Abstract
Introduction
Glucagon-like peptide-1 receptor agonists reduce insulin requirements and improve glucose control when administered before cardiac surgery. An increase in endogenous insulin release is the most likely mechanism, but this has never been studied in the setting of cardiac surgery. We hypothesized that liraglutide increases pancreatic insulin secretion during cardiac surgery with cardiopulmonary bypass (CPB).
Methods
We performed a planned prospective substudy of a multicenter randomized-controlled trial (GLOBE trial, NTR6323). Patients undergoing cardiac surgery with CPB were randomized to receive either two preoperative subcutaneous injections of liraglutide or a matching placebo. We measured hormone concentrations before and after surgery, including insulin, glucagon, C-peptide and free fatty acid (FFA), and calculated HOMA-B, HOMA-IR and insulin/glucagon ratios. We compared between-group and before and after surgery differences in outcomes.
Results
Metabolic hormone concentrations were measured in 37 participants. HOMA-B revealed that liraglutide increased insulin secretion relative to glycemia (258 ± 179 vs 116 ± 180, difference (95% CI): 142 (24–261), P = 0.004). While insulin, C-peptide and glucagon levels did not differ significantly between groups, the insulin/glucagon ratios were significantly higher in the liraglutide group (preoperatively: 1.09 ± 0.45 vs 0.79 ± 0.35 difference (95% CI): −0.30 (−0.57 to −0.03), P = 0.039). Overall, postoperative insulin levels decreased >60% from preoperative insulin levels (55 ± 31 to 21 ± 9.8, difference (95% CI): −29 (−36 to −22), P < 0.001).
Conclusion
Preoperative liraglutide administration increased beta-cell function, measured as HOMA-B, and higher insulin/glucagon ratios. These results could explain the lower glucose and FFA concentrations in the liraglutide-treated patients. Interestingly, in both groups, we observed a remarkable drop in insulin and other hormone levels over the course of surgery.
Introduction
Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) might be an attractive adjunct to insulin for glycemic control in the perioperative period. Provided that beta-cell function is intact, GLP-1 stimulates insulin and suppresses glucagon release in a glucose-dependent way. Thus, GLP-1 RAs effectively lower glucose during hyperglycemia, yet are less effective during normoglycemia and have a low risk of hypoglycemia (1, 2). Furthermore, GLP-1 increases the insulin/glucagon ratio (IGR). While these effects are well-described in the outpatient population of people with type 2 diabetes mellitus (T2D) (3), these effects have not been studied in patients with stress hyperglycemia due to cardiac surgery.
Cardiac surgery and cardiopulmonary bypass (CPB) have a significant impact on normal physiology through the associated surgical stress, anesthesia, hemodilution, hemofiltration and cooling. This also impacts the ‘endocrine milieu’, for instance, by increasing insulin resistance, leading to stress hyperglycemia. While perioperative hyperglycemia is traditionally treated with insulin, we demonstrated in a recent randomized-controlled trial that preoperative liraglutide effectively reduced insulin requirements and blood glucose concentrations in patients undergoing cardiac surgery (1). In this study population, we aimed to investigate the mechanism of action of the trial intervention. Given the difference between stress hyperglycemia following cardiac surgery and hyperglycemia in patients with T2D, we investigated the mechanism of action of a GLP-1 RA on blood glucose concentrations in a substudy of the GLOBE trial (1). Therefore, we measured the most relevant glucose regulatory hormones and determined insulin resistance parameters. We hypothesized that preoperative administration of liraglutide would increase pancreatic insulin secretion and attenuate the development of insulin resistance during cardiac surgery with CPB.
Materials and methods
Study design
This study was a prospective, single-center substudy of the ‘GLP-1 for bridging of hyperglycemia during cardiac surgery’ (GLOBE) trial (4). The initial study protocol and the amendment that included the additional measurements for this substudy were approved by the Medical Research Ethics Committee (METC) of the Amsterdam UMC (number 2017_012) before the start of the study. Before initiation, this study was included in the Netherlands trial registry (NTR6323). A summary of the GLOBE trial is given here.
The GLOBE trial
The GLOBE trial was a multicenter, triple-blind, placebo-controlled, parallel-group, phase III, randomized superiority trial. The primary objective of the GLOBE trial was to evaluate whether perioperative administration of liraglutide resulted in a lower number of patients requiring insulin for glycemic control during cardiac surgery compared to placebo. The study protocol and primary results of the GLOBE trial have been published (4, 5). In this study, randomization allocated patients to a liraglutide or placebo group, with stratification per center and T2D. The liraglutide group received a first dose of 0.6 mg liraglutide on the evening before surgery and a second dose of 1.2 mg liraglutide after induction of anesthesia. The placebo group received matching placebo medication. Intraoperative glycemic control was performed using a sliding scale algorithm with intravenous insulin (Supplementary Table 1 (see section on Supplementary materials given at the end of the article)), targeting intraoperative blood glucose concentrations between 4.0 and 8.0 mmol/L, using NovoRapid (Novo Nordisk, Denmark). All patients, care providers and study personnel were blinded to treatment allocation.
Participants
Adult patients (18–80 years) scheduled for elective cardiac surgery were eligible for inclusion. Criteria for exclusion were: type 1 diabetes, insulin use >0.5 IU/kg/day, use of GLP-1 or corticosteroids, heart failure (only New York Heart Association (NYHA) class IV), impaired renal function (creatinine ≥133 μmol/L for men and ≥115 μmol/L for women), allergy to GLP-1 (study intervention), previous pancreatic surgery, pancreatitis, medullary thyroid cancer or multiple endocrine neoplasia syndrome type 2 and pregnancy or breastfeeding.
Study procedures
All participants provided written informed consent before trial participation. In addition to the primary trial procedures, a (PRE) blood sample was taken preoperatively after placement of the arterial catheter, before administration of the second dose of liraglutide. The second (POST) blood sample was drawn upon completion of the procedure before transfer to the Intensive Care Unit (ICU). POST samples were collected with a minimum interval of 45 min from the last insulin bolus dose. Both blood samples were drawn from the arterial catheter, immediately placed on ice and stored at −80°C until analysis.
Perioperative procedures
The bypass circuit consists of a phosphorylcholine-coated tubing system (LivaNova Nederland BV, Netherlands) with an oxygenator (Inspire 8F) and arterial line, a soft-shell venous reservoir (BMR 1900, Medtronic/LivaNova Nederland NV, Netherlands) and a cardiotomy reservoir (Medtronic/LivaNova Nederland NV, Netherlands). The bypass circuit was primed with 1000–1200 mL lactated Ringers (Baxter BV, The Netherlands), 5000 IE heparin and 100 mL mannitol (15%, Baxter BV, the Netherlands). Myocardial protection was achieved with a cold crystalloid cardioplegia solution (St. Thomas). A cell-saving device (Autolog, Medtronic, USA) was used to re-transfuse pericardial shed blood. The intravenous fluid of choice was Sterofundin ISO (B. Braun, Germany). The CPB priming, cardioplegia and intravenous fluids were all glucose-free.
Data collection and outcomes
Insulin and C-peptide samples were analyzed using an immunoluminometric assay (Atellica®, Siemens, Germany). The insulin assay’s lower limit of quantification was 10 pmol/L. Cross-reactivity of the insulin assay with NovoRapid insulin used perioperatively was determined between 117 and 140%. Radioimmunoassay (EMD Millipore, USA) determined total endogenous GLP-1 and glucagon levels. For free fatty acids (FFA) concentrations, the samples were analyzed by spectrophotometric assay (Abbott Architect, USA). In case values were outside the measurement range, the threshold value was recorded for analysis. We calculated change from baseline as the difference between PRE and POST measurements. We calculated IGR as insulin (pmol/L)/glucagon (ng/mL), homeostasis model assessment of insulin resistance (HOMA-IR), a measure of insulin resistance, as glucose (mmol/L)*insulin (pmol/L)/135 and homeostasis model assessment of beta-cell function (HOMA-B), a measure of beta-cell function, as 20/3*insulin (pmol/L)/(glucose (mmol/L)-3.5) (6).
The outcomes of this exploratory study included the between-group differences in concentrations of insulin, glucose, C-peptide, glucagon, GLP-1, FFA and the HOMA-IR, HOMA-B and IGR at both time points (PRE and POST) and the change between measurements.
Sample size
A previous study evaluating the efficacy of GLP-1 infusion included a similar patient group (7). Based on serial insulin measurements ranging from 6.7 (±3.4) to 23.1 (±37.6) μU/mL, we assumed a standard deviation of 15 μU mL−1 to detect a difference of at least 15 μU/mL. Based on a two-sided Z-test with a power of 80% and a significance level of 5%, we needed 16 patients per group. Since insulin concentrations were not normally distributed, we added a 15% correction factor, (8) requiring the inclusion of 36 patients.
Statistical analyses
Discrete variables are presented as count (%) and compared between the intervention and control group using chi-quadrate or Fisher’s exact test. Continuous variables are presented as the mean ± SD or median (IQR) and compared using Student’s t-test or Mann–Whitney U tests, if data were not normally distributed. Normality was assessed with histograms, Q–Q plots and the Shapiro–Wilk test. For all outcomes, we analyzed the between-group differences with the unpaired Student T-test and the before-to-after-surgery differences with the paired Student T-test. Given the post-hoc design of this substudy, all outcomes are secondary and hypotheses-generating only. A P-value below 0.05 was considered significant.
Results
From the approval of the amendment that included this substudy, all subsequent participants in the Amsterdam University Medical Centers were included until the end of the GLOBE trial. Between April 2018 and August 2018, we included 37 patients in this study, of which 21 were allocated to the liraglutide group and 16 to the placebo group. Baseline characteristics and perioperative context are presented in Table 1. Eight patients had a history of type 2 diabetes mellitus, none of whom used a form of insulin therapy. The laboratory found significant hemolysis in several samples, disabling analysis. This resulted in missing outcomes for one preoperative and six postoperative insulin measurements. Likewise, three GLP-1 and three FFA samples were unavailable.
Baseline characteristics.
| Group | All (n = 37) | Liraglutide (n = 21) | Placebo (n = 16) |
|---|---|---|---|
| Age, years | 61.8 (± 14.4) | 63.2 (± 14.6) | 60.0 (± 14.6) |
| Sex, female | 3 (8.1%) | 1 (4.8%) | 2 (12.5%) |
| BMI, kg/m2 | 27.5 (± 3.7) | 27.9 (± 4.0) | 27.1 (± 3.3) |
| Type 2 DM | 8 (21.6%) | 5 (23.8%) | 3 (18.8%) |
| ASA physical status | |||
| 2 | 16 (43.2%) | 11 (52.4%) | 5 (31.3%) |
| 3 | 18 (48.6%) | 9 (42.9%) | 9 (56.3%) |
| 4 | 3 (8.1%) | 1 (4.8%) | 2 (12.5%) |
| Duration, minutes | |||
| Surgery | 264 (218–333) | 263 (211–321) | 297 (218–335) |
| CPB | 145 (99–177) | 125 (93–169) | 167 (113–186) |
| Aortic cross clamp | 111 (65–140) | 87 (64–87) | 130 (75–148) |
| Type of surgery | |||
| CABG only | 8 (21.6%) | 5 (23.8%) | 3 (18.8%) |
| Single non-CABG | 14 (37.8%) | 8 (38.1%) | 6 (37.5%) |
| ≥2 procedures* | 15 (40.1%) | 8 (38.1%) | 7 (43.8%) |
| Insulin administered † | 10 (27.0%) | 4 (19.0%) | 6 (37.5%) |
| Insulin dosage ‡ , IU | 2.4 (± 6.4) | 0.8 (± 1.8) | 4.4 (± 9.3) |
Values are displayed as n (%), mean (±SD) and median (Q1–Q3). BMI, body mass index; ASA, American Society of Anesthesiologists; CPB, cardiopulmonary bypass; CABG, coronary artery bypass grafting; IU, international units.
Multiple procedures at once, e.g., CABG and valvular surgery.
Number of patients who had received insulin during the perioperative period.
Mean dosage of insulin during the perioperative period.
Between-group comparisons of outcomes are summarized in Table 2, while Table 3 presents the entire cohort, comparing preoperative to postoperative concentrations. PRE and POST measurements of GLP-1 revealed higher concentrations in the intervention group in accordance with the GLP-1 receptor agonist, liraglutide, which was administered in this group (Table 2). This was accompanied by lower glucose measurements preoperatively and postoperatively, although the latter did not reach statistical significance (P = 0.061).
Between-group comparisons of glucose, glycemic control hormones and insulin resistance.
| Glucose (mmol/L) | Liraglutide | Placebo | Difference & 95% CI | P-value |
|---|---|---|---|---|
| Preoperative | 5.5 ± 1.0 | 6.5 ± 2.0 | 1.0 (0.1–2.1) | 0.047 |
| Postoperative | 6.2 ± 1.1 | 7.3 ± 1.2 | 1.1 (−0.1–2.2) | 0.061 |
| Insulin (pmol/L) | ||||
| Preoperative | 61 ± 30 | 47 ± 22 | −14 (−34–6) | 0.17 |
| Postoperative | 24 ± 11 | 17 ± 11 | −7 (−14–1) | 0.081 |
| Glucagon (ng/L) | ||||
| Preoperative | 58 ± 19 | 61 ± 19 | 3 (−10–16) | 0.69 |
| Postoperative | 36 ± 14 | 36 ± 14 | 2 (−9–12) | 0.76 |
| Insulin/glucagon ratio | ||||
| Preoperative | 1.09 ± 0.45 | 0.79 ± 0.35 | −0.30 (−0.57–−0.03) | 0.039 |
| Postoperative | 0.78 ± 0.33 | 0.52 ± 0.35 | −0.25 (−0.50–−0.06) | 0.032 |
| C-peptide (pmol/L) | ||||
| Preoperative | 734 ± 334 | 650 ± 278 | −84 (−286–119) | 0.40 |
| Postoperative | 554 ± 239 | 536 ± 221 | −18 (−168–133) | 0.82 |
| GLP-1 (ng/L) | ||||
| Preoperative | 27 ± 7 | 14 ± 7 | −13 (−18–−9) | <0.001 |
| Postoperative | 36 ± 23 | 13 ± 13 | −23 (−36–−9) | <0.001 |
| FFA (μmol/L) | ||||
| Preoperative | 391 ± 166 | 525 ± 172 | 134 (23–246) | 0.014 |
| Postoperative | 416 ± 206 | 558 ± 209 | 142 (−13–298) | 0.063 |
| HOMA-IR | ||||
| Preoperative | 2.55 ± 1.42 | 2.56 ± 1.48 | −0.0 (−1.27–1.31) | 0.98 |
| Postoperative | 1.14 ± 0.64 | 0.80 ± 0.64 | −0.3 (−0.82–0.15) | 0.13 |
| HOMA-B | ||||
| Preoperative | 258 ± 179 | 116 ± 180 | −142 (−261–−24) | 0.004 |
| Postoperative | 77 ± 75 | 40 ± 76 | −37 (−93–19) | 0.12 |
FFA, free fatty acids; HOMA-IR, homeostasis model assessment of insulin resistance; HOMA-B, homeostasis model assessment of beta-cell function.
Comparisons of preoperative to postoperative measurements in the overall cohort.
| Preoperative | Postoperative | Difference & 95% CI | P-value | |
|---|---|---|---|---|
| Glucose (mmol/L) | 5.9 ± 1.6 | 6.7 ± 1.7 | 0.8 (0.2–1.4) | 0.018 |
| Insulin (pmol/L) | 55 ± 31 | 21 ± 9.8 | −29 (−36–−22) | <0.001 |
| Glucagon (ng/L) | 59 ± 20 | 35 ± 15 | −24 (−30–−18) | <0.001 |
| C-peptide (pmol/L) | 698 ± 295 | 546 ± 221 | −151 (−218–−81) | <0.001 |
| FFA (μmol/L) | 447 ± 165 | 475 ± 227 | 36 (−46–119) | 0.39 |
FFA, free fatty acids.
Our primary endpoint, the postoperative insulin concentrations, did not differ significantly between the study groups, liraglutide: 24 ± 11 vs placebo: 17 ± 11, the difference (95% CI) = −7 (−14 to 1), P = 0.081. We also observed no between-group differences in preoperative insulin concentrations (Table 2). However, in both groups, insulin concentrations declined significantly from preoperative levels, overall: PRE: 55 ± 31 vs POST: 21 ± 9.8 difference: −29 (−36 to −22), P < 0.001, (Table 3). This difference was not significantly different for the between-group comparison.
Although insulin concentrations did not differ between groups, HOMA-B was significantly higher in the liraglutide group compared to placebo, liraglutide: 258 ± 179 vs placebo: 116 ± 180, difference 142 (24–261), P = 0.004. The decline in insulin concentrations postoperatively lowered HOMA-B scores, and we observed no significant between-group difference at this time (P = 0.12). In contrast to HOMA-B scores, insulin resistance measured as HOMA-IR did not reveal between-group differences.
Compared to preoperative concentrations, C-peptide and glucagon decreased significantly postoperatively, without any between-group differences at either time point (Table 2). Despite the comparable insulin and glucagon measurements, liraglutide treatment resulted in a significantly higher IGR before and after surgery (Table 2). The higher IGR in the liraglutide-treated patients were accompanied by lower FFA concentrations in the placebo group preoperatively: liraglutide 391 ± 166 vs placebo: 525 ± 172 μmol/L, difference: 134 (23–246), P = 0.014. Postoperatively, we observed similar differences in FFA concentrations, although these did not reach statistical significance (P = 0.063).
Discussion
In this study, we observed several interesting and hypothesis-generating findings. First, beta-cell function, measured as HOMA-B, was significantly higher in the liraglutide group. Second, while insulin and glucagon concentrations did not differ between groups, IGRs were significantly higher in the liraglutide group, coinciding with lower FFA concentrations. Finally, insulin, glucagon and C-peptide concentrations all declined significantly during the intraoperative period.
Interestingly, HOMA-B was a more sensitive outcome parameter for detecting insulin secretion relative to glucose levels. It is well-known that GLP-1 RAs increase insulin secretion in response to blood glucose concentrations, with no effect in the hypoglycemic range (<4 mmol/L) and increasing effects depending on the level of hyperglycemia (9). Since 78% patients in our study had no history of T2D and all patients were required to fast during the night before surgery, preoperative glucose concentrations were in the normoglycemic range, likely attenuating the effect of liraglutide on pancreatic insulin secretion. As a result, while insulin concentrations were not statistically higher in the liraglutide group, HOMA-B supports the GLP-1-induced increase in beta-cell function.
Of note, the glucose-lowering effect of liraglutide in cardiac surgery might not be solely attributed to an insulin-increasing effect. Besides stimulating insulin secretion, GLP-1 RAs also have a glucagon-lowering effect, and hence, increase IGRs. We found that liraglutide does indeed positively affect the IGR, suggesting that this might be an additional mechanism affecting perioperative glucose metabolism. Together, insulin and glucagon regulate the balance between storage and mobilization of nutrients. A low IGR also promotes the breakdown of adipose tissue into FFA and glycerol (10). Interestingly, in our study, we also observed that the higher IGRs in the liraglutide group coincided with lower FFA concentrations at both measurement moments. Of note, the postoperative increase in FFA resulting from lipolysis is an expected finding, resulting from the increase in adrenaline, cortisol and growth hormone previously described as part of the surgical stress response. While the clinical relevance of lowering FFA concentrations is not fully elucidated, a correlation exists between higher FFA concentrations and cardiovascular complications in patients with coronary artery disease (11). As such, the lower FFA concentrations in the liraglutide group may constitute a potential cardiovascular protective effect of this intervention.
Intriguingly, we found a decrease in insulin and glucagon concentrations during the intraoperative period (12). In addition, the postoperatively lower insulin concentrations resulted in a comparable lowering of insulin resistance calculated by HOMA-IR. Previous studies in cardiac surgery patients have reported significantly higher insulin concentrations at the end of surgery (13, 14). However, these studies are not conclusive as to whether this is due to increased pancreatic insulin secretion or iatrogenic insulin administration. On the contrary, older studies showed a drop in insulin and glucagon levels during the perioperative phase. Several explanations for this discrepancy can be put forward. In our study, perioperative glucose control was achieved with short-acting intravenous insulin bolus according to the study protocol sliding-scale regimen. This resulted in relatively few patients receiving insulin with a low total dose intraoperatively. Other significant differences with previous studies include the use of hyperinsulinemic normoglycemic clamp technique in some studies, (15, 16, 17) the administration of steroids and variable doses of glucose infusions. As CPB circuits are primed with a crystalloid solution, the initiation of bypass circulation results in significant hemodilution. However, hemodilution cannot explain the observed drop in insulin concentrations in excess of 60%. This is supported by the observation that while other hormone concentrations decreased intraoperatively, these were less significant. Another hypothesis for the insulin-lowering effect of cardiac surgery is also related to the CPB circuit, as the plastic-binding properties of insulin have long been known (18). To enable oxygenation, the CPB comprises a large surface area of plastic tubing in contact with blood circulation (19). An in vivo study reported that CPB resulted in a loss of >60% of circulating insulin, which could be reduced to 30–40% through different forms of tube coatings (19). While this effect is well-documented, perhaps it is an underappreciated mechanism behind cardiac surgery-induced hyperglycemia. Recently, another ex-vivo study determined insulin concentrations during CPB. They found that biocoating of the plastic tubing, in combination with fresh frozen plasma (used to saturate binding sites for peptides such as insulin), minimized the loss of insulin from circulation (20). In contrast, this study argued that the loss of insulin in clinical practice might be better explained by hemolysis through either the release of insulin, an enzyme that breaks down insulin, or insulin binding to free hemoglobin molecules (20). While hemolysis was present in some samples (primarily postoperatively), we did not quantify the presence or degree of hemolysis in all samples. We strongly recommended future research on perioperative glucose control and insulin concentrations to include quantification of hemolysis as a confounding factor.
Limitations
Our study comprises a relatively small sample size with an unbalance in group allocation, limiting the strength of our findings. Preoperative insulin and C-peptide concentrations were both higher in the liraglutide group, although the differences were not significant in this limited sample size. Results should be interpreted as hypothesis-generating. Furthermore, six of the postoperative insulin measurements could not be analyzed due to the hemolysis of the sample, most likely related to the use of CPB and concentrated red cell transfusions. The insulin assay used has a 100–140% cross-reactivity with the NovoRapid insulin used for perioperative glucose control. Theoretically, this may have confounded IGR and HOMA measurements. However, NovoRapid is a short-acting form of insulin, with a half-life of 4–6 min. Importantly, since POST samples were collected >45 min after insulin administration, any potential confounding effect of exogenous insulin on these measurements is likely minimal.
It is important to note that the Millipore RIA kit used for glucagon measurement in this study was produced before the 2019 antibody change that has been associated with flawed outcomes (21). Future studies should validate glucagon levels using updated methods to confirm our findings. Finally, we have not performed any intraoperative measurements, limiting our ability to interpret the moment of decline in insulin and its temporal relation to perioperative events such as the start of CPB and the resulting hemodilution.
Conclusion
In this study, we observed a higher HOMA-B index and higher IGRs in the liraglutide-treated group compared to the placebo group. These results correlated with lower glucose and FFA concentrations. Liraglutide did not result in a significant increase in insulin concentrations in this study, while we observed remarkable declines in postoperative insulin and glucagon concentrations compared to preoperative levels. The effects of liraglutide on glucose and FFA concentrations through improved beta-cell function and IGRs are interesting and potentially protective mechanisms for the improvement of perioperative care for cardiac surgery patients.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/EC-24-0427.
Declaration of interest
A H Hulst, J H DeVries, B Preckel and J Hermanides received research support from Novo Nordisk.
Funding
This investigator-initiated study was partly supported by a research grant from the Investigator-Initiated Studies Program of Novo Nordisk (Bagsvaerd, Denmark). The opinions expressed in this paper are those of the authors and do not necessarily represent those of Novo Nordisk. This publication is part of the Rubicon Research Programme NEPHRITIC (project number 452020104), which is financed by the Dutch Research Council (NWO). ‘This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 101024833.’
References
- 1↑
Hulst AH , Visscher MJ , Godfried MB , et al. Liraglutide for perioperative management of hyperglycaemia in cardiac surgery patients: a multicentre randomized superiority trial. Diabetes Obes Metab 2020 22 557–565. (https://doi.org/10.1111/dom.13927)
- 2↑
Hulst AH , Plummer MP , Hollmann MW , et al. Systematic review of incretin therapy during peri-operative and intensive care. Crit Care 2018 22 299. (https://doi.org/10.1186/s13054-018-2197-4)
- 3↑
Guenifi A , Ahrén B & Abdel-Halim SM . Differential effects of glucagon-like peptide-1 (7-36)amide versus cholecystokinin on arginine-induced islet hormone release in vivo and in vitro. Pancreas 2001 22 58–64. (https://doi.org/10.1097/00006676-200101000-00010)
- 4↑
Hulst AH , Visscher MJ , Godfried MB , et al. Study protocol of the randomised placebo-controlled GLOBE trial: GLP-1 for bridging of hyperglycaemia during cardiac surgery. BMJ Open 2018 8 e022189. (https://doi.org/10.1136/bmjopen-2018-022189)
- 5↑
Hulst AH , Visscher MJ , Godfried MB , et al. Liraglutide for perioperative management of hyperglycaemia in cardiac surgery patients: a multicentre randomized superiority trial. Diabetes Obes Metab 2019 22 557–565. (https://doi.org/10.1111/dom.13927)
- 6↑
Matthews DR , Hosker JP , Rudenski AS , et al. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985 28 412–419. (https://doi.org/10.1007/bf00280883)
- 7↑
Kohl BA , Hammond MS , Cucchiara AJ , et al. Intravenous GLP-1 (7-36) amide for prevention of hyperglycemia during cardiac surgery: a randomized, double-blind, placebo-controlled study. J Cardiothorac Vasc Anesth 2014 28 618–625. (https://doi.org/10.1053/j.jvca.2013.06.021)
- 8↑
Lehmann EL . Nonparametrics: Statistical Methods Based On Ranks, pp 76–81. New York, NY, USA: Springer.
- 9↑
Meloni AR , Deyoung MB , Lowe C , et al. GLP-1 receptor activated insulin secretion from pancreatic β-cells: mechanism and glucose dependence. Diabetes Obes Metab 2013 15 15–27. (https://doi.org/10.1111/j.1463-1326.2012.01663.x)
- 10↑
Kalra S & Gupta Y . The insulin:glucagon ratio and the choice of glucose-lowering drugs. Diabetes Ther 2016 7 1–9. (https://doi.org/10.1007/s13300-016-0160-4)
- 11↑
Pilz S , Scharnagl H , Tiran B , et al. Free fatty acids are independently associated with all-cause and cardiovascular mortality in subjects with coronary artery disease. J Clin Endocrinol Metab 2006 91 2542–2547. (https://doi.org/10.1210/jc.2006-0195)
- 12↑
Duncan AE . Hyperglycemia and perioperative glucose management. Cuur Pharm Des 2012 18 6195–6203. (https://doi.org/10.2174/138161212803832236)
- 13↑
Mescheryakov AV , Kozlov IA , Dementyeva II , et al. Glucose metabolism and insulin activity during cardiac surgery. J Cardiothorac Anesth 1989 3 536–543. (https://doi.org/10.1016/0888-6296(89)90149-x)
- 14↑
Allison SP . Changes in insulin secretion during open heart surgery. Br J Anaesth 1971 43 138–143. (https://doi.org/10.1093/bja/43.2.138)
- 15↑
Németh Á , Cziráki A , Sulyok E , et al. Short-term response of metabolic hormones to coronary artery bypass surgery. Adv Med Sci 2014 59 213–220. (https://doi.org/10.1016/j.advms.2014.04.002)
- 16↑
Visser L , Zuurbier CJ , Hoek FJ , et al. Glucose, insulin and potassium applied as perioperative hyperinsulinaemic normoglycaemic clamp: effects on inflammatory response during coronary artery surgery. Br J Anaesth 2005 95 448–457. (https://doi.org/10.1093/bja/aei220)
- 17↑
Smith CE , Styn NR , Kalhan S , et al. Intraoperative glucose control in diabetic and nondiabetic patients during cardiac surgery. J Cardiothorac Vasc Anesth 2005 19 201–208. (https://doi.org/10.1053/j.jvca.2005.01.032)
- 18↑
Hirsch JI , Wood JH & Thomas RB . Insulin adsorption to polyolefin infusion bottles and polyvinyl chloride administration sets. Am J Hosp Pharm 1981 38 995–997. (https://doi.org/10.1093/ajhp/38.7.995)
- 19↑
Urban K , Redford D & Larson DF . Insulin binding to the cardiopulmonary bypass biomaterials. Perfusion 2007 22 207–210. (https://doi.org/10.1177/0267659107081632)
- 20↑
Schweizer T , Nossen CM , Galova B , et al. In vitro investigation of insulin dynamics during 4 hours of simulated cardiopulmonary bypass. Anesth Analg 2024 [epub]. (https://doi.org/10.1213/ANE.0000000000007106)
- 21↑
Wu T , Rayner CK , Jones KL , et al. Measurement of plasma glucagon in humans: a shift in the performance of a current commercially available radioimmunoassay kit. Diabetes Obes Metabol 2022 24 1182–1184. (https://doi.org/10.1111/dom.14673)
