The isolated perfused Langendorff heart

In general, the experimental setup was realized as described before [19, 20]: male Wistar rats aged 2 to 3 months with a weight between 250 g and 350 g were used for the experiment. They were sedated with an intraperitoneal injection of pentobarbital (80 mg/kg body weight; Narcoren, Merial, Germany) in a final volume of 2 ml NaCl (0.9%) containing 0.2 ml heparin (25,000 IU/5 ml; B. Braun, Melsungen, Germany), placed latero-proximal to the hind legs. After 5 min, sufficient depth of sedation was ensured by testing for the absence of reflexes. The rats were decapitated using a guillotine, followed by an immediate thoracotomy to remove the heart and connect it to the Langendorff system. The hearts were perfused with a previously prepared buffer solution at a constant system pressure of 80 mmHg. The buffer, a modified Krebs–Henseleit solution, was composed of the following components:

118 mM sodium chloride (VWR Chemicals Internationals GmbH, Belgium)

25 mM sodium hydrogen carbonate (Carl Roth GmbH + Co. KG, Germany)

11 mM D-glucose (Carl Roth GmbH + Co. KG, Germany)

4.7 mM potassium chloride (Merck Chemicals GmbH, Germany)

2.25 mM calcium chloride (Merck Chemicals GmbH, Germany)

1.2 mM magnesium sulfate hepta-hydrate (Carl Roth GmbH + Co. KG, Germany)

1.2 mM potassium dihydrogen phosphate (Merck Chemicals GmbH, Germany)

1 mM L-lactic acid sodium salt (AppliChem GmbH, Germany)

0.5 mM ethylenediaminetetraacetic acid (Carl Roth GmbH + Co. KG, Germany)

During the experiments, oxygen supply was ensured with a carbogen mixture of 95% O2 and 5% CO2. The maximal and the minimal left ventricular pressure (LVPmax and LVPmin) was measured by inserting a balloon into the left ventricle as a pressure gauge after removal of the left auricle. After a stabilization period of 15 min, electrocardiogram (ECG) electrodes were attached to the heart to measure cardiac electrical activity and the heart rate (HR). A total of three electrodes were placed, one at the aorta, one around the left coronary artery in close proximity to the location of the left auricle, and one at the cardiac apex. Coronary flow (CF) was measured by collecting and weighing the buffer solution that passed through the heart during a 1-min period.

Data were digitized using an analog to digital converter (PowerLab/8SP, ADInstruments Pty Ltd, Castle hill, Australia) and continuously recorded on a personal computer using LabChart Reader (ADInstruments) for Windows v5.0. The evaluation of the records regarding the functional parameters was done manually for the time points presented below.

Study group

Based on the knowledge of concentrations of the predecessor drug Cumyl-PEGACLONE and of THC, two concentrations were determined that served as low and high doses of 5F-Cumyl-PEGACLONE. The low dose was defined as 50 ng/ml and the high dose as 100 ng/ml.

For the high dose of 100 ng/ml, 80 µl 5F-Cumyl-PEGACLONE were mixed with 7920 µl of the modified Krebs–Henseleit carrier solution. For the low dose of 50 ng/ml, 40 µl 5F-Cumyl-PEGACLONE were added to 7960 µl of modified Krebs–Henseleit carrier solution.

In total, the study group comprised 25 hearts. Of these, 13 hearts were exposed to 5 ml of the low dose solution of 5F-Cumyl-PEGACLONE (low dose group) and 12 hearts were exposed to 5 ml of the high dose solution of 5F-Cumyl-PEGACLONE (high dose group) by using a perfusor running at 30 ml/h over a time span of 10 min (drug phase).

The recording of the relevant functional parameters (see above) started right before the onset of the perfusor, determining the individual base line values for each heart. After starting the perfusor, the designated parameters were measured at minutes 1, 2, 3, 4, 5, and 6 during the drug phase. Further measurements were performed at minutes 15, 30, 40, 50, 60, and 70 while hearts were, again, perfused only by the buffer solution (observation phase). Due to the manual measurement of CF, it was determined only before the onset of the perfusor and then at minutes 1 and 6 during the drug phase; during the observation phase, like the other parameters, it was measured at minutes 15, 30, 40, 50, 60, and 70. An overview of the study protocol is provided in Fig. 1.

Fig. 1

Study protocol of the experiments with the isolated perfused Langendorff heart

Control group

The control group comprised 13 hearts. Like in the study group, the hearts were connected to the Langendorff system and baseline values for each parameter were determined. During the drug phase, control hearts were only perfused with additional 5 ml of pure modified Krebs–Henseleit solution over a 10-min period. The measurements of the functional parameters were realized in the same way as in the study group.

Electrocardiography

Besides a general assessment of all ECG graphs in terms of a “screening” for anomalies, the recorded data was manually evaluated by measuring the duration of the QRS complexes, as markers for the excitation propagation in the heart chambers, as well as the duration of RR intervals, as markers for the duration of an electrical heart action. They were measured at the same time points as all the other functional parameters, including the evaluation of a base line value right before the onset of the perfusor.

A general assessment of the QT interval was not possible because most ECGs did not show clear T waves. Therefore, this parameter was not included in the statistical analysis. Only two ECGs could be evaluated in this respect; the results are presented in a descriptive way.

Statistics

One heart each from the high and the low dose group had to be excluded from the analysis because recording problems occurred during the experiments and thus, in retrospect, the documented values were not considered to be valid. Furthermore, one heart of the control group had to be excluded since the measured values presented massive and unexplainable deviations compared to the other hearts in this group. In summary, 11 hearts of the high dose group and 12 hearts each of the low dose group and of the control group, meaning 35 hearts in total, were included in the statistical analysis.

The acquired data were first analyzed by a univariate analysis procedure, the so-called Kruskal–Wallis analysis (one-way ANOVA of ranks), a statistical approach based on so-called rank sums which is suitable for the comparison of more than two groups. The null hypothesis assumes that there is no difference between the groups. Thus, in our model, the null hypothesis states that 5F-Cumyl-PEGACLONE has no measurable effect on the hearts’ function. In contrast, the H1 hypothesis states that 5F-Cumyl-PEGACLONE has a measurable effect on the hearts’ function.

The calculations were not based on the absolute measuring values, but on the deviation from the baseline as their reference value. For each measured parameter (LVP min, LVP max, CF, HR, QRS complex, RR interval), p-values were calculated at each measuring time point to detect a possible effect of the administered 5F-Cumyl-PEGACLONE. p-values < 0.05 were considered significant.

As a post hoc analysis, a Bonferroni test was performed on the results of the Kruskal–Wallis analysis.

In addition, a predictive model in terms of a multivariate analysis was applied. The model was trained on the data to learn differences in measurement patterns between the three groups, which could then be used to assign new, previously unseen cases (hearts) to the correct group (low dose group, high dose group, control group). The process was repeated several times so that different hearts could be left out and serve as the new, unseen case. The hit rate was put into a chart, with the X-axis showing the group which was predicted by the model and the Y-axis showing the true group in which the heart belonged.