Empagliflozin Normalizes Fasting Hyperglycemia and Improves Postprandial Glucose Tolerance in Totally Pancreatectomized Patients: A Randomized, Double-Blind, Placebo-Controlled Crossover Study

OBJECTIVE

Insulin remains the only glucose-lowering treatment modality recommended for totally pancreatectomized patients. We investigated the effects of the sodium-glucose cotransporter 2 inhibitor empagliflozin on fasting and postprandial glucose concentrations in pancreatectomized patients and matched healthy control participants.

RESEARCH DESIGN AND METHODS

In a randomized, double-blind, placebo-controlled crossover study, 10 pancreatectomized patients and 10 matched control participants underwent two 3-h liquid mixed meal tests preceded by two doses of 25 mg empagliflozin (administered the night before and in the morning of the meal test) or placebo, respectively. Basal insulin was administered as usual, but bolus insulin was omitted before the meal test during experimental days.

RESULTS

Compared with placebo, empagliflozin lowered fasting plasma glucose (5.0 ± 0.4 vs. 7.9 ± 0.9 mmol/L [mean ± SEM], P = 0.007) and postprandial plasma glucose excursions as assessed by baseline-subtracted area under the curve (1,080 [733; 1,231] vs. 1,169 [1,036; 1,417] pmol/L × min [median (25th and 75th percentiles)], P = 0.014) in the pancreatectomized patients. In the control participants, empagliflozin lowered fasting plasma glucose compared with placebo (5.1 ± 0.1 vs. 5.5 ± 0.1 mmol/L, P = 0.008) without affecting postprandial glucose excursions significantly. The pancreatomy group exhibited greater postprandial glucagon excursions compared with the control group on both experimental days (P ≤ 0.015); no within-group differences between days were observed.

CONCLUSIONS

Empagliflozin administered the day before and immediately before a standardized liquid mixed meal test normalized fasting hyperglycemia and improved postprandial glucose tolerance in pancreatectomized patients.

Pancreaticoduodenectomy with preoperative total embolization of the hepatic arteries (PD-HAE)-a novel treatment with sacrifice of the hepatic arterial blood supply without the need for arterial reconstruction

PURPOSE

Tumors with involvement of common hepatic and gastroduodenal arteries (CHA and GDA) or GDA and the proper hepatic artery (PHA) are traditionally considered nonresectable. We have devised a new procedure that includes pancreaticoduodenectomy with preoperative hepatic artery embolization (PD-HAE) to facilitate an R0 resection of tumors involving the hepatic arteries without vascular anastomoses and complete sacrifice of normal hepatic arterial blood supply.

METHODS

To allow resection of the hepatic arteries, preoperative embolization of the PHA was performed to induce an increased collateral arterial blood flow from the periphery of the liver, far from the hepatic hilum 10-14 days prior to the operation. Between May 1, 2017 and December 31, 2019, eight patients with ductal adenocarcinoma were operated with the PD-HAE procedure.

RESULTS

The embolizations were uneventful apart from a transient marginal elevation of alanine aminotransferase in three patients. All patients had N disease with perineural invasion of tumor cells around the adventitia of the artery and severe perivascular inflammation. An R0 resection (> 1.0 mm to all resection margins) was obtained in six patients (75%). Mean hospital stay was 12 days. Median survival was 23 months (95% CI: 19.5-26.5 months). Six patients (75%) are still alive 11 to 36 months after the operation. There was perioperative fatality, and morbidity was comparable to standard pancreaticoduodenectomy.

CONCLUSION

PD-HAE is a safe procedure and may provide the opportunity for curative resection in otherwise unresectable patients. However, larger studies are needed to evaluate this procedure.

Postoperative and long-term survival in relation to life-expectancy after pancreatic surgery in elderly patients (cohort study)

Background

An evaluation of the outcome after pancreatic surgery with focus on post-operative and late survival in elderly patients was performed.

Methods

The study included 1.556 patients from a single HBP unit operated from 1. January 2010 to 31. December 2019. Patients were divided into two cohorts, < 75 years (n = 1.296) and ≥75 years (n = 260). Post-operative outcome was evaluated in all patients and late outcome in patients with adenocarcinoma in the pancreas (n = 765) and the duodenum (n = 117). The follow-up of patients with benign disease and adenocarcinoma was 57.95 (12.1–132.7) and 39.85 (12.0–131.7) months, respectively.

Results

Length of hospital-stay and surgical complications were not significantly different in the two cohorts, but in-hospital death was 1.1% (<75 years) and 3.5% (≥75 years) (p = 0.008). The median overall survival of adenocarcinoma was 29.7 (<75 years) and 24.3 months (≥75 years) (p = 0.3228) with a one, two, and five-years survival of 74.5%, 56.6% and 28.6% vs. 73.6%, 51.1%, and 25.5%. Median time to relapse (46.2% of patients <75 years and 40.5% of patients ≥75 years) was 9 (1 - 51) and 8 (1 - 78) months (p = 0.534), respectively. Adjuvant chemotherapy did not have impact on the survival of the old cohort. Patients who died during the observation period had lost 94% (<75 years) and 87% (≥75 years) of expected remnant life. Estimated years lost in the old cohort was 4.2 in males and 4.9 in females (p = 0.025)

Conclusion

Elderly patients may undergo pancreatic surgery with a low mortality and for adenocarcinoma with an acceptable long-term survival.

•

Surgery is the only potentially curable treatment to pancreatic cancer.

•

Elderly patients may tolerate pancreatic surgery with low mortality.

•

Radical pancreatic surgery improves long-term survival, also in elderly patients.

•

Operability should be evaluated from morbidity and biological age.

•

Pancreatic surgery in elderly should only be performed in high volume centers.

Glycemic Control and Variability of Diabetes Secondary to Total Pancreatectomy Assessed by Continuous Glucose Monitoring

CONTEXT

The extent of the glycemic variability in diabetes secondary to total pancreatectomy is not fully understood.

OBJECTIVE

To evaluate glycemic variability in totally pancreatectomized (PX) patients and compare it to glycemic variability in hemoglobin A1c (HbA1c)-matched patients with long-standing type 1 diabetes (T1D).

DESIGN

A case-control study was performed.

SETTING

Center for Clinical Metabolic Research, Gentofte Hospital, University of Copenhagen, Hellerup, Denmark.

PATIENTS OR OTHER PARTICIPANTS

Ten PX patients (mean [SEM]: age 64.3 [9.8] years; body mass index (BMI) 34.4 [5.0] kg/m2; duration of diabetes 3 [2.8] years), 10 HbA1c-matched patients with T1D (63.9 [8.6] years; 24.6 [3.1] kg/m2; 22 [4] years), and 10 gender-, age-, and BMI-matched healthy controls. All patients were managed on multiple daily injections of insulin.

INTERVENTION

Continuous glucose monitoring (CGM) (Medtronic MiniMed iPro 2) during 12 consecutive days.

MAIN OUTCOME MEASURES

Glycemic variability.

RESULTS

HbA1c levels were similar in the PX group and the T1D group. The PX group had greater continuous overall net glycemic action per 60 minutes (CONGA60 min) compared with the T1D group (mean [SEM]: 9.5 [0.3] vs 8.3 [0.2] mmol/L, P < 0.003) and mean plasma glucose values were higher in the PX group (10.6 [0.9] vs 9.0 [0.9] mmol/L, P < 0.001), whereas coefficient of variation for plasma glucose and standard deviation of mean plasma glucose, respectively, were similar in the 2 groups. Time spent below range was not different between the PX and the T1D group (2.3 [0.8] vs 4.5 [0.8]%, P = 0.065), whereas time spent above range was higher in the PX group (51.4 [3.3] vs 37.6 [1.9]%, P < 0.001).

CONCLUSIONS

CGM-assessed glycemic variability showed higher CONGA60 min and time spent above range in our PX patients compared with HbA1c-matched T1D patients. This study is registered at www.ClinicalTrials.gov (NCT02944110).

Spontaneous intestinal bleeding due to pseudoaneurism of the gastroduodenal artery: case report of a rare complication to median arcuate ligament syndrome

Median arcuate ligament syndrome (MALS) is the compression of the celiac artery (CA) by the median arcuate ligament. MALS can cause pseudoaneurysm of the gastroduodenal artery, which can lead to fatal bleeding. A 40-year-old male with no prior medical history presented with symptoms of upper gastrointestinal hemorrhage (UGIH). Severe duodenal bleeding was confirmed although endoscopic hemostasis was impossible and final hemostasis was achieved following a subsequent open duodenotomy. A postoperative computed tomographic angiography (CTA) visualized a significant CA stenosis, post-stenotic dilatation and an aneurysm on a jejunal branch artery. The patient underwent coiling of the gastroduodenal artery, gastroepiploic artery and two pancreaticoduodenal arterial branches. The patient was diagnosed with MALS and 6 months later underwent open resection of the median arcuate ligament. MALS should be considered as a rare cause of upper gastrointestinal bleeding. The literature and proposed treatments are discussed.

[Minimally invasive surgery for malignant diseases in pancreas]

Unlike most other abdominal procedures, pancreatic resection for malignant tumours is still predominantly performed as open surgery. However, recent published randomised trials suggest that a laparoscopic approach is safe and may offer advantages in the early postoperative period. Likewise, early reports of robot-assisted pancreatic resection suggest advantages. This reveiw describes the current status of minimally invasive pancreatic surgery for malignant tumours, including both laparoscopic and robot-assisted approaches for pancreaticoduodenectomy and distal pancreatectomy.

Preoperative TruCulture® whole blood cytokine response predicts post-operative inflammation in pancreaticoduodenectomy patients-A pilot cohort study

Major surgery is associated with substantial morbidity and mortality with early post-operative adverse events (POAE) occurring in 30% of patients within the first 30 days. The underlying pathogenesis is multifactorial, including immune dysfunction and increased inflammatory response to surgery. We investigated preoperative immune function by the TruCulture® whole blood technique in a cohort of patients undergoing pancreaticoduodenectomy (PD), hypothesizing that patients developing inflammatory POAE defined as leucocytosis, fever or high (above median) area under the curve (AUC) C-reactive protein (CRP) the first post-operative week would display perturbed preoperative immune function. Sixty-two adult patients were screened, 30 included and 11 excluded post-inclusion due to other surgical procedures than PD and post-operative complications directly attributed to surgery, leaving 19 patients for analysis of preoperative immune function. Patients developing leucocytosis (n = 5, 26%) had lower Toll-like receptor (TLR)-3-stimulated IL-12p40 and higher Candida Albicans (TLR1/2/4/6, Dectin-1)-stimulated TNF-α, compared to patients without leucocytosis (all P < .05). Patients developing fever (n = 7, 37%) had lower TLR7/8-stimulated IFN-γ and patients with high AUC CRP (n = 9, 47%) had lower TLR3-stimulated IFN-γ and IL-6 and lower TLR7/8-stimulated IL-10 (all P < .05), compared to patients without fever or low CRP, respectively. In conclusion, patients with inflammatory POAE displayed lower preoperative stimulated IL-12p40, IFN-γ, IL-6 and IL-10 and higher TNF-α response, compared to patients without inflammatory POAE. This finding suggests that TruCulture is a feasible immunologic screening tool in surgical patients, with a potential for preoperative identification of patients at increased risk for inflammatory POAE, allowing for risk-based intervention trials.

The GLP-1 receptor agonist lixisenatide reduces postprandial glucose in patients with diabetes secondary to total pancreatectomy: a randomised, placebo-controlled, double-blinded crossover trial

AIMS/HYPOTHESIS

Treatment of diabetes secondary to total pancreatectomy remains a challenge and insulin constitutes the only glucose-lowering treatment for these patients. We hypothesised that the glucagon-like peptide 1 (GLP-1) receptor agonist lixisenatide would improve postprandial glucose tolerance in totally pancreatectomised patients.

METHODS

In a double-blinded, randomised, crossover study, 12 totally pancreatectomised individuals (age: 65.0 ± 9.5 mean±SD years; BMI: 22.9 ± 3.9 kg/m²) and 12 healthy control individuals (age 66.1 ± 7.6 years; BMI: 24.0 ± 2.9 kg/m²) underwent two 3 h liquid mixed-meal tests (with paracetamol for assessment of gastric emptying) after single-dose injection of 20 μg of lixisenatide or placebo. Basal insulin was given the night before each experimental day; no insulin was given during study days.

RESULTS

Compared with placebo, lixisenatide reduced postprandial plasma glucose excursions in the pancreatectomy group (baseline-subtracted AUC [bsAUC] [mean±SEM]: 548 ± 125 vs 1447 ± 95 mmol/l × min, p < 0.001) and in the control group (-126 ± 12 vs 222 ± 51 mmol/l × min, p < 0.001). In the pancreatectomy group a mean peak glucose concentration of 23.3 ± 1.0 mmol/l was reached at time point 134 ± 11 min with placebo, compared with a mean peak glucose concentration of 18 ± 1.4 mmol/l (p = 0.008) at time point 148 ± 13 min (p = 0.375) with lixisenatide. In the control group a mean peak concentration of 8.2 ± 0.4 mmol/l was reached at time point 70 ± 13 min with placebo, compared with a mean peak concentration of 5.5 ± 0.1 mmol/l (p < 0.001) at time point 8 ± 25 min (p = 0.054) with lixisenatide. Lixisenatide also reduced gastric emptying and postprandial glucagon responses in the pancreatectomy group (66 ± 84 vs 1190 ± 311 pmol/l × min, p = 0.008) and in the control group (141 ± 100 vs 190 ± 100 pmol/l × min, p = 0.034). In the pancreatectomy group, C-peptide was undetectable in plasma. In the control group, postprandial plasma C-peptide responses were reduced with lixisenatide (18 ± 17 vs 189 ± 31 nmol/l × min, p < 0.001).

CONCLUSIONS/INTERPRETATION

The GLP-1 receptor agonist lixisenatide reduces postprandial plasma glucose excursions in totally pancreatectomised patients. The mode of action seems to involve deceleration of gastric emptying and reduced postprandial responses of gut-derived glucagon.

TRIAL REGISTRATION

ClinicalTrials.gov NCT02640118.

FUNDING

This study was funded by an unrestricted investigator-initiated study grant from Sanofi. Support was also received from from the Novo Nordisk Foundation Center for Basic Metabolic Research, the A.P. Møller Foundation for the Advancement of Medical Science and the Faculty of Health and Medical Sciences, University of Copenhagen.

Evidence of Extrapancreatic Glucagon Secretion in Man

Glucagon is believed to be a pancreas-specific hormone, and hyperglucagonemia has been shown to contribute significantly to the hyperglycemic state of patients with diabetes. This hyperglucagonemia has been thought to arise from α-cell insensitivity to suppressive effects of glucose and insulin combined with reduced insulin secretion. We hypothesized that postabsorptive hyperglucagonemia represents a gut-dependent phenomenon and subjected 10 totally pancreatectomized patients and 10 healthy control subjects to a 75-g oral glucose tolerance test and a corresponding isoglycemic intravenous glucose infusion. We applied novel analytical methods of plasma glucagon (sandwich ELISA and mass spectrometry-based proteomics) and show that 29-amino acid glucagon circulates in patients without a pancreas and that glucose stimulation of the gastrointestinal tract elicits significant hyperglucagonemia in these patients. These findings emphasize the existence of extrapancreatic glucagon (perhaps originating from the gut) in man and suggest that it may play a role in diabetes secondary to total pancreatectomy.

Mechanical properties and collagen content differ between isolated guinea pig duodenum, jejunum, and distal ileum

We compared the stress-strain distributions obtained from isolated segments of the guinea pig duodenum, jejunum, and distal ileum, and the relation between the elastic properties and the collagen content. The segments were immersed in Krebs-Ringer solution containing 10(-2) M MgCl2 to abolish contractile activity. Stepwise inflation of an intraluminal balloon in which the cross-sectional area (CSA) was measured provided the luminal pressure-loading stimulus. The wall thickness was measured by means of 20-MHz A-mode ultrasound. The stress-strain and the incremental elastic modulus-strain distributions were derived from the steady-state values of wall thickness, internal radius, and applied pressure. The CSA-pressure relations and the wall thickness-pressure relations were nonlinear and both differed between the segments (P < 0.01). The stress-strain distributions showed an exponential behavior that fitted well to the equation Y = a x Exp(b x X) (r2 = 0.97 +/- 0.01). The intercept with they axis (a) and the slope of the curves (b) differed between the segments (P < 0.01 and P < 0.05). The collagen contents were 3.99 +/- 0.18 microg/mg, 2.51 +/- 0.13 microg/mg, and 2.10 +/- 0.11 microg/mg in the duodenum, jejunum, and distal ileum, respectively. This difference was significant among all three locations (P > 0.05). An association was found between the collagen content and the incremental elastic modulus (stiffness) at a stress level of 70 kPa (P < 0.05).

Impedance planimetric characterization of esophagus in systemic sclerosis patients with severe involvement of esophagus

This study was designed to evaluate the distensibility and secondary peristalsis of the esophagus in patients suffering from systemic sclerosis with severe esophageal involvement. Balloon distension with impedance planimetric measurement of luminal cross-sectional area was done 7 and 15 cm above the lower esophageal sphincter in 13 patients and nine healthy controls. The controls were studied both with and without receiving the anticholinergic drug butylscopolamine. The cross-sectional area--pressure relations were nonlinear with the largest cross-sectional area in patients at both measuring sites when compared to controls (P < 0.001). The anticholinergic drug butylscopolamine increased the cross-sectional area in controls (P < 0.001). The cross-sectional area distensibility, defined as CSA0(-1) delta CSA delta P-1 did not differ between patients and controls. Balloon distensions elicited contractions proximal to the distension site. The amplitude and frequency of contractions at the distal distension site were significantly reduced in the patients when compared to the controls (P < 0.05). In conclusion, the distal esophagus is most severely affected in patients with systemic sclerosis with increased cross-sectional area and impaired peristalsis.

Dimensions and mechanical properties of porcine aortic segments determined by combined impedance planimetry and high-frequency ultrasound

The aim of the study is to devise a method for direct measurements of pressure, cross-sectional area (CSA) and wall thickness of a vessel when pressurised. Segments of porcine descending thoracic aorta from 22 and 35 kg pigs (in groups 1 and 2, respectively) and of abdominal aorta from 35 kg pigs are studied in vitro. Impedance planimetry provides measurements of the luminal pressure and CSA of the aorta. The wall thickness is measured simultaneously by means of 20 MHz A-mode ultrasound. The pressure, CSA and wall thickness are registered at different pressure at steady state. At maximum pressure, the CSAs are 107 +/- 10, 215 +/- 19 and 257 +/- 17 mm2 in the abdominal, group 1 and 2 segments, respectively. There is a difference between the abdominal group and groups 1 and 2 (p < 0.05), but not between group 1 and 2 segments (p > 0.2). At maximum pressure, the wall thickness is 0.52 +/- 0.04, 0.79 +/- 0.07 and 1.07 +/- 0.02 mm, in the abdominal group 1 and 2 segments, respectively (p < 0.05). The circumferential stress-strain relationship accorded well with stress = a x exp(b x strain). The stress-strain curve of the abdominal aortas is shifted to the left, indicating an increase (p < 0.05).

Static elastic wall properties of the abdominal porcine aorta in vitro and in vivo

OBJECTIVES

To characterise possible differences in vascular mechanics between in vitro and in vivo settings.

DESIGN

Experimental animal study.

MATERIAL

Segments of abdominal aorta from five 30kg pigs were studied in vitro and five porcine abdominal aortas were investigated in vivo.

METHODS

Vessel luminal cross-sectional area (CSA) was assessed in vitro and in vivo by measuring the electrical impedance of fluid inside a distensible balloon mounted on a catheter. Wall thickness was measured in vitro by means of 20 MHz A-mode ultrasound. The obtained values in vitro were used to calculate wall thickness in vivo.

RESULTS

The CSAs were larger in vitro than in vivo (p < 0.05). All segments showed stress-strain relations that accorded well with the exponential equation: Stress = a.exp(b. Strain). The y-axis intercepts (a) and the slope of the curves (b) were different in vitro and in vivo (p < 0.05). Differences were found between the aorta in vivo and in vitro regarding the circumferential stress-strain relations, i.e. the stiffest abdominal aorta was found in vivo.

CONCLUSIONS

The differences in dimensions and wall stiffness may be attributed to the tethering to the surrounding tissues in vivo. Impedance planimetry combined with high frequency ultrasound is an accurate way of measuring segmental vascular elasticity in vivo.

In vivo assessment of luminal cross-sectional areas and circumferential tension-strain relations of the porcine aorta

The objectives were to measure the pressure--cross-sectional area relations and intrinsic stiffness of the porcine aorta in vivo.

METHODS

Measurements were made in 12 pigs, weight 30, 50 or 75 kg, proximal and distal to the bifurcation of the renal arteries, using an electrical impedance system which was located inside a balloon mounted on a catheter. Vessel cross-sectional area (CSA) was assessed by measuring impedance of the fluid inside the balloon during distension.

RESULTS

In vitro testing demonstrated the accuracy and reproducibility of impedance planimetry. In vivo steady-state CSA values showed non-linear relationship with rising distension pressures. Mean CSA values rose with increasing weight of the pigs. the suprarenal aorta was larger than the infrarenal only in 75 kg pigs (p < 0.05). At three measurements the within-subject variation was 0.89% and the between-subject variation 99.11%. Nitroglycerine infusion produced only minor, insignificant CSA increase, indicating negligible tone in the abdominal aorta. All segments showed circumferential wall strain-tension relations with exponential behaviour uninfluenced by weight class or site in the aorta.

CONCLUSION

Information of vascular mechanics in vivo is obtainable by catheterization of the abdominal aorta.

Passive elastic wall properties in isolated guinea pig small intestine

The aim was to study and compare the passive biomechanical wall properties in the isolated duodenum and distal ileum of the guinea pig in vitro. The organ bath contained a Krebs-Ringer solution with 10(-2) M MgCl2 to abolish smooth muscle contractile activity. Stepwise inflation of an intraluminal balloon, in which the cross-sectional area (CSA) was measured, provided the distension stimulus. The circumferential wall tension-strain distributions and wall stiffness-strain relations were computed from steady-state values of these measurements in order to evaluate the passive elastic properties. The CSA always reached equilibrium within the 2-min distension period. The CSAs obtained in the distal ileum were higher than those in the duodenum (P < 0.001). The basal CSA was 17.31 +/- 1.14 mm2 and 12.96 +/- 0.42 mm2 for the distal ileum and the duodenum, respectively (P < 0.01). At a maximum pressure of 6 kPa, the CSA of the ileum was 56.63 +/- 1.81 mm2 and 36.86 +/- 1.76 mm2 for the duodenum (P < 0.01). The circumferential wall tension-strain distributions showed an exponential behavior that accorded well with the equation Y = exp(a+bX) with determination coefficients of 0.96 +/- 0.01 and 0.99 +/- 0.00 in the duodenal segments in the distal ileal segments, respectively. The values of a (intercept with the y-axis) were 0.54 +/- 0.11 and -0.35 +/- 0.19 for the duodenal and ileal segments, respectively (P < 0.001). The slope of the curves (b values) were 4.34 +/- 0.35 in the duodenal and 5.23 +/- 0.37 in the ileal segments (0.1 > P > 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)