As patients with diabetes have a significantly increased risk of coronary heart disease including risk of myocardial infarction , it is of major importance that new treatment options do not increase myocardial infarct size during ischemia or increase the risk of arrhythmia during ischemia. Liraglutide, a long-acting GLP1 analogue, is a promising new treatment option for patients with type 2 diabetes. There is, however currently limited data on the cardiovascular effects of especially the GLP1 analogues.
In the present study, we examined the effect of Liraglutide administered subcutaneously as clinically intended, in contrast to previous experimental studies with intravenous infusions immediately before, during or after ischemia. The effect was neutral on myocardial infarct size in our porcine model. Heart rate was the only parameter differing significantly between the groups. This is in accordance with a previous experimental study reporting effects of GLP1 on heart rate . Blood pressure did not differ significantly.
The obtained plasma concentrations of Liraglutide immediately before ischemia was comparable to the same body weight adjusted doses in humans .
Previously it has been reported in patients given GLP1 infusion in the reperfusion phase after successful PCI based reperfusion, that ejection fraction (EF) increased in the treated group . Whether this effect was due to inotropic effects previously reported for GLP1 or a true reduction of reperfusion injury is not obvious, and furthermore, the mentioned study must be interpreted cautiously because it was a non-randomised study with 10 and 11 patients in the groups. However the effect of GLP1 on myocardial ischemia may differ depending on whether the ischemic insult is occurring during ongoing treatment with GLP1 (or its analogues) or whether the treatment is instituted after the ischemic insult.
A previous porcine study with GLP1 infusion during and just prior to ischemia reported equal infarct sizes in treated and control groups . This is in accordance with our study on Liraglutide. A recent study in a porcine model using Exanatide showed beneficial effects when administered at reperfusion . Both timing of treatment as well as dosing regimen may thus influence the effect as well as the specific analogue.
Studies in rodent models have shown reduction in infarct size in animals given GLP1 in both in-vivo models and in Langendorff-perfused rat hearts [5, 6]. It is, however, well known that results may differ considerably between animal models. Larger animal models such as pigs are probably more predictive of results in humans .
GLP1 has, as mentioned, been reported to possess inotropic and chronotropic effects, as well as preconditioning-like effects . Such effects are in accordance with the reported intracellular increase in cAMP levels in cardiomyocytes, possibly mediating these responses. It is well known that preischemic intermittent exposure to sympathomimetic effects can act as an ischemic preconditioning stimulus [19, 20]. However continued activation of the sympathetic system and inotropic and chronotropic effects, leading to increased myocardial oxygen requirements, are certainly not beneficial in myocardial ischemia. The possibility therefore exists that continued exposure to GLP1 and analogues could be potentially harmful in myocardial ischemia. This issue is therefore obviously very important to clarify.
Despite that heart rate was significantly larger in the treated group, the infarct sizes did not differ significantly and liraglutide therefore seems to be safe to use in this respect.
We chose a dosing regimen giving injections on a daily basis on three previous days before the ischemic insult. The cardiovascular effects could very well be different on such a dosage scheme compared to intravenous infusions as in most of the previous studies. However, subcutaneous injections are the intended route of administration with this GLP1 analogue. It cannot be excluded that differences could be anticipated with long-term administration. Additionally, effects may differ between GLP1 and the analogues such as Liraglutide.
Even though we did not include a direct measure of myocardial contractility, the indirect measures such as blood pressure and cardiac output did not differ between groups. The pigs used in the present study are non-diabetic animals. It is possible that diabetic subjects may differ in the response compared to healthy subjects. However, to our knowledge no porcine model of type 2 diabetes exists, and models using streptozocin etc simulate type 1 diabetes, and Liraglutide treatment is not intended for type 1 diabetes. Furthermore, rodent studies reporting beneficial effects in myocardial ischemia-reperfusion injury suggests an ischemic preconditioning-like effect of GLP1. This raises the possibility, that GLP1 analogues could gain use in a broader category of patients than diabetics, if such effects could be substantiated in larger animal models and humans. It is therefore necessary to conduct studies in both non-diabetic and diabetic animals.
The study is an acute study with no long-term follow up. However, since the treatment with Liraglutide was ongoing before the insult and during ischemia, it would be expected that potential effects would be manifested early. We assessed infarct size early, but we have previously found in the present model, that infarct size does not differ significantly when measured after 3 days . Finally, the porcine model, as compared to rodent models, may be less sensitive for demonstrating cardioprotection. We have, however, previously demonstrated myocardioprotective effects in our porcine model [14, 22].