Abstract
Background/Aim: We investigated the postoperative treatment status for diabetes mellitus and perioperative HbA1c levels in patients with diabetes mellitus and examined the effects of clinical factors on the remission of diabetes mellitus. Patients and Methods: In this study, 126 patients with gastric cancer were considered to have diabetes mellitus preoperatively, of whom 79 were treated with oral antidiabetic drugs and/or insulin treatment. We compared diabetic treatment status and HbA1c values between the preoperative and postoperative periods in patients who underwent gastrectomy and examined the effects of clinical factors on improving diabetes mellitus. Results: Of the 79 patients treated preoperatively for diabetes mellitus, 34 (43%) discontinued all medications for diabetes mellitus and for 37 (47%) the therapeutic dose was reduced or switched from insulin to oral antidiabetic drugs. Total gastrectomy was an independent factor for remission of antidiabetic treatments after gastrectomy. Concerning HbA1c levels, only the absence of preoperative insulin use was an independent factor for improvement. However, reconstruction was not a significantly correlated factor for the improvement of postoperative HbA1c levels and reduction of antidiabetic medications after distal gastrectomy. Conclusion: Almost all patients discontinued or had their dose of antidiabetic medications reduced after gastrectomy in clinical practice, and special attention should be paid in the management methods for diabetes mellitus in patients who underwent total gastrectomy for gastric cancer.
Gastric cancer (GC) remains one of the most common cancers in Asia because of the highly carcinogenic subtype of Helicobacter pylori (1) and salty food (2). In more than half GC patients, the disease is detected in early stages due to increased use of upper gastrointestinal endoscopy as a cancer screening modality (3), and their prognoses have improved along with the development of minimally invasive surgery and the improvement of postoperative management (4). Therefore, many patients survive long postoperative periods after surgical treatment.
In contrast, several patients with GC and diabetes mellitus have been recently increasing due to the consumption of high-calorie diets and the lack of exercise (5), which subsequently increased the opportunity of performing gastrectomy for patients with diabetes mellitus. In that context, we have sometimes experienced cases with diabetes mellitus who improve the disease postoperatively after gastrectomy for GC. Similar postoperative remission of glucose tolerance has been reported in patients with obesity who underwent surgical treatments (6, 7). Several mechanisms, such as weight loss due to decreased food intake and changes in the secretion of gastrointestinal hormones, have been considered causes of improvement. Moreover, it has also been reported whether the reconstruction method might affect the glucose tolerance after gastrectomy (8, 9). According to previous reports, gastrojejunostomy [Billroth-II (B-II) and Roux-en-Y (R-Y) methods] are superior to the Billroth-I (B-I) method, and the R-Y method is superior to the B-II method in improving glucose tolerance (8, 9). The reconstruction method after distal gastrectomy for GC, however, is usually mainly selected according to the size of the remnant stomach and/or the presence or absence of hiatal hernia, among others (10). If the clinical status permits, surgeons perform their preferred reconstruction after gastrectomy.
In this study, we investigated the postoperative treatment status for diabetes mellitus and perioperative HbA1c levels in patients with diabetes mellitus and examined the effects of clinical factors on the remission of diabetes mellitus.
Patients and Methods
Patients. At the Yamanashi University Hospital, a total of 668 patients with GC underwent curative gastrectomy with regional lymph node dissection between 2004 and 2015. In this study, patients with diabetes mellitus were defined as those with HbA1c ≥6.5% (NGSP) and those treated with oral antidiabetic drugs and/or insulin, excluding those who received insulin treatment only within the preoperative period. Of the 668 patients, 126 had diabetes mellitus preoperatively, of whom 79 were treated with oral antidiabetic drugs and/or insulin.
These patients principally underwent gastrectomy with lymph node dissection according to the Japanese Gastric Cancer Treatment Guidelines. Distal gastrectomy was performed for those with middle- and/or lower-third GC, and total gastrectomy was performed for those with advanced GC, including the upper-third stomach. In contrast, proximal gastrectomy was performed for those with early upper-third GC. As for lymph node dissection, D1+ lymph node dissection was performed for early GC and D2 dissection for advanced GC. Billroth-I reconstruction (gastroduodenostomy) was our principal reconstructive method after distal gastrectomy; however, R-Y reconstruction (gastrojejunostomy) was selected for cases with extended resection or esophageal hiatal hernia. R-Y reconstruction was performed after total gastrectomy, and esophagogastrostomy reconstruction was performed after proximal gastrectomy.
Correlation between perioperative diabetic status and clinical factors. The clinicopathological features of the patients were reviewed based on data recorded in the hospital database. The Clavien-Dindo classification was used for classifying postoperative complications, and morbidities including surgical and systemic complications, were defined as grades III and higher in the Clavien-Dindo classification. After starting postoperative oral intake, oral hyperglycemic drugs and/or insulin was not restarted if the fasting blood glucose concentration was <140 mg/dl or the casual blood glucose concentration was <200 mg/dl. Remission of diabetic status was judged 1 month after the operation, and the correlation of the remission of diabetic status with various clinical factors was examined in this study. HbA1c levels were also compared between the preoperative period and 3 months after gastrectomy, and the association between the results and various clinical factors were investigated. This study was approved by the Ethics Committee of the Yamanashi University and was performed according to the ethical standards of the Declaration of Helsinki and its later amendments.
Statistical analysis. All continuous data are presented as means±standard error or median values. Statistical analyses were performed using the chi-square test and Student’s t-test. Continuous variables are cut off by a median, and only variables with p<0.10 in the univariate analysis were included in the multivariate analysis. p-Values <0.05 were used to indicate statistical significance. All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria).
Ethics approval and consent to participate. This study was approved by the Ethics Committee of the Yamanashi University and was performed according to the ethical standards of the Declaration of Helsinki and its later amendments. All patients or their authorizers gave written informed consent before the operation.
Results
Diabetes medications after surgery. Patients’ characteristics are shown in Table I. Of the 79 patients under treatment for diabetes mellitus before surgery, 60 were treated with oral antidiabetic drugs only (oral group), 13 were treated with insulin only (insulin group), and six were treated with both oral antidiabetic drugs and insulin (oral+insulin group).
Patient background.
In the oral group, the oral antidiabetic drugs were discontinued postoperatively in 28 patients (47%) and reduced in 31 patients (51%) after gastrectomy. One patient (2%) was treated with the same medications postoperatively as those preoperatively. In the insulin group, insulin treatment was discontinued in four patients (31%) postoperatively, and switched to oral antidiabetic drugs in three patients (23%) after gastrectomy. However, six patients (46%) required the same insulin therapy as that before surgery. In the oral+insulin group, two patients (33%) discontinued all medications, three patients (50%) reduced the dose of insulin or switched to only oral antidiabetic drugs, and one patient (17%) required the same treatment even after gastrectomy as that before the surgery. Overall, 34 of the 79 patients (43%) discontinued all antidiabetic medications, and 37 (47%) patients reduced the therapeutic dose or switched from insulin to oral antidiabetic drugs. However, eight (10%) patients required the same treatment even after gastrectomy as that before the surgery.
Comparison of clinical features according to postoperative antidiabetic treatments. Various clinical factors were compared between the 34 patients who discontinued all antidiabetic medications postoperatively (remission group) and the remaining 45 patients (non-remission group). The extent of gastrectomy was correlated with postoperative antidiabetic treatment, and total gastrectomy was significantly more frequent in the remission group than in the non-remission group (56% vs. 29%; p=0.021). In contrast, no correlation was observed between other factors, including patient, tumor, and surgical factors, and the remission of antidiabetic drugs (Table II). The multivariate analysis also showed the TG was an independent factor for the remission of antidiabetic treatments after gastrectomy.
Analysis of all cases.
Correlation between HbA1c remission and clinical features. Of the 126 patients with preoperative diabetes mellitus included in this study, 73 patients, for whom both preoperative and 3-month postoperative HbA1c levels were available, were examined for correlation between improvement and clinical features. The preoperative non-insulin use was significantly correlated with the improvement in postoperative HbA1c (p=0.036). Except for the preoperative insulin use, patients with low preoperative HbA1c levels and those who underwent total gastrectomy tended to have improved HbA1c levels after gastrectomy. However, multivariate analysis demonstrated that the preoperative non-insulin use was an independent factor for the improvement in postoperative HbA1c levels (Table II).
Analyses of antidiabetic medications and HbA1c levels in patients who underwent distal gastrectomy. Although R-Y reconstruction tended to be correlated with the improvement of postoperative HbA1c levels, the reconstruction method was not an independently correlated factor on the multivariate analyses for the improvement of postoperative HbA1c and the remission of antidiabetic medications after distal gastrectomy. No significant correlations were observed between any clinical factors and the improvements in antidiabetic medications and HbA1c levels after distal gastrectomy (Table III).
Analysis of distal gastrectomy cases.
Discussion
Recent advancements in diagnostic devices and the widespread of cancer screening have increased the early detection of GC. In addition to the early detection of GC, minimally invasive surgery and improvements in perioperative managements have improved the short- and long-term surgical outcomes of patients with GC. However, the aging of the population has increased the opportunity to perform surgery for elderly patients. Such elderly patients often have various comorbidities; subsequently, postoperative follow-up should become a more important challenge not for only GC, but also for related comorbidities. Among various comorbidities, diabetes mellitus, particularly obesity-related diabetes mellitus, has also recently been increasing. These findings prompted us to examine the disease status after gastrectomy in patients with GC and diabetes mellitus.
Glycemic control can be exacerbated by surgical invasiveness in the perioperative period in not only patients with diabetes mellitus but also patients without diabetes mellitus. Perioperative appropriate glycemic control is important, particularly in patients with diabetes mellitus, to decrease various surgical complications, such as surgical site infection and anastomotic leakages, in addition to the complications of diabetes mellitus itself (11). Perioperative glycemic control in patients with GC is conducted similar to that in patients with other cancers; however, we have often experienced that glycemic control would be improved even in patients with diabetes mellitus after gastrectomy, partially due to their postoperative poor oral intake.
This study demonstrated that almost all patients discontinued or had their antidiabetic medication reduced after gastrectomy in clinical practice. The reduction in gastric capacity with/without bypass mainly contributes to improvement in diabetic status via a mechanism similar to bariatric surgery (12). As expected, total gastrectomy was an independent correlated factor for the remission of diabetes after gastrectomy in this study, although no statistically significant difference was observed in the analyses on HbA1c levels.
Other possible mechanisms have been reported to be involved in the improvement, such as changes in the secretion of gastrointestinal hormones, changes in bile acid signaling and intestinal bacterial flora, changes in the gut–brain–liver neural network, and changes in food preferences and eating behavior. Concerning the changes in the secretion of gastrointestinal hormones, two hypotheses have been suggested for the improvement of postoperative good glycemic control in patients after gastrectomy. One is the foregut hypothesis in which gastrectomy with vagal nerve denervation might decrease the secretion of ghrelin, which consequently results in appetite loss, a decrease in growth hormone secretion, and the downregulation of the suppression of insulin secretion (13-15). The other reason is the hindgut hypothesis in which food passage to the ileum is accelerated by gastrectomy, which results in the stimulation of insulin secretion by GLP-1-producing cells located at the terminal ileum (15, 16).
The changes in bile acid signaling and intestinal bacteria are also possibly associated with the improved glycemic control after gastrectomy. Bile acids are converted into various molecules by intestinal bacteria (17). In patients after gastrectomy, changes in the intestinal bacterial flora produce changes in the molecules converted by the bacteria from bile acids, which decreases the efficiency of bile acid reabsorption at the terminal ileum (18). Therefore, the production of bile acids from fatty acids increases and subsequently decreases visceral fat and improves insulin sensitivity (19). The changes in the gut-brain-liver neural network and food preferences and eating behavior are also possibly associated with the improvement. Vagal nerve denervation during gastrectomy might disrupt the gut–brain–liver neural network and decrease the efficiency of food absorption (20), and postoperative nutritional guidance might lead to changes in eating preferences and behaviors. These factors are supposed to be involved in the improvement of postoperative glycemic control in patients after gastrectomy.
Furthermore, the reconstruction method can be associated with the improvement in glucose tolerance. Dietary stimulation is transmitted to ghrelin-producing duodenal cells in the Billroth-I reconstruction method; therefore, the improvement in glucose tolerance may be less effective in the Billroth-I method than that in the Billroth-II and R-Y methods (15). R-Y reconstruction method may have advantages, particularly the earlier stimulation of GLP-1 secretion at the terminal ileum (10). Several scoring systems, such as the ABCD (21) and DiaRem (22) scores, have been reported to be useful in predicting diabetic remission after gastrectomy. From these findings, the likeliness of improving glucose tolerance might be one of the key considerations in determining the reconstruction method in patients with diabetes mellitus and GC who underwent gastrectomy, and the R-Y method may be the most favorable reconstruction method for patients with diabetes mellitus lower remission rate scores (23). In this study, however, the reconstruction method was not a significant factor associated with remission or the improvement in HbA1c levels after distal gastrectomy. No statistically significant difference was observed, which might be due to the small number of patients analyzed in this study; therefore, further large- scale studies should be conducted for definite conclusions.
This study was a retrospective study conducted at a single Institution and has several limitations. First, this study included only patients whose HbA1c values could be compared before and after surgery. It may have caused selection bias. Second, the follow-up period was short and relapse after remission could not be evaluated in this study. Oral intake increases gradually during the postoperative period, and further large-scale studies with long-term follow-up periods are warranted.
In addition, it is possible that diseases treated with steroids (e.g., liver disease as complication), except for diabetes mellitus, may affect glycemic control after gastrectomy. However, there was no evidence that these diseases affect glycemic control after gastrectomy in this study (Data not shown). Further studies are required to clarify the influence of complications excepting diabetes mellitus on glycemic control after gastrectomy.
Conclusion
Most patients with diabetes mellitus discontinued or reduced antidiabetic medication after gastrectomy for GC, and total gastrectomy was an independent factor for the remission of diabetes mellitus. Total gastrectomy is one of the factors that should be considered when determining surgical technique, because postoperative glycemic control is associated with the risk of postoperative complications.
Acknowledgements
The authors would like to express their gratitude to Enago (https://www.enago.jp/) for the expert linguistic services provided.
Footnotes
Authors’ Contributions
Akihito Mizukami was involved in drafting the manuscript. Yoshihiko Kawaguchi designed this study. Daisuke Ichikawa revised the manuscript. All Authors read and approved the final manuscript.
Conflicts of Interest
The Authors declare that they have no competing interests.
- Received June 16, 2023.
- Revision received July 21, 2023.
- Accepted July 25, 2023.
- Copyright © 2023, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).