, 0.65; 95% confidence interval, 0.49–0.87; P = 0.004) had the lowest type 2 diabetes risk. A non‐linear association occurred between postoperative weight decrease and the type 2 diabetes risk in gastrectomy recipients (Akaike's information criterion [AIC] for non‐linear model, 5423.52; AIC for linear model, 5425.61).
Conclusions
A U‐shaped non‐linear association occurred between the type 2 diabetes risk and postoperative weight decrease in gastric cancer survivors who underwent gastrectomy. The lowest type 2 diabetes risk occurred in patients with ≥ −15% to < −10% of postoperative weight decrease at 3 years.
Keywords: gastrectomy, gastric cancer, type 2 diabetes, weight loss
Introduction
The development of various types of cancer is associated with an increased risk of type 2 diabetes. 1 Gastric cancer patients have a 35% increased risk of developing type 2 diabetes, 1 which contributes to higher mortality in gastric cancer survivors. 2 Five‐year survival rate of gastric cancer has drastically increased—from 12% in the 1950s to 28% in the 2010s in the United States 3 and from 44% in the early 1990s to 77% in the late 2010s in South Korea 4 , 5 —and gastric cancer survivor care has highlighted improvement in non‐cancer mortality as well as cancer mortality. Gastric cancer survivors have a higher likelihood of dying from non‐cancer causes compared with other cancers, and type 2 diabetes is one of the major non‐cancer causes of death. 6 Therefore, guidance in lifestyle management (e.g., weight recommendation) after gastric cancer surgery to prevent type 2 diabetes is imperative in terms of gastric cancer survivor care.
Patients undergoing gastrectomy for cancer have distinct characteristics of glucose homeostasis. First, gastric cancer surgery recipients experience postoperative weight loss due to the combined effect of altered food intake, nutritional malabsorption and systemic endocrinologic changes due to altered foregut anatomy. 7 , 8 Postoperative weight decrease is pronounced until the first 6 months after the surgery, as the patients generally lose 5–15% of their baseline weight; subsequently, the body weight stabilizes or changes even after 12 months or more post‐surgery. 9 , 10 In the general population, intentional weight loss was associated with a reduced risk of type 2 diabetes. 11 There is a paucity of literature investigating the impact of unintentional weight loss after gastric cancer surgery on the risk of type 2 diabetes in gastric cancer survivors.
Second, previous studies have underlined improvements in glycaemic control after gastric cancer surgery in type 2 diabetes patients. Approximately 25% of patients with comorbid type 2 diabetes undergoing gastric cancer surgery experience type 2 diabetes remission, defined as normoglycaemia without hypoglycaemic drugs, within 2 years post‐surgery, more than half of whom did not develop type 2 diabetes for more than 4 years. 12 These beneficial effects of gastric cancer surgery on type 2 diabetes are considered attributable to modified foregut physiology following gastrectomy, resulting in reduced nutritional intake and absorption, weight loss, and alteration in gastric hormones related to glucose homeostasis. 13 , 14 , 15 Modifying the course of type 2 diabetes, even leading to remission, is characteristic of gastric cancer surgery and is distinctive from other gastrointestinal cancer surgeries.
Therefore, we investigated (1) the impact of weight decrease, following gastrectomy for gastric cancer, on the risk of type 2 diabetes, and (2) optimal postoperative weight goal to minimize the risk of type 2 diabetes in gastric cancer survivors undergoing gastrectomy. We used a nationwide population‐based cohort in South Korea, the country with the world's highest rate of gastric cancer survival. 5
Materials and methods
Data source and study population
The National Health Insurance Service (NHIS) is a single insurer managed by the Korean government. The NHIS manages a mandatory universal insurance system that covers the entire Korean population (97% via the health insurance system and 3% via medical aid). Therefore, the NHIS maintains representative population‐based cohort data, including a claims database containing medical information of the entire Korean population based on insurance claims (e.g., patient demographics, disease diagnoses, medical treatments and procedures). The NHIS database was described extensively in a previous study. 16
The NHIS provides biannual health check‐ups for all insured Koreans aged ≥40 years. Among patients who underwent gastrectomy for cancer between 2004 and 2014, 47 032 who underwent health check‐ups within 2 years pre‐surgery and 4 years post‐surgery were identified using the International Classification of Diseases, 10th Revision, Clinical Modification (ICD‐10‐CM) codes for total gastrectomy (Q2533–Q2537, QA533, QA535 and QA536) or subtotal gastrectomy (Q0251–Q0259, Q2594, Q2595, Q2597, Q2598, QA595, QA597 and QA598), and gastric cancer (C16) (Supporting Information, Figure S1 ).
The following patients were excluded: those with pre‐existing diabetes (n = 12 141; with ICD‐10‐CM codes [E11–E14] or antidiabetic medication) within 2 years prior to surgery; those who had fasting plasma glucose (FPG) level ≥ 125 mg/dL (n = 1301); those who died within 5 years post‐surgery (n = 3430); those with a history of cancer other than gastric cancer (n = 12 442); those who received adjuvant chemotherapy 6 months before or after gastrectomy (n = 3946); those who developed type 2 diabetes within 3 years after gastrectomy (n = 7022); and those with any missing variables (n = 1132). Eventually, 5618 patients were enrolled and evaluated for type 2 diabetes development after surgery according to postoperative weight decrease. This study was approved by the institutional review board (No. 2019AN0156), which waived the requirement for informed consent because the customized database was released after de‐identification and anonymization.
Weight decrease after gastrectomy for gastric cancer
We used postoperative weight measured at 3 years (±6 months) after gastrectomy for cancer, as it has been reported that body weight stabilizes ~2–3 years after gastric cancer surgery following drastic weight changes occurring within 1 year post‐surgery. 9 , 10 The lowest weight was selected when it was measured more than once. Postoperative weight decrease (in percentage) was defined as [postoperative weight at 3 years (±6 months) − baseline weight]/baseline weight × 100.
Primary outcome
The primary outcome was the incidence of type 2 diabetes after 3‐year weight decrease following gastrectomy for cancer. Type 2 diabetes was defined as a combination of antidiabetic medications use and ICD‐10‐CM codes E11–E14. Patients were followed up until the date of diagnosis of type 2 diabetes or until 31 December 2017 depending on which one occurred earlier.
Clinical variables
A standardized self‐report questionnaire was used to obtain demographic and lifestyle data. Smoking status was categorized as never, former or current smoker. Individuals who consumed ≥30 and <30 g of alcohol per day were considered heavy and mild drinkers, respectively. Exercise was recorded as days of moderate‐to‐vigorous physical activity for more than 20–30 min/week (none, 1–4 times per week and ≥5 times per week). A low‐income status was defined as a lower 30% level. The FPG was measured after an overnight fast. Blood pressure was measured after at least 10 min of rest in the sitting position and was repeated if the initial measurement was >120/80 mmHg. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in metres, and obesity was defined as BMI ≥ 25 kg/m2 according to the Asia‐Pacific criteria of the WHO guidelines. 17 Hospitals in which these health examinations were performed were certified by the NHIS and subjected to regular quality control. Baseline comorbidities (hypertension, dyslipidaemia, coronary artery disease, stroke, dementia and end‐stage renal disease) were defined using the combinations of ICD‐10‐CM diagnosis, prescription codes or laboratory results. Detailed definitions of the comorbidities are presented in Supporting Information, Table S1 .
Statistical analysis
Summary data were presented as frequencies with percentages for categorical variables and means with standard deviations (SDs) for continuous variables and compared between postoperative weight decrease quintiles using one‐way analysis of variance (ANOVA) test for continuous variables and chi‐square test for categorical variables. Patients were stratified into five categories based on their weight decrease (< −15%, ≥ −15% to < −10%, ≥ −10% to < −5%, ≥ −5% to <0%, and ≥0%), considering that the median postoperative weight change was −7.5%. The incidence rate of type 2 diabetes was calculated by dividing the number of incident type 2 diabetes cases by the total follow‐up duration (person‐years), and hazard ratios (HRs) and 95% confidence intervals (95% CIs) were estimated using multivariate Cox regression analyses for five groups of postoperative weight decrease. The Cox model was adjusted for age, sex, BMI, systolic blood pressure, FPG, smoking status, alcohol consumption, exercise, income status, surgical procedures (total or subtotal gastrectomy) and comorbidities (hypertension, dyslipidaemia, coronary artery disease, stroke, dementia and end‐stage renal disease). Given the competing risks of type 2 diabetes and death in patients undergoing gastrectomy for cancer, a competing risk regression model was considered using the Fine and Gray method. 18 Sensitivity analysis, to avoid the potential effects of reverse causality, was performed after washout type 2 diabetes within 2 years following 3‐year weight decrease after gastrectomy for cancer. Furthermore, we sought to explore the association between baseline BMI and type 2 diabetes after gastrectomy for cancer. Patients were stratified into the following categories based on their baseline BMI per definition of underweight, overweight and obesity applied for East Asians: <18.5 kg/m2, ≥18.5 to <23 kg/m2, ≥23 to <25 kg/m2, ≥25 to <30 kg/m2, and ≥30 kg/m2.
Because of the possible non‐linear association between postoperative weight decrease and the risk of type 2 diabetes identified from Cox regression analyses, we conducted restricted cubic spline regression using three knots placed at the 25th, 50th and 75th percentiles to develop HR curves with postoperative weight decrease as a continuous variable. Akaike's information criterion (AIC) was adopted to compare the linear or non‐linear model's fit. 19 Subgroup analyses were performed to assess whether the association of postoperative weight decrease and type 2 diabetes after gastrectomy for cancer varied among different baseline BMIs (≥18.5 to <23 kg/m2, ≥23 to <25 kg/m2, and ≥25 kg/m2), age groups (<65 and ≥65 years) and surgical procedures (subtotal and total gastrectomy). We also analysed the association between BMI after gastric cancer surgery and the risk of postoperative type 2 diabetes. Patients were stratified into five groups according to postoperative BMIs measured 3 years after surgery: <18.5 kg/m2, ≥18.5 to <23 kg/m2, ≥23 to <25 kg/m2, and ≥25 kg/m2. After identifying significantly low risk of type 2 diabetes in patients with normal BMI measured 3 years after surgery, we performed an additional sensitivity analysis with this population for the association between postoperative weight decrease and the risk of type 2 diabetes. The a priori level of statistical significance was set at P < 0.05. All analyses were two‐tailed and were performed using SAS software Version 9.3 (SAS Institute Inc., Cary, NC, USA).
Results
Baseline characteristics
The mean age of the study population was 55.7 (SD, 10.9) years, 21.9% were female, the mean baseline BMI was 23.7 (SD, 2.9) kg/m2, the mean baseline FPG level was 92.7 (SD, 11.5) mg/dL, and 86.1% of patients underwent subtotal gastrectomy for cancer. Baseline characteristics categorized by postoperative weight decrease are presented in Table 1 . Patients with excessive postoperative weight decrease were more likely to have higher baseline BMI and higher blood pressure, to have never smoked, to not be heavy drinkers and to not perform moderate‐to‐vigorous exercise. Additionally, patients with excessive postoperative weight decrease were more likely to have undergone total gastrectomy and to have hypertension, dyslipidaemia and end‐stage renal disease. The most excessive postoperative weight decrease category (≤ −15%) had the highest proportion of females, were more likely to be older and had the highest mean baseline FPG levels.
Table 1.
Baseline characteristics
| Variables | Weight decrease after gastric cancer surgery (%) | P value | ||||
|---|---|---|---|---|---|---|
| < −15% (n = 886) | ≥ −15% to < −10% (n = 1389) | ≥ −10% to < −5% (n = 1544) | ≥ −5% to <0% (n = 1103) | ≥0% (n = 696) | ||
| Age (years), mean (SD) | 56.2 (11.6) | 55.9 (10.6) | 55.2 (10.4) | 55.5 (10.9) | 55.7 (11.2) | 0.235 |
| Female sex, n (%) | 250 (28.2) | 295 (21.2) | 308 (20.0) | 224 (20.3) | 154 (22.1) | <0.0001 |
| BMI, kg/m2, mean (SD) | 25.0 (3.0) | 24.1 (2.5) | 23.6 (2.6) | 23.0 (2.7) | 22.0 (2.9) | <0.0001 |
| BMI ≥ 25 kg/m2, n (%) | 427 (48.2) | 485 (34.9) | 472 (30.6) | 268 (24.3) | 112 (16.1) | <0.0001 |
| Systolic blood pressure, mmHg, mean (SD) | 126.2 (15.1) | 125.5 (15.7) | 124.4 (14.9) | 124.0 (15.0) | 121.5 (16.0) | <0.0001 |
| Diastolic blood pressure, mmHg, mean (SD) | 78.8 (9.8) | 78.6 (10.4) | 78.1 (10.0) | 77.5 (10.0) | 76.0 (10.6) | <0.0001 |
| FPG, mg/dL, mean (SD) | 93.4 (11.7) | 92.8 (11.1) | 92.9 (11.6) | 92.5 (11.7) | 91.3 (11.1) | 0.007 |
| FPG (100–125 mg/dL), n (%) | 261 (29.5) | 357 (25.7) | 424 (27.5) | 299 (27.1) | 155 (22.3) | 0.019 |
| Smoking status, n (%) | ||||||
| Never | 479 (54.1) | 679 (48.9) | 696 (45.1) | 475 (43.1) | 293 (42.1) | <0.0001 |
| Former | 196 (22.1) | 329 (23.7) | 314 (20.3) | 181 (16.4) | 69 (9.9) | |
| Current | 211 (23.8) | 381 (27.4) | 534 (34.6) | 447 (40.5) | 334 (48.0) | |
| Alcohol consumption, n (%) | ||||||
| Non‐user | 519 (58.6) | 739 (53.2) | 775 (50.2) | 556 (50.4) | 334 (48.0) | <0.0001 |
| Mild drinker | 292 (33.0) | 487 (35.1) | 583 (37.8) | 390 (35.4) | 251 (36.1) | |
| Heavy drinker | 75 (8.5) | 163 (11.7) | 186 (12.1) | 157 (14.2) | 111 (16.0) | |
| Moderate‐to‐vigorous exercise, n (%) | ||||||
| None | 519 (58.6) | 739 (53.2) | 775 (50.2) | 556 (50.4) | 334 (48.0) | <0.0001 |
| 1–4 times per week | 292 (33.0) | 487 (35.1) | 583 (37.8) | 390 (35.4) | 251 (36.1) | |
| ≥5 times per week | 75 (8.5) | 163 (11.7) | 186 (12.1) | 157 (14.2) | 111 (16.0) | |
| Low income, n (%) | 106 (12.0) | 169 (12.2) | 182 (11.8) | 130 (11.8) | 91 (13.1) | 0.926 |
| Surgical procedures, n (%) | ||||||
| Total gastrectomy | 245 (27.7) | 229 (16.5) | 183 (11.9) | 76 (6.9) | 48 (6.9) | <0.0001 |
| Subtotal gastrectomy | 641 (72.4) | 1160 (83.5) | 1361 (88.2) | 1027 (93.1) | 648 (93.1) | |
| Comorbidities, n (%) | ||||||
| Hypertension | 251 (28.3) | 343 (24.7) | 367 (23.8) | 259 (23.5) | 152 (21.8) | 0.029 |
| Dyslipidaemia | 244 (27.5) | 333 (24.0) | 352 (22.8) | 249 (22.6) | 152 (21.8) | 0.039 |
| Coronary artery disease | 19 (2.1) | 28 (2.0) | 30 (1.9) | 29 (2.6) | 13 (1.9) | 0.753 |
| Stroke | 45 (5.1) | 63 (4.5) | 62 (4.0) | 35 (3.2) | 24 (3.5) | 0.195 |
| Dementia | 6 (0.7) | 6 (0.4) | 3 (0.2) | 3 (0.3) | 4 (0.6) | 0.348 |
| End‐stage renal disease | 8 (0.9) | 7 (0.5) | 2 (0.1) | 2 (0.2) | 2 (0.3) | 0.026 |
Note: Postoperative weight decrease was defined as the per cent body weight loss at 3 years post‐surgery compared with the baseline. Abbreviations: BMI, body mass index; FPG, fasting plasma glucose; SD, standard deviation.
Type 2 diabetes risk after gastric cancer surgery according to postoperative weight decrease
There were 331 cases of type 2 diabetes (5.9%) during the follow‐up period (median, 8.1 years; interquartile range, 4.8 years; maximum, 15.2 years). Compared with patients who gained weight at 3 years after gastrectomy for cancer (postoperative weight change ≥ 0%), those who had postoperative weight decrease of ≥ −15% to < −10% (HR, 0.68; 95% CI, 0.51–0.89; P = 0.005) and ≥ −10% to < −5% (HR, 0.65; 95% CI, 0.49–0.87; P = 0.004) had a significantly lower risk of type 2 diabetes after gastrectomy for cancer after full adjustment for covariates (Table 2 ). In the sensitivity analysis after washout incident type 2 diabetes within 2 years following 3‐year weight decrease after gastrectomy for cancer, the significance of the original outcome was maintained (Table S2 ).
Table 2.
Type 2 diabetes risk after gastric cancer surgery according to postoperative weight decrease, and baseline body mass index
| Categories | Total (n) | Event (%) | Person‐years | Incidence rate | Adjusted hazard ratios (95% confidence intervals) | P value |
|---|---|---|---|---|---|---|
| Postoperative weight decrease (%) | ||||||
| ≥0 | 696 | 50 (7.2) | 5815 | 8.60 | [Reference] | |
| ≥ −5 to <0 | 1103 | 84 (7.6) | 9140 | 9.19 | 0.90 (0.69–1.18) | 0.483 |
| ≥ −10 to < −5 | 1544 | 73 (4.7) | 12 653 | 5.77 | 0.68 (0.51–0.89) | 0.005 |
| ≥ −15 to < −10 | 1389 | 62 (4.5) | 11 465 | 5.41 | 0.65 (0.49–0.87) | 0.004 |
| < −15 | 886 | 62 (7.0) | 7275 | 8.52 | 0.89 (0.66–1.20) | 0.451 |
| Baseline body mass index (kg/m2) | ||||||
| <18.5 | 131 | 5 (3.8) | 1053 | 4.75 | 1.32 (0.84–2.06) | 0.221 |
| ≥18.5 to <23 | 2158 | 116 (5.4) | 17 693 | 6.56 | [Reference] | |
| ≥23 to <25 | 1565 | 88 (5.6) | 13 083 | 6.73 | 0.91 (0.74–1.12) | 0.382 |
| ≥25 to <30 | 1650 | 112 (6.8) | 13 620 | 8.22 | 0.97 (0.79–1.19) | 0.804 |
| ≥30 | 114 | 10 (8.8) | 900 | 11.11 | 1.06 (0.60–1.86) | 0.838 |
Note: Postoperative weight decrease was defined as the per cent body weight loss at 3 years post‐surgery compared with the baseline.
Type 2 diabetes risk after gastric cancer surgery according to baseline weight status
When the adjusted risk of type 2 diabetes was evaluated according to baseline BMI quintiles, no baseline BMI interval was significantly associated with the risk of type 2 diabetes after gastrectomy for cancer, compared with normal weight (BMI ≥ 18.5 to <23 kg/m2) (Table 2 ).
Non‐linear association between weight decrease following gastric cancer surgery and type 2 diabetes risk
Given that (1) the HR of type 2 diabetes development after gastric cancer surgery was lowest in patients with postoperative weight decrease of ≥ −15% to < −10% and (2) HR increased with excessive (< −15%) or small postoperative weight decrease (≥ −10%), a cubic spline plot was generated to visualize the association between postoperative weight decrease and type 2 diabetes risk, and a characteristic U‐shaped association was observed (Figure 1 ). The non‐linear model was superior to that of the linear model (AIC for non‐linear model, 5423.52; AIC for linear model, 5425.61). Type 2 diabetes risk after gastric cancer surgery was higher for both with excessive postoperative weight decrease and with small weight decrease with the nadir of the U‐shaped curve around postoperative weight decrease of −15%.
Figure 1.
Cubic spline plot presenting the association between weight decrease following gastric cancer surgery and postoperative type 2 diabetes risk in gastric cancer survivors. Shaded areas represent 95% confidence intervals. BMI, body mass index.
Subgroup analyses
A non‐linear association was observed across all subgroups according to baseline BMI, and no significant interaction was identified (Figure 1 ); significant effect modification and variations in the shapes of the cubic spline plot were observed in subgroups of ≥65 years (P < 0.0001) and total gastrectomy recipients (P < 0.0001) (Figure 2 ). The protective effect of postoperative weight decrease of ≥ −10% to < −5% on type 2 diabetes risk was pronounced in subtotal gastrectomy recipients. However, a significant association between type 2 diabetes risk and postoperative weight decrease was not observed in total gastrectomy recipients.
Figure 2.
Subgroup analyses according to age and surgical procedures: The association between weight decrease following gastric cancer surgery and postoperative type 2 diabetes risk in gastric cancer survivors. Shaded areas represent 95% confidence intervals.
Effects of postoperative body mass index on the risk of type 2 diabetes after gastric cancer surgery
When patients were stratified according to their postoperative BMI measured 3 years after gastric cancer surgery, we observed that those with normal BMI (≥18.5 to <23 kg/m2) had the lowest risk of incident type 2 diabetes after surgery (HR, 0.73, compared with patients with postoperative BMI lower than 18.5 kg/m2; 95% CI, 0.57–0.94; P = 0.016) (Table S3 ). Furthermore, in the sensitivity analysis of patients with normal BMI 3 years after gastric cancer surgery, we consistently observed the lowest risk of type 2 diabetes in patients with postoperative weight decrease of ≥ −15% to < −10% (HR, 0.64; 95% CI, 0.41–0.96; Table S4 ) and a non‐linear association between the postoperative weight decrease and the risk of type 2 diabetes following gastric cancer surgery (Figure S2 ).
Discussion
Cancer survivor care requires cancer‐specific strategies considering the effects of diverse treatment modalities across different cancers in cancer survivors. 20 The foregut is known to be responsible for nutrient absorption and glucose metabolism, 21 and surgical modification of foregut anatomy induces metabolic changes in patients with gastric cancer. 12 , 15 In particular, weight trends after gastrectomy for gastric cancer follow a characteristic course; unintentional weight loss is induced by decrease in nutritional intake and anatomical modification of the foregut even after 12 months or more post‐surgery. 9 , 10 , 22 In this study, we observed a non‐linear U‐shaped association between postoperative weight decrease and type 2 diabetes risk after gastrectomy for gastric cancer and identified that postoperative weight decrease of approximately −15% to −10% maximizes the preventive effect of type 2 diabetes. This association was maintained in the subgroup analysis according to the baseline BMI. This study (1) is the first to identify that weight decrease following gastric cancer surgery affects type 2 diabetes risk and (2) suggests the need for a type 2 diabetes prevention strategy customized for gastric cancer survivors by providing optimal weight decrease intervals.
We identified that the optimal weight decrease following gastrectomy for the prevention of type 2 diabetes in gastric cancer survivors is −15% to −10%. This finding not only provides evidence of the beneficial effects of postoperative weight decrease following gastric cancer surgery but also leads to a change in attitudes regarding weight decrease during postoperative management for patients and their healthcare professionals. Earlier, with the concept of cancer‐related cachexia, involuntary weight decrease in cancer patients was perceived as a malnutrition condition that affects survival and quality of life and needed to be corrected. 23 Several studies have reported postoperative weight decrease as a risk factor for poor outcomes in patients undergoing gastric cancer surgery. 24 , 25 However, two subjects should be considered separately: (1) improving gastric cancer‐related outcomes and (2) preventing chronic diseases in gastric cancer survivors. Considering that chemotherapy is administered to 94.4% of patients with stage II or III gastric cancer in Korea, 26 >95% of the investigated study population was thought to be constituted of patients with stage I gastric cancer, which shows a 5‐year survival rate of >90% in South Korea. Our results regarding the optimal weight decrease interval for prevention of type 2 diabetes were investigated in gastric cancer survivors (and highly expected to be in early stages of cancer), which should be differentiated from the previous studies that investigated cancer‐related outcomes in heterogeneous patients, including those with advanced stages of gastric cancer.
We found a specific interval of weight decrease that minimizes the risk of type 2 diabetes after gastric cancer surgery, which is distinct from previous studies on the association between the risk of type 2 diabetes and surgical or non‐surgical weight loss interventions. 11 , 27 It is widely known that intentional weight loss has protective effects against incident type 2 diabetes. A study from the Diabetes Prevention Program Research Group showed that lifestyle modification targeting at least a 7% weight decrease reduced the risk of type 2 diabetes by 58% (95% CI, 4–66%). 11 Bariatric surgery, which is the most widely performed surgical weight loss intervention, has also been proven to reduce type 2 diabetes risk. 27 However, these studies cannot be directly applied to weight management in gastric cancer survivors to prevent type 2 diabetes, as we found that the optimal postoperative weight decrease is between −15% and −10%, and cannot explain the non‐linear association between weight decrease after gastric cancer surgery and the risk of type 2 diabetes, which was characteristic in our research.
We highlighted the significance of postoperative weight change rather than one‐point weight status measured before or after surgery as a predictor of type 2 diabetes. Baseline BMI has been discussed as a predictor of the remission and relapse of type 2 diabetes in patients undergoing gastric cancer surgery. 12 , 22 However, unlike the higher likelihood of type 2 diabetes remission in patients with higher BMI following gastric cancer surgery, we observed the protective effect of normal BMI on type 2 diabetes (Table S3 ). Furthermore, in a sensitivity analysis of patients with normal BMI after gastric cancer surgery, we found a U‐shaped non‐linear association between the postoperative weight change and the risk of type 2 diabetes (Table S4 and Figure S2 ). We propose that a one‐point body weight measurement cannot fully explain the risk of incident type 2 diabetes after gastric cancer surgery, and postoperative weight change needs to be considered.
This study highlighted the protective effect of weight decrease after gastric cancer surgery on incident type 2 diabetes in gastric cancer survivors, whereas a large portion of the mechanism remains unclear. A recent study observed that an increase in small bowel glucose utilization and white adipose tissue glycolysis, observed after gastric cancer surgery, was correlated with greater weight loss. 15 This may explain the lower risk of type 2 diabetes in the group of patients who achieved postoperative weight decrease after gastric cancer surgery compared with those who did not. However, this cannot explain the increase in the risk of type 2 diabetes observed on the left side of the nadir of the U‐shaped curve. Thus, further studies of the effects of gastrointestinal hormones on type 2 diabetes after gastric cancer surgery are warranted. Changes in gastrointestinal hormones after foregut surgery affect body weight, satiety and insulin resistance, 28 , 29 and duodenal mucosal resurfacing is performed to treat type 2 diabetes, appreciating such roles of the foregut in metabolism. 30
An increase in the risk of type 2 diabetes with excessive weight decrease might be related to malnutrition and sarcopenia, as a decrease in skeletal muscle contributes to insulin resistance, which is well known to predispose patients to type 2 diabetes. 31 Moreover, patients with excessive postoperative weight decrease might have been predisposed to type 2 diabetes with metabolic syndrome components (high baseline BMI, overweight patients, impaired fasting glucose and high prevalence of hypertension and dyslipidaemia [Table 1 ]). This finding suggests a novel hypothesis that baseline metabolic status may characterize excessive weight decrease and increased risk of type 2 diabetes after gastric cancer surgery, as shown on the left side of the U‐shaped curve.
The non‐linear model showed heterogeneity between subgroups according to surgical procedures, and the protective effect of postoperative weight decrease on incident type 2 diabetes after gastric cancer surgery was not observed in patients undergoing total gastrectomy (Figure 2 ). Greater weight loss after gastric cancer surgery in total gastrectomy recipients 9 and a greater decrease in ghrelin, a hormone that controls insulin sensitivity, after total gastrectomy 32 , 33 have been reported in previous observational studies. Gastric peptides, including gastric inhibitory polypeptide and glucagon‐like peptide‐1, which regulate insulin sensitivity, 34 could also contribute to the interaction of the extent of surgery and the association between weight loss and the risk of type 2 diabetes. However, these are insufficient to explain why total gastrectomy recipients did not have protective effects of weight decrease post‐surgery for incident type 2 diabetes. We suggest that postoperative weight management for these populations needs to be performed per type 2 diabetes preventive strategy developed for the general population.
Some limitations of this study need to be acknowledged. First, there might be some cancer‐related confounding factors that may affect weight status. We tried to minimize the effects of chemotherapy or cachexia by excluding patients who died within 5 years post‐surgery and who received chemotherapy. However, excluding patients who had received adjuvant chemotherapy could potentially limit the generalizability of our findings. Further research is warranted to evaluate the association between weight change and the risk of incident diabetes in patients with advanced gastric cancer. Second, the limitations of using claims data, including possible coding errors, misdiagnoses or misclassification, require consideration. For example, when including patients with type 2 diabetes using diagnosis codes and medication prescription records, there is a concern of underestimation of incident type 2 diabetes cases by excluding those who did not receive antidiabetic drugs after diagnosis. Third, although all available variables were used, unavailable potential confounders might not be adjusted due to the retrospective design of the study (e.g., waist circumference and family history of type 2 diabetes). Fourth, reconstruction methods after gastrectomy for cancer were not considered in the analysis because of the unavailability of the variables. However, the extent of gastrectomy is known to play a more dominant role in metabolic effects and postoperative weight decrease than do reconstruction methods. 9 , 35 Fifth, concern remains regarding reverse causation. We performed a sensitivity analysis after washout type 2 diabetes within 2 years following 3‐year weight decrease after gastrectomy for cancer, and the significance of the main outcome was maintained.
This study presented a non‐linear U‐shaped association between postoperative weight decrease and type 2 diabetes risk in gastric cancer survivors undergoing gastrectomy; this association remained in subgroup analyses according to baseline BMI. Our findings suggest that the optimal postoperative weight decrease for incident type 2 diabetes after gastrectomy for cancer is >10% to ≤15% of the baseline weight. A significant prophylactic effect of postoperative weight decrease on incident type 2 diabetes was not observed in total gastrectomy recipients. These results may provide evidence and insights for establishing lifestyle guidelines for gastric cancer survivors to prevent type 2 diabetes.
Conflict of interest
None declared.
Funding
This work was supported by a grant of Korea University Anam Hospital, Seoul, Republic of Korea (for Y. Kwon, K2118501), and a Korea University grant (for S. Park).
Supporting information
Table S1. Definition of Comorbidities
Table S2. Sensitivity Analysis After Exclusion of Type 2 Diabetes Within 5 Years After Gastrectomy for Cancer
Table S3. Risk of type 2 diabetes after gastric cancer surgery according to 3‐year‐postoperative body mass index
Table S4. Sensitivity analysis of patients with normal body mass index (18.5 to 23.0 kg/m2) 3 years after gastric cancer surgery
Figure S1. A flow diagram of the selection of the study population
Figure S2. Association between weight decrease following gastric cancer surgery and the risk of type 2 diabetes in patients with normal postoperative body mass index. Shaded areas represent the 95% confidence intervals
Acknowledgements
The authors of this manuscript certify that they comply with the ethical guidelines for authorship and publishing in the Journal of Cachexia, Sarcopenia and Muscle. 36
Kwon Y., Ha J., Kim D., Hwang J., Park S.‐H., Kwon J.‐W., et al (2023) The association between weight change after gastric cancer surgery and type 2 diabetes risk: A nationwide cohort study, Journal of Cachexia, Sarcopenia and Muscle, 14, 826–834, 10.1002/jcsm.13206
Yeongkeun Kwon and Jane Ha equally contributed to this work.
Contributor Information
Jin‐Won Kwon, Email: jwkwon@knu.ac.kr.
Sungsoo Park, Email: kugspss@korea.ac.kr.
References
- 1. Hwangbo Y, Kang D, Kang M, Kim S, Lee EK, Kim YA, et al. Incidence of diabetes after cancer development: a Korean national cohort study. JAMA Oncol 2018;4:1099–1105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Tseng CH. Diabetes conveys a higher risk of gastric cancer mortality despite an age‐standardised decreasing trend in the general population in Taiwan. Gut 2011;60:774–779. [DOI] [PubMed] [Google Scholar]
- 3. Howlader N, Noone AM, Krapcho M, Miller D, Bres TA, Yu M, et al. SEER Cancer Statistics Review, 1975–2018. National Cancer Institute; 2021. https://seer.cancer.gov/csr/1975_2018/. Accessed 25 September 2021. [Google Scholar]
- 4. Hong S, Won YJ, Park YR, Jung KW, Kong HJ, Lee ES. Cancer statistics in Korea: incidence, mortality, survival, and prevalence in 2017. Cancer Res Treat 2020;52:335–350. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Allemani C, Matsuda T, Di Carlo V, Harewood R, Matz M, Nikšić M, et al. Global surveillance of trends in cancer survival 2000–14 (CONCORD‐3): analysis of individual records for 37 513 025 patients diagnosed with one of 18 cancers from 322 population‐based registries in 71 countries. Lancet 2018;391:1023–1075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Shin DW, Ahn E, Kim H, Park S, Kim YA, Yun YH. Non‐cancer mortality among long‐term survivors of adult cancer in Korea: national cancer registry study. Cancer Causes Control 2010;21:919–929. [DOI] [PubMed] [Google Scholar]
- 7. Jeon TY, Lee S, Kim HH, Kim YJ, Lee JG, Jeong DW, et al. Long‐term changes in gut hormones, appetite and food intake 1 year after subtotal gastrectomy with normal body weight. Eur J Clin Nutr 2010;64:826–831. [DOI] [PubMed] [Google Scholar]
- 8. Lee YH, Han SJ, Kim HC, Hyung WJ, Lim JS, Lee K, et al. Gastrectomy for early gastric cancer is associated with decreased cardiovascular mortality in association with postsurgical metabolic changes. Ann Surg Oncol 2013;20:1250–1257. [DOI] [PubMed] [Google Scholar]
- 9. Davis JL, Selby LV, Chou JF, Schattner M, Ilson DH, Capanu M, et al. Patterns and predictors of weight loss after gastrectomy for cancer. Ann Surg Oncol 2016;23:1639–1645. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Park KB, Kwon OK, Yu W. Midterm body composition changes after open distal gastrectomy for early gastric cancer. Ann Surg Treat Res 2018;95:192–200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Knowler WC, Barrett‐Connor E, Fowler SE, Hamman RF, Lachin JM, Walker EA, et al. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002;346:393–403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Kwon Y, Kwon JW, Kim D, Ha J, Park SH, Hwang J, et al. Predictors of remission and relapse of diabetes after conventional gastrectomy for gastric cancer: nationwide population‐based cohort study. J Am Coll Surg 2021;232:973–981. [DOI] [PubMed] [Google Scholar]
- 13. Kwon Y, Abdemur A, Lo Menzo E, Park S, Szomstein S, Rosenthal RJ. The foregut theory as a possible mechanism of action for the remission of type 2 diabetes in low body mass index patients undergoing subtotal gastrectomy for gastric cancer. Surg Obes Relat Dis 2014;10:235–242. [DOI] [PubMed] [Google Scholar]
- 14. Kim WJ, Kwon Y, Lee CM, Lim SH, Li Y, Wang J, et al. Oncometabolic surgery: emergence and legitimacy for investigation. Chin J Cancer Res 2020;32:252–262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Cho A, Kwon IG, Kim S, Noh SH, Ku CR. Altered systematic glucose utilization after gastrectomy: correlation with weight loss. Surg Obes Relat Dis 2020;16:900–907. [DOI] [PubMed] [Google Scholar]
- 16. Seong CS, Kim YY, Khang YH, Park JH, Kang HJ, Lee H, et al. Data resource profile: the national health information database of the National Health Insurance Service in South Korea. Int J Epidemiol 2017;46:799–800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Seo MH, Lee WY, Kim SS, Kang JH, Kang JH, Kim KK, et al. 2018 Korean Society for the Study of Obesity guideline for the management of obesity in Korea. J Obes Metab Syndr 2019;28:40–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Austin PC, Lee DS, Fine JP. Introduction to the analysis of survival data in the presence of competing risks. Circulation 2016;133:601–609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Burnham KP, Anderson DR. Multimodel inference—understanding AIC and BIC in model selection. Sociol Method Res 2004;33:261–304. [Google Scholar]
- 20. Shapiro CL. Cancer survivorship. N Engl J Med 2018;379:2438–2450. [DOI] [PubMed] [Google Scholar]
- 21. Korner J, Cline GW, Slifstein M, Barba P, Rayat GR, Febres G, et al. A role for foregut tyrosine metabolism in glucose tolerance. Mol Metab 2019;23:37–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Kwon Y, Kwon J‐W, Ha J, Kim D, Cho J, Jeon SM, et al. Remission of type 2 diabetes after gastrectomy for gastric cancer: diabetes prediction score. Gastric Cancer 2022;25:265–274. [DOI] [PubMed] [Google Scholar]
- 23. Huhmann MB, Cunningham RS. Importance of nutritional screening in treatment of cancer‐related weight loss. Lancet Oncol 2005;6:334–343. [DOI] [PubMed] [Google Scholar]
- 24. Climent M, Munarriz M, Blazeby JM, Dorcaratto D, Ramón JM, Carrera MJ, et al. Weight loss and quality of life in patients surviving 2 years after gastric cancer resection. Eur J Surg Oncol 2017;43:1337–1343. [DOI] [PubMed] [Google Scholar]
- 25. Park YS, Park DJ, Lee Y, Park KB, Min SH, Ahn SH, et al. Prognostic roles of perioperative body mass index and weight loss in the long‐term survival of gastric cancer patients. Cancer Epidemiol Biomarkers Prev 2018;27:955–962. [DOI] [PubMed] [Google Scholar]
- 26. Health Insurance Review and Assessment Service . National healthcare quality report for the gastric cancer in 2014. 2016. https://www.hira.or.kr/cms/open/04/04/12/2015_09.pdf. Accessed 25 September 2021. [Google Scholar]
- 27. Booth H, Khan O, Prevost T, Reddy M, Dregan A, Charlton J, et al. Incidence of type 2 diabetes after bariatric surgery: population‐based matched cohort study. Lancet Diabetes Endocrinol 2014;2:963–968. [DOI] [PubMed] [Google Scholar]
- 28. Santarpia L, Pagano MC, Cioffi I, Alfonsi L, Cuomo R, Labruna G, et al. Impaired enterohormone response following a liquid test meal in gastrectomized patients. Ann Nutr Metab 2017;71:211–216. [DOI] [PubMed] [Google Scholar]
- 29. Delzenne N, Blundell J, Brouns F, Cunningham K, De Graaf K, Erkner A, et al. Gastrointestinal targets of appetite regulation in humans. Obes Rev 2010;11:234–250. [DOI] [PubMed] [Google Scholar]
- 30. van Baar AC, Beuers U, Wong K, Haidry R, Costamagna G, Hafedi A, et al. Endoscopic duodenal mucosal resurfacing improves glycaemic and hepatic indices in type 2 diabetes: 6‐month multicentre results. JHEP Rep 2019;1:429–437. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Larsen BA, Wassel CL, Kritchevsky SB, Strotmeyer ES, Criqui MH, Kanaya AM, et al. Association of muscle mass, area, and strength with incident diabetes in older adults: the Health ABC Study. J Clin Endocrinol Metab 2016;101:1847–1855. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Soleyman‐Jahi S, Abdirad A, Fallah AA, Ghasemi S, Sadeghi F, Heidari R, et al. Prognostic significance of preoperative and postoperative plasma levels of ghrelin in gastric cancer: 3‐year survival study. Clin Transl Gastroenterol 2017;8:e209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Takachi K, Doki Y, Ishikawa O, Miyashiro I, Sasaki Y, Ohigashi H, et al. Postoperative ghrelin levels and delayed recovery from body weight loss after distal or total gastrectomy. J Surg Res 2006;130:1–7. [DOI] [PubMed] [Google Scholar]
- 34. Vigneshwaran B, Wahal A, Aggarwal S, Priyadarshini P, Bhattacharjee H, Khadgawat R, et al. Impact of sleeve gastrectomy on type 2 diabetes mellitus, gastric emptying time, glucagon‐like peptide 1 (GLP‐1), ghrelin and leptin in non‐morbidly obese subjects with BMI 30–35.0 kg/m2: a prospective study. Obes Surg 2016;26:2817–2823. [DOI] [PubMed] [Google Scholar]
- 35. Peng D, Cheng YX, Zhang W. Does Roux‐en‐Y construction really bring benefit of type 2 diabetes mellitus remission after gastrectomy in patients with gastric cancer? A systematic review and meta‐analysis. Diabetes Ther 2020;11:2863–2872. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. von Haehling S, Morley JE, Coats AJS, Anker SD. Ethical guidelines for publishing in the Journal of Cachexia, Sarcopenia and Muscle: update 2021. J Cachexia Sarcopenia Muscle 2021;12:2259–2261. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1. Definition of Comorbidities
Table S2. Sensitivity Analysis After Exclusion of Type 2 Diabetes Within 5 Years After Gastrectomy for Cancer
Table S3. Risk of type 2 diabetes after gastric cancer surgery according to 3‐year‐postoperative body mass index
Table S4. Sensitivity analysis of patients with normal body mass index (18.5 to 23.0 kg/m2) 3 years after gastric cancer surgery
Figure S1. A flow diagram of the selection of the study population
Figure S2. Association between weight decrease following gastric cancer surgery and the risk of type 2 diabetes in patients with normal postoperative body mass index. Shaded areas represent the 95% confidence intervals