2023 Volume 70 Issue 11 Pages 1077-1085
Residual pancreatic endocrine function is important for maintaining metabolic status after pancreatectomy and is closely related to patient nutritional status and prognosis. In contrast to insulin secretion, the significance of glucagon secretion following pancreatectomy remains unclear. In this study, we assessed the changes in pancreatic glucagon secretion during pancreatectomy to determine their pathophysiological significance. We evaluated glucagon and insulin secretion using a liquid meal tolerance test before and after pancreatectomy in patients scheduled to undergo pancreaticoduodenectomy (PD) or distal pancreatectomy (DP). After pancreatectomy, fasting plasma glucagon levels were significantly decreased in both the PD (n = 10) and DP (n = 5) groups (PD: from 18.4 to 10.5 pg/mL, p = 0.037; DP: from 21.0 to 12.1 pg/mL, p = 0.043), whereas postprandial plasma glucagon levels were not changed. In the liquid meal tolerance test after pancreatectomy, 60-min plasma glucagon levels and the area under the curve (AUC) for 0–120 min of PD were significantly higher than those for DP (60-min plasma glucagon: PD 49.0 vs. DP 21.7 pg/mL, p = 0.040; AUC0–120min: PD 4,749 vs. DP 3,564 μg min/mL, p = 0.028). Postoperative plasma glucose, serum insulin, and serum C-peptide levels during the liquid meal tolerance test were not significantly different between the two groups. Although fasting plasma glucagon levels decreased, postprandial glucagon responses were maintained after both PD and DP. The difference in residual meal-stimulated glucagon response between PD and DP suggests that a relative excess of postprandial glucagon is involved in the postoperative nutritional status after PD through its impact on systemic metabolic status.
PANCREATECTOMY, including pancreaticoduodenectomy (PD) and distal pancreatectomy (DP), is performed in patients with periampullary malignant tumors including pancreatic cancers. Despite of its higher invasiveness, operative mortality has gradually decreased to an acceptable level, and the number of long-term survivors has increased [1-5]. However, nutritional impairment adversely affects patient prognosis and quality of life (QOL) in the postoperative period. Impaired pancreatic exocrine function after pancreatectomy can lead to malnutrition via maldigestion and malabsorption and other risks; thus, nutritional interventions were critical for the successful management of patients after pancreatectomy with maintaining the QOL [6-10]. Pancreatectomy includes two different surgical approaches, PD and DP, which have different postoperative anatomical alterations that could influence systemic status. PD involves bypass of the proximal intestine and resection of the pancreatic head parenchyma, whereas DP involves resection of the pancreatic body and tail parenchyma without gastric bypass [1, 2]. In a comparison of postoperative complications, PD was found to be correlated with an increased risk of exocrine insufficiency, whereas DP was correlated with an increased risk of glucose tolerance impairment [7]. As the different surgical approaches, PD and DP, include completely different resection sites and anatomical dislocations, they may induce different post-operative physiological changes in both exocrine and endocrine functions. Thus, it is important to correctly determine post-operative endocrine disorders, including diabetes mellitus, together with classical post-operative complications, including exocrine dysfunction.
On the other hand, orchestrating endocrine hormone secretion is important for the metabolism of various nutrients after digestion and absorption [11]. Glucagon produced and secreted by pancreatic α-cells plays a critical role in systemic energy supply by enhancing hepatic glucose output and amino acid metabolism, while insulin from pancreatic β-cells activates cellular energy uptake and protein synthesis in the target tissue. Properly balanced secretion of glucagon and insulin plays a pivotal role in metabolic turnover, thus contributing to maintaining nutritional status [12-15]. The majority of reports on post-pancreatectomy endocrine dysfunction have focused solely on insulin deficiency. For example, resection of 25%–44% of the pancreatic volume by DP has been reported to lead to the development of glucose impairment [16-18]. Newly diagnosed diabetes mellitus (DM) has been observed after PD, while in some patients DM resolution is also observed after PD [16, 17, 19]. Importantly, glucose intolerance after pancreatectomy is characterized by hyperglycemia and hypoglycemia possibly due to enhanced peripheral insulin sensitivity and glucagon deficiency [20].
Glucagon is a pleiotropic hormone that not only regulates blood glucose but also amino acid metabolism [21-24]. Thus, glucagon is proposed to be closely associated with the nutritional status of post-pancreatectomy patients. However, the effect of pancreatectomy on glucagon secretion was not investigated in detail until recently. Examination of postoperative glucagon secretion will provide a better understanding of the pathophysiology and provide future improvement in the nutritional status of patients after pancreatectomy.
In this study, we attempted to integrate the clinical problems of different surgical approaches for endocrine disorders by evaluating both insulin and glucagon levels together in patients undergoing PD or DP. We evaluated the glucagon secretion in a liquid meal tolerance test before and after pancreatectomy in patients undergoing PD or DP. Although fasting plasma glucagon levels were decreased, postprandial responses of glucagon were maintained after PD and DP, and were elevated in PD than in DP. These differences in residual meal-stimulated glucagon response between PD and DP could affect the systemic metabolic status of the patients, suggesting the postoperative nutritional alteration after PD. These results emphasize the pathophysiological significance of glucagon in addition to insulin in patients undergone pancreatectomy.
Patients with periampullary disease scheduled to undergo PD or DP were enrolled between July 2019 and January 2021 at the Hiroshima City North Medical Center Asa Citizens Hospital. Patients with change of surgical therapy (total pancreatectomy, n = 1), infection (n = 1), insulin dependent (n = 1), and neuroendocrine tumor (n = 1), were all excluded. Physical and biochemical data were collected before and 1 month after pancreatectomy. The study protocol was approved by the local ethics committee of Hiroshima City North Medical Center Asa Citizens Hospital (approval no. 01-2-18) and conducted in accordance with the tenets of the Declaration of Helsinki. All participants provided informed consent to participate in this study.
Liquid meal tolerance testWe measured plasma glucose, serum insulin, serum C-peptide reactivity (CPR), and plasma glucagon levels at baseline and at 30, 60, 120, 180, and 240 min during the liquid meal tolerance test before and 1 month after pancreatectomy. For the liquid meal tolerance test, Peptamen AF® (300 kcal/200 mL; 25% of the energy provided by proteins, 39% by fats, and 36% by carbohydrates; Nestle, Japan) was administered after the patient had fasted overnight. Exogenous insulin infusion and oral hypoglycemic agents were discontinued for more than 12 hours and during the liquid meal tolerance test. Plasma glucagon analyses were performed by SRL, Inc. (Tokyo, Japan). Plasma glucagon concentration was measured using a specific dual-antibody sandwich enzyme-linked immunosorbent assay (ELISA) (Mercodia AB, Uppsala, Sweden/Cosmic Corporation Co., Ltd., Tokyo, Japan), according to the manufacturer’s instructions. The serum insulin concentration was measured using the Lumipulse® Presto Insulin kit (FujirebioInc., Tokyo, Japan).
Glucagon stimulation testIn addition, we measured serum CPR levels at baseline and 6 min after injection of 1 mg glucagon for the glucagon stimulation test as an assessment of β-cell function and to exclude insulin-dependent cases the day after the liquid meal tolerance test before and 1 month after pancreatectomy. The glucagon stimulation test was performed after the patient had fasted overnight. Exogenous insulin infusion and oral hypoglycemic agents were discontinued for more than 12 hours and during the glucagon stimulation test. The change in CPR (ΔCPR) was calculated using the following formula: ΔCPR = 6-min CPR (6min-CPR) – fasting CPR (F-CPR). Serum CPR concentration was measured using an Elecsys® C-peptide kit (Roche Diagnostics GmbH, Mannheim, Germany).
Statistical analysisFor analyzing serum CPR and plasma glucagon levels, which were below the measurement sensitivity (<0.2 ng/mL and <3.5 pg/mL, respectively), were assigned a value of 0.2 ng/mL and 3.5 pg/mL, respectively. Results are expressed as medians and interquartile ranges. The Wilcoxon signed-rank test was used to evaluate changes from before to after pancreatectomy. Differences between PD and DP were assessed using the Mann-Whitney U test. Chi-squared analysis with Yates’ continuity correction was used to examine the difference in the sex between PD and DP. Unless otherwise indicated, differences between median values were considered significant at p < 0.05. All analyses were performed using SPSS 19.0 J (SPSS Inc., Chicago, IL, USA) for Windows.
This study included 15 patients (10 with PD and 5 with DP) (Table 1). There were no significant differences in clinical characteristics between the PD and DP groups. Glucagon secretion is closely related to liver function and disease [25]. In this study, we did not find any statistical correlation between ALT and glucagon levels after pancreatectomy, suggesting that liver function has less of an influence on glucagon secretion and response (data not shown). Patients with obesity and diabetes were included in this study. Out of 15 patients with PD and DP included in this study, 8 had adenocarcinomas (Stage IA, n = 1; Stage IB, n = 1; Stage IIA, n = 4; Stage IIB, n = 2), 3 intraductal papillary mucinous carcinomas (Stage IA, n = 2; Stage IB, n = 1), 3 distal common bile duct carcinomas (Stage IB, n = 3), and 1 duodenal cancer (Stage IA, n = 1).
Clinical characteristics of the study patients
| All (n = 15) | PD (n = 10) | DP (n = 5) | p value (PD vs. DP) | |
|---|---|---|---|---|
| Age, years | 75 (70–76) | 73 (70–76) | 76 (74–80) | 0.124 |
| Male sex, % | 67 | 50 | 100 | 0.175 |
| BMI, kg/m2 | 22.7 (20.2–24.8) | 22.1 (19.8–24.4) | 23.6 (19.5–24.9) | 0.953 |
| AST, IU/L | 19 (17–46) | 27 (17–69) | 17 (16–23) | 0.310 |
| ALT, IU/L | 16 (11–33) | 22 (11–71) | 16 (10–20) | 0.440 |
| Alb, g/L | 3.9 (3.7–4.1) | 4.0 (3.7–4.1) | 3.8 (3.7–4.0) | 0.594 |
| TG, mg/dL | 114 (87–137) | 120 (104–176) | 87 (64–132) | 0.298 |
| Cre, mg/dL | 0.81 (0.65–0.89) | 0.78 (0.60–0.88) | 0.85 (0.71–1.02) | 0.206 |
| eGFR, mL/min/1.73 m2 | 66.9 (56.9–76.6) | 69.0 (61.6–78.9) | 66.9 (55.5–80.8) | 0.859 |
| HbA1c, % | 6.2 (5.3–7.2) | 6.4 (5.5–6.9) | 7.2 (5.5–8.0) | 0.371 |
| CRP, mg/dL | 0.85 (0.04–0.22) | 0.10 (0.05–0.40) | 0.06 (0.03–0.14) | 0.298 |
Abbreviations: PD, pancreaticoduodenectomy; DP, distal pancreatectomy.
Data are shown as median values (interquartile range). P values were determined by Mann-Whitney U test or Chi-squared analysis with Yates’ continuity correction.
We compared the serum CPR levels in the glucagon stimulation test before and after pancreatectomy (Table 2). Serum CPR levels were significantly decreased after PD [F-CPR: from 1.60 (1.35–2.78) to 1.35 (0.53–1.60) ng/mL, p = 0.036; 6min-CPR: from 5.35 (2.75–6.15) to 2.85 (2.05–3.75) ng/mL, p = 0.008; ΔCPR: from 2.35 (1.28–3.90) to 1.50 (0.88–2.60) ng/mL, p = 0.012]. Similarly, in the DP group, postoperative 6min-CPR, and ΔCPR levels were significantly decreased [F-CPR: from 1.50 (0.75–2.60) to 1.20 (0.20–2.00) ng/mL, p = 0.068; 6min-CPR: from 4.50 (1.25–6.25) to 3.40 (0.25–3.65) ng/mL, p = 0.043; ΔCPR: from 3.00 (0.50–3.65) to 1.30 (0.05–2.10) ng/mL, p = 0.043]. We compared the serum CPR levels between the PD and DP groups. There were no statistically significant differences between the two groups.
Serum CPR levels during the glucagon stimulation test
| PD | DP | p value | |||
|---|---|---|---|---|---|
| PD vs. DP | PD Pre vs. Post | DP Pre vs. Post | |||
| Preoperative | |||||
| F-CPR, ng/mL | 1.60 (1.35–2.78) | 1.50 (0.75–2.60) | 0.679 | — | — |
| 6min-CPR, ng/mL | 5.35 (2.75–6.15) | 4.50 (1.25–6.25) | 0.594 | — | — |
| ΔCPR, ng/mL | 2.35 (1.28–3.90) | 3.00 (0.50–3.65) | 0.953 | — | — |
| Postoperative | |||||
| F-CPR, ng/mL | 1.35 (0.53–1.60) | 1.20 (0.20–2.00) | 0.953 | 0.036 | 0.068 |
| 6min-CPR, ng/mL | 2.85 (2.05–3.75) | 3.40 (0.25–3.65) | 1.000 | 0.008 | 0.043 |
| ΔCPR, ng/mL | 1.50 (0.88–2.60) | 1.30 (0.05–2.10) | 0.440 | 0.012 | 0.043 |
Abbreviations: PD, pancreaticoduodenectomy; DP, distal pancreatectomy; CPR, C-peptide immunoreactivity; F-CPR, fasting serum CPR; 6min-CPR, 6-min CPR level after injection of 1mg glucagon; ΔCPR was calculated on the basis of the formula ΔCPR = 6min-CPR - F-CPR.
Data are shown as median values (interquartile range). P values were determined by Mann-Whitney U test.
Although fasting glucose levels were significantly increased after DP [from 91 (82–108) to 120 (97–122) mg/dL, p = 0.043], there was no significant change in postprandial glucose levels after pancreatectomy in either group (Fig. 1A and 1B). There was also no statistically significant difference in postprandial glucose levels between the two groups. In the PD group, although F-CPR and fasting serum insulin levels were significantly decreased after pancreatectomy [F-CPR: from 1.95 (1.28–2.23) to 1.68 (0.80–1.53) ng/mL, p = 0.011; fasting serum insulin: from 5.1 (2.9–7.7) to 3.1 (1.2–4.0) μU/mL, p = 0.005], postprandial CPR and insulin levels were not statistically significantly different (Fig. 1C and 1D). At the same time, in the DP group, the postoperative 180-min serum insulin levels were significantly decreased [from 19.9 (6.1–26.9) to 8.2 (3.5–20.0) μU/mL, p = 0.043; Fig. 1E and 1F]. There were no statistically significant differences in the serum CPR and serum insulin levels between the two groups before and after the pancreatectomy.
Concentrations of plasma glucose, serum CPR, serum insulin, and plasma glucagon during the liquid-meal tolerance test before and after pancreatectomy. Plasma glucose (A), serum CPR (C), and serum insulin (E) concentrations before pancreatectomy. Plasma glucose (B), serum CPR (D), and serum insulin (F) concentrations after pancreatectomy. The solid lines show the concentrations of each factor in the PD group, and dashed lines show the concentrations of each factor in the DP group. Data are presented as median and interquartile range. # p < 0.05: before vs. after pancreatectomy in the PD group; + p < 0.05: before vs. after pancreatectomy in the DP group.
Fasting plasma glucagon levels were significantly decreased in both groups [PD: from 18.4 (8.6–25.4) to 10.5 (5.5–16.2) pg/mL, p = 0.037; DP: from 21.0 (18.7–29.9) to 12.1 (8.1–15.1) pg/mL, p = 0.043; Fig. 2A and 2B]. The postprandial plasma glucagon levels did not differ significantly before and after pancreatectomy in either group. On the other hand, in the PD group, postprandial plasma glucagon levels tended to increase after pancreatectomy, and plasma glucagon levels after 60 min during liquid meal tolerance test were significantly increased in comparison to DP [PD 49.0 (30.9–65.6) vs. DP 21.7 (12.3–35.9) pg/mL, p = 0.040; Fig. 2A and 2B]. Furthermore, the postoperative area under the curve (AUC) for 0–120 min of PD was significantly higher than that of DP [PD 4,749 (3,221–7,058) vs. DP 3,564 (1,097–3,794) μg min/mL, p = 0.028]. However, the postoperative AUC for 0–240 min was not significantly different between the PD and DP groups [PD 8,609 (5,003–12,039) vs. DP 6,725 (2,337–7,203) μg min/mL, p = 0.206].
Concentrations of plasma glucagon during the liquid-meal tolerance test before (A) and after (B) pancreatectomy. The solid lines show the concentrations of plasma glucagon in the PD group, and dashed lines show the concentrations of plasma glucagon in the DP group. Data are presented as median and interquartile range. * p < 0.05: PD vs. DP; # p < 0.05: before vs. after pancreatectomy in PD group; + p < 0.05: before vs. after pancreatectomy in DP group.
Pancreatectomy is a highly invasive surgery that is widely used for the treatment of patients with periampullary disease. In addition to critical perioperative management, postoperative management of complications is also important, as they are closely associated with patient prognosis and maintenance of QOL. Among postoperative complications, both pancreatic endocrine and exocrine insufficiencies are major problems due to impairment in the absorption and metabolism of nutrients and are thus closely associated with poor postoperative performance status. Glucose intolerance after pancreatectomy is characterized by postprandial hyperglycemia and fasting hypoglycemia [20]. Importantly, the incidence of glucose intolerance after pancreatectomy differs depending on the pancreatic resection procedure [14, 26], suggesting that glucose intolerance after pancreatectomy is not simply a result of decreased insulin secretion [20, 27]. Thus, exogenous insulin injection in proportion to food intake is the main treatment approach. However, the appropriate nutritional management of patients presenting with hyperglycemia and hypoglycemia remains insufficient [20].
In contrast, glucagon produced and secreted by pancreatic α-cells plays a critical role in glucose metabolism, similar to insulin [12-14], and is closely related to amino acid metabolism [21-24]. However, the effect of pancreatectomy on glucagon secretion remains unclear. Here, we evaluated glucagon secretion using a liquid meal tolerance test before and after pancreatectomy in patients undergoing PD or DP. The fasting plasma glucagon levels were significantly decreased after both PD and DP. While postprandial responses of glucagon were maintained after both PD and DP, the postprandial peak concentration was significantly elevated in PD compared to DP. In contrast, we did not detect any differences in postprandial plasma glucose levels and serum insulin in the same settings. These differences in glucagon responses independent of glucose and insulin suggest a significant impact on the nutritional and metabolic status of patients with PD and DP.
In this study, fasting plasma glucagon levels were significantly decreased after both PD and DP. Serum CPR levels in the glucagon stimulation test decreased significantly after PD and DP. It is assumed that this decrease is simply due to a reduction in islet cell mass and α-cell mass after pancreatectomy. In contrast, the postprandial peak of plasma glucagon concentration was significantly elevated in patients with PD than in those with DP. Moreover, this could be caused by differences in the residual α-cell mass or structure of the islets at the resection site. However, serum CPR levels during the glucagon stimulation test after pancreatectomy were not significantly different between the PD and DP groups. Therefore, differences in reductions of islet cell mass and also α-cell mass depending on the resection site may have minimally contributed to this difference in glucagon secretion during the liquid meal tolerance tests. Indeed, it has been reported that the relative contribution of β-cells and α-cells to the composition of human islets is high throughout the human pancreas [28].
Pancreatic resection with or without gastric bypass and anatomical alterations can affect pancreatic α-cells and their secretory regulation of glucagon through pancreatic functional alterations, including pancreatic exocrine function and hemodynamics. PD with bypass of the proximal intestine and suture for approximating the pancreatic parenchyma to the jejunum is correlated with an increased risk of exocrine insufficiency [7]. Thus, maldigestion and malabsorption of nutrients may influence glucagon secretion. Furthermore, it is possible that elevated glucagon secretion is caused by changes in the blood flow and nerves. Since abdominal vagus nerve stimulation modulates glucose levels by affecting the secretion of insulin and glucagon [29], resection of the abdominal vagus nerve is also considered to be a cause of the difference in postprandial glucagon secretion between pancreatic resection procedures. However, the insulin secretion status was not altered between the pancreatic resection procedures, suggesting that they are not exclusive factors for glucagon secretion. Alternatively, the degree or pattern of α-cell dysfunction could differ between the groups after pancreatectomy. Frey’s procedure, including both exocrine drainage and resection of the pancreas in participants with chronic calcific pancreatitis, does not affect the β-cell function and insulin resistance but induces α-cell dysfunction after surgery [30]. Although the detailed mechanisms of α-cell dysfunction after PD remain unknown, anatomical dislocation and/or functional alterations due to the surgical technique have been suggested. This study showed no difference in glucagon secretion before and after pancreatectomy. This indicates that the total pancreatic function of glucagon secretion was maintained despite pancreatic resection. However, the post-operative glucagon response differed between the PD and DP surgical approaches, which could mean that the surgeries affected the responsive function of glucagon secretion at different levels. This emphasizes the clinical significance of surgical approaches for pancreatic endocrine function, not only for insulin but also glucagon. Patients who underwent PD exhibited relative hyperglucagonemia, but not significant hyperglycemia, at 60 min in the post-operative meal-challenge test compared to those who underwent DP. Maintaining insulin secretion during hyperglucagonemia could prevent apparent hyperglycemia in the hyperglucagonemic state. Long-term follow-up of both glucagon and insulin secretion, together with glycemic status, is necessary to confirm whether prolonged hyperglucagonemia affects glycemic and nutritional statuses.
Extra-pancreatic factors, such as food digestion and incretin hormones, can be attributed as possible etiologies for altered glucagon secretion. As mentioned above, the difference in impairment of pancreatic exocrine function and the presence of gastric bypass may affect the time variation in food digestion and the absorption of nutrients, thereby affecting the postprandial glucagon response. Additionally, incretin hormones, which regulate the secretion of pancreatic hormones, might also affect the response. Glucose-dependent insulinotropic polypeptide (GIP) produced by K cells in the duodenum and upper jejunum stimulates glucagon secretion, whereas glucagon-like peptide-1 (GLP-1) produced by L cells in the small and large intestines suppresses glucagon secretion [31, 32]. Anatomical alteration of the digestive tract induces nutrients to reach the small intestine in a short time, after which they can drive faster and increase GIP secretion, leading to an increase in the postprandial secretion of glucagon. Indeed, Roux-en-Y, the procedure for gastric bypass, which is also performed in PD, is used in bariatric surgeries, resulting in significant reductions in body weight and increased secretion of GIP and GLP-1 [32]. The increase in postprandial glucagon secretion after PD could have been caused by GIP hypersecretion due to Roux-en-Y reconstruction [33]. In summary, the differences in postprandial glucagon levels between pancreatic resection procedures could be caused by multiple mechanisms. Further studies targeting digestive tract function, enterokinesis, and incretin hormones, as well as insulin and glucagon, are necessary to elucidate the detailed mechanisms underlying the alteration in glucagon secretion. In long-term observations, it had been reported that more significant weight loss was observed in patients undergoing PD than in those undergoing DP [10]. Prolonged greater glucagon secretion worsens post-operative nutritional status through its catabolism-enhancing effects; thus, this could be considered one of the reasons for weight loss. Long-term follow-up of both glucagon secretion and nutritional status, including body weight alteration after pancreatectomy, will clarify the association between glucagon secretion and the underlying mechanisms and provide clues for finding a therapeutic strategy for post-operative malnutrition.
For the following reasons, we performed a meal tolerance test using Peptamen AF® (300 kcal/200 mL, 25% of the energy provided by proteins, 39% by fats, and 36% by carbohydrates; Nestle, Japan), a peptide-based formula. First, in the glucose tolerance test, glucose suppressed glucagon secretion; therefore, it was not suitable for the assessment of secretion capacity. Second, the liquid meal contained more protein and fewer carbohydrate than that of other enteral nutrients at our hospital. In addition to glucose-dependent mechanisms, amino acids are known to enhance glucagon secretion, and glucagon enhances amino acid metabolism in the liver. Some fatty acids also regulate glucagon secretion both positively and negatively [21-24]. Therefore, this formula can prevent a rapid increase in blood plasma glucose levels, which also suppresses glucagon secretion, which was our main focus. Finally, a peptide-based formula was used to avoid the effects of putative exocrine deficiency after pancreatectomy and of meal composition [34]. However, it is not a widely established endocrine function test, therefore, the data and results should be interpreted with caution. It is important to accumulate data from this liquid meal tolerance test under various conditions to validate our findings.
It is reported that postprandial pseudo-hyperglucagonemia could be displayed when the plasma glucagon level was measured by glucagon sandwich ELISA after partial but large pancreatectomy (one-seventh of the pancreas remaining) and total pancreatectomy, possibly due to the cross-reactivity of glucagon antibodies to the remaining gut-derived glicentin [35]. In the current study, we could not exactly evaluate the dissected and remaining volumes of pancreatic parenchyma. However, pancreatic resection lines of all study patients were in the middle of the superior mesenteric vein; therefore, we estimated that the volumes of the remnant pancreas were around 40 mL of the pancreas in both PD and DP groups, according to a previous report (PD: from 68.9 ± 17.5 to 38.7 ± 14.3 mL, DP: from 67.4 ± 21.7 to 43.9 ± 16.4 mL) [36]. Indeed, the absolute values of glucagon in the current study were much larger than those of totally pancreatectomized patients in the previous report but comparable with partially pancreatectomized patients [35], suggesting that our cases still maintained certain pancreatic endocrine capacities for both insulin and glucagon. Considering the cross-reactivity of the glucagon ELISA antibodies used in this study to glicentin, the possibility of glicentin-related pseudohyperglucagonemia should be considered. However, considering the maintained basal and stimulated glucagon secretion from the remaining pancreas and the lower reactivity of the antibody to glicentin than to glucagon, we believe that the impact of glicentin on the measured values is minimal. We also understand that our glucagon results should ideally be reconfirmed by more sensitive liquid chromatography-high-resolution mass spectrometry, but this is difficult mainly due to technical issues so far.
In addition to the above, this study had several limitations. First, the sample size is small. Second, we were unable to examine nutritional status such as amino acid metabolism. Third, long-term follow-up for glucose tolerance was not performed. There was no difference in diabetes prevalence 1 month after surgery in this study. However, differences in glucose tolerance between PD and DP have been observed 6 months after surgery [37]. We believe that long-term follow-up is necessary to clarify the relationship between glucagon secretion and glucose tolerance after pancreatectomy. Studies with large sample sizes and comprehensive evaluations, as stated above, are desirable to confirm these experimental results.
In this study, the postprandial peak concentration of glucagon after pancreatectomy was significantly higher in the PD group than that in the DP group. The primary effects of glucagon are the promotion of hepatic glucose output and amino acid metabolism [21-24]. Thus, glucagon can be considered as a supplier of stored energy to the whole body. The elevated glucagon secretion in PD could compensate for malnutrition caused by exocrine insufficiency [7]. Although this is an appropriate physiological response, a relative excess of glucagon can induce postprandial hyperglycemia and amino acid metabolism, ultimately leading to protein and amino acid complications, such as sarcopenia. Taken together, we propose that increased postprandial glucagon secretion is a potential therapeutic target in patients who have undergone PD.
In conclusion, we evaluated pancreatic endocrine function for both glucagon and insulin using a liquid meal tolerance test before and after pancreatectomy in patients who underwent PD or DP. Although fasting plasma glucagon levels were decreased, postprandial responses of glucagon were maintained after both PD and DP, and were elevated in PD compared to DP. These differences in the residual meal-stimulated glucagon response between PD and DP could impact the nutritional status of patients through imbalanced glucagon and insulin effects. Therefore, from this study we understand that maintaining metabolic status by proper intervention with glucagon in addition to insulin is important to improve post-operative nutritional status, leading to better prognosis.
This work was carried out in cooperation with the Clinical Examination Unit, Hiroshima City North Medical Center Asa Citizens Hospital.
All authors in the manuscript have met the criteria as co-authors including study design, collection and analyses of data, writing the manuscript together with substantial edits and discussions.
All the authors declare no conflicts of interest.
The study protocol was approved by the local ethics committee of Hiroshima City North Medical Center Asa Citizens Hospital (approval no. 01-2-18) and was conducted in accordance with the tenets of the Declaration of Helsinki. Informed consent was obtained from all patients in this study.
